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.
14 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
15 When redis_servers is set to a list of > 1 element, and the Redis servers
16 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
17 case of MOVED into a DEFER case instead, thus moving the query onto the
18 next server in the list. For a cluster of N elements, all N servers must
19 be defined in redis_servers.
21 GF/02 Catch and remove uninitialized value warning in exiqsumm
22 Check for existence of @ARGV before looking at $ARGV[0]
24 JH/01 Replace the store_release() internal interface with store_newblock(),
25 which internalises the check required to safely use the old one, plus
26 the allocate and data copy operations duplicated in both (!) of the
29 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
30 modifier. This matches the restriction on the commandline.
32 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
33 Previously only the last row was returned.
35 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
36 we assumed that tags in the header were well-formed, and parsed the
37 element content after inspecting only the first char of the tag.
38 Assumptions at that stage could crash the receive process on malformed
41 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
42 While running the DKIM ACL we operate on the Permanent memory pool so that
43 variables created with "set" persist to the DATA ACL. Also (at any time)
44 DNS lookups that fail create cache records using the Permanent pool. But
45 expansions release any allocations made on the current pool - so a dnsdb
46 lookup expansion done in the DKIM ACL releases the memory used for the
47 DNS negative-cache, and bad things result. Solution is to switch to the
48 Main pool for expansions.
49 While we're in that code, add checks on the DNS cache during store_reset,
50 active in the testsuite.
51 Problem spotted, and debugging aided, by Wolfgang Breyha.
53 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
54 When none of the hosts presented to a transport match an already-open
55 connection, close it and proceed with the list. Previously we would
56 queue the message. Spotted by Lena with Yahoo, probably involving
59 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
60 Previously a spurious "250 OK id=" response was appended to the proper
63 JH/08 The "support for" informational output now, which built with Content
64 Scanning support, has a line for the malware scanner interfaces compiled
65 in. Interface can be individually included or not at build time.
67 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
68 by the template makefile "src/EDITME". The "STREAM" support for an older
69 ClamAV interface method is removed.
71 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
72 rows affected is given instead).
74 JH/11 The runtime Berkeley DB library version is now additionally output by
75 "exim -d -bV". Previously only the compile-time version was shown.
77 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
78 SMTP connection. Previously, when one had more receipients than the
79 first, an abortive onward connection was made. Move to full support for
80 multiple onward connections in sequence, handling cutthrough connection
81 for all multi-message initiating connections.
83 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
84 routers. Previously, a multi-recipient message would fail to match the
85 onward-connection opened for the first recipient, and cause its closure.
87 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
88 a timeout on read on a GnuTLS initiating connection, resulting in the
89 initiating connection being dropped. This mattered most when the callout
90 was marked defer_ok. Fix to keep the two timeout-detection methods
93 JH/15 Relax results from ACL control request to enable cutthrough, in
94 unsupported situations, from error to silently (except under debug)
95 ignoring. This covers use with PRDR, frozen messages, queue-only and
98 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
100 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
101 metadata, resulting in a crash in free().
103 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
104 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
105 Broken also in d185889f4, with init system revamp.
107 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
108 Previously we abruptly closed the connection after reading a malware-
109 found indication; now we go on to read the "scan ok" response line,
112 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
113 ACL. Previously, a crash would result.
115 JH/19 Speed up macro lookups during configuration file read, by skipping non-
116 macro text after a replacement (previously it was only once per line) and
117 by skipping builtin macros when searching for an uppercase lead character.
119 JH/20 DANE support moved from Experimental to mainline. The Makefile control
120 for the build is renamed.
122 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
123 was allocated for every new TLS startup, meaning one per message. Fix
124 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
126 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
127 reported the original. Fix to report (as far as possible) the ACL
128 result replacing the original.
130 JH/23 Fix memory leak during multi-message connections using STARTTLS under
131 OpenSSL. Certificate information is loaded for every new TLS startup,
132 and the resources needed to be freed.
134 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
136 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
139 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
140 DATA response info to the (existing) per-recipient response info for
141 the "C=" log element. It can have useful tracking info from the
142 destination system. Patch from Simon Arlott.
144 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
145 length value. Previously this would segfault.
147 HS/02 Support Avast multiline protoocol, this allows passing flags to
148 newer versions of the scanner.
150 JH/28 Ensure that variables possibly set during message acceptance are marked
151 dead before release of memory in the daemon loop. This stops complaints
152 about them when the debug_store option is enabled. Discovered specifically
153 for sender_rate_period, but applies to a whole set of variables.
154 Do the same for the queue-runner and queue-list loops, for variables set
155 from spool message files. Do the same for the SMTP per-message loop, for
156 certain variables indirectly set in ACL operations.
158 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
159 as a multi-recipient message from a mailinglist manager). The coding had
160 an arbitrary cutoff number of characters while checking for more input;
161 enforced by writing a NUL into the buffer. This corrupted long / fast
162 input. The problem was exposed more widely when more pipelineing of SMTP
163 responses was introduced, and one Exim system was feeding another.
164 The symptom is log complaints of SMTP syntax error (NUL chars) on the
165 receiving system, and refused recipients seen by the sending system
166 (propating to people being dropped from mailing lists).
167 Discovered and pinpointed by David Carter.
169 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
170 replaced by the ${authresults } expansion.
172 JH/31 Bug 2257: Fix pipe transport to not use a socket-only syscall.
174 HS/03 Set a handler for SIGTERM and call exit(3) if running as PID 1. This
175 allows proper process termination in container environments.
177 JH/32 Bug 2258: Fix spool_wireformat in combination with LMTP transport.
178 Previously the "final dot" had a newline after it; ensure it is CR,LF.
180 JH/33 SPF: remove support for the "spf" ACL condition outcome values "err_temp"
181 and "err_perm", deprecated since 4.83 when the RFC-defined words
182 "temperror" and "permerror" were introduced.
184 JH/34 Re-introduce enforcement of no cutthrough delivery on transports having
185 transport-filters or DKIM-signing. The restriction was lost in the
186 consolidation of verify-callout and delivery SMTP handling.
187 Extend the restriction to also cover ARC-signing.
189 JH/35 Cutthrough: for a final-dot response timeout (and nonunderstood responses)
190 in defer=pass mode supply a 450 to the initiator. Previously the message
193 PP/02 DANE: add dane_require_tls_ciphers SMTP Transport option; if unset,
194 tls_require_ciphers is used as before.
196 HS/03 Malware Avast: Better match the Avast multiline protocol. Add
197 "pass_unscanned". Only tmpfails from the scanner are written to
198 the paniclog, as they may require admin intervention (permission
199 denied, license issues). Other scanner errors (like decompression
200 bombs) do not cause a paniclog entry.
202 JH/36 Fix reinitialisation of DKIM logging variable between messages.
203 Previously it was possible to log spurious information in receive log
206 JH/37 Bug 2255: Revert the disable of the OpenSSL session caching. This
207 triggered odd behaviour from Outlook Express clients.
209 PP/03 Add util/renew-opendmarc-tlds.sh script for safe renewal of public
212 JH/38 DKIM: accept Ed25519 pubkeys in SubjectPublicKeyInfo-wrapped form,
213 since the IETF WG has not yet settled on that versus the original
214 "bare" representation.
216 JH/39 Fix syslog logging for syslog_timestamp=no and log_selector +millisec.
217 Previously the millisecond value corrupted the output.
218 Fix also for syslog_pid=no and log_selector +pid, for which the pid
219 corrupted the output.
225 JH/01 Rework error string handling in TLS interface so that the caller in
226 more cases is responsible for logging. This permits library-sourced
227 string to be attached to addresses during delivery, and collapses
228 pairs of long lines into single ones.
230 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
231 during configuration. Wildcards are allowed and expanded.
233 JH/02 Rework error string handling in DKIM to pass more info back to callers.
234 This permits better logging.
236 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
237 do not close it down and have the child transport start it up again on
238 the passed-on TCP connection. Instead, proxy the child (and any
239 subsequent ones) for TLS via a unix-domain socket channel. Logging is
240 affected: the continued delivery log lines do not have any DNSSEC, TLS
241 Certificate or OCSP information. TLS cipher information is still logged.
243 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
244 identical IP addresses on different listening ports. Will also affect
247 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
248 add noisy ifdef guards to special-case this sillyness.
249 Patch from Bernd Kuhls.
251 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
252 than 255 are no longer allowed.
254 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
255 Disable the session-cache too, which might reduce our load. Since we
256 currrectly use a new context for every connection, both as server and
257 client, there is no benefit for these.
258 GnuTLS appears to not support tickets server-side by default (we don't
259 call gnutls_session_ticket_enable_server()) but client side is enabled
260 by default on recent versions (3.1.3 +) unless the PFS priority string
263 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
264 <https://reproducible-builds.org/specs/source-date-epoch/>.
266 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
267 the check for any unsuccessful recipients did not notice the limit, and
268 erroneously found still-pending ones.
270 JH/08 Pipeline CHUNKING command and data together, on kernels that support
271 MSG_MORE. Only in-clear (not on TLS connections).
273 JH/09 Avoid using a temporary file during transport using dkim. Unless a
274 transport-filter is involved we can buffer the headers in memory for
275 creating the signature, and read the spool data file once for the
276 signature and again for transmission.
278 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
279 4.77 as the kernel support then wasn't solid, having issues in 64bit
280 mode. Now, it's been long enough. Add support for FreeBSD also.
282 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
283 case where the routing stage had gathered several addresses to send to
284 a host before calling the transport for the first, we previously failed
285 to close down TLS in the old transport process before passing the TCP
286 connection to the new process. The new one sent a STARTTLS command
287 which naturally failed, giving a failed delivery and bloating the retry
288 database. Investigation and fix prototype from Wolfgang Breyha.
290 JH/12 Fix check on SMTP command input synchronisation. Previously there were
291 false-negatives in the check that the sender had not preempted a response
292 or prompt from Exim (running as a server), due to that code's lack of
293 awareness of the SMTP input buffering.
295 PP/04 Add commandline_checks_require_admin option.
296 Exim drops privileges sanely, various checks such as -be aren't a
297 security problem, as long as you trust local users with access to their
298 own account. When invoked by services which pass untrusted data to
299 Exim, this might be an issue. Set this option in main configuration
300 AND make fixes to the calling application, such as using `--` to stop
303 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
304 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
305 responses to those, into a single TLS record each way (this usually means
306 a single packet). As a side issue, smtp_enforce_sync now works on TLS
309 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
310 affects you only if you're dancing at the edge of the param size limits.
311 If you are, and this message makes sense to you, then: raise the
312 configured limit or use OpenSSL 1.1. Nothing we can do for older
315 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
316 cmdline element to get the documented default one. Previously it was
319 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
320 are unsuable for non-message-specific reasons - eg. connection timeout,
321 banner-time rejection.
323 JH/16 Fix logging of delivery remote port, when specified by router, under
326 PP/06 Repair manualroute's ability to take options in any order, even if one
327 is the name of a transport.
330 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
332 JH/17 Change the list-building routines interface to use the expanding-string
333 triplet model, for better allocation and copying behaviour.
335 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
336 Previously it was constructed the first time a possibly-matching string
337 was met in the configuration file input during startup; now it is done
340 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
341 compatible one, to avoid the (poorly documented) possibility of a config
342 file in the working directory redirecting the DB files, possibly correpting
343 some existing file. CVE-2017-10140 assigned for BDB.
345 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
346 cache-hot. Previously, although the result was properly cached, the
347 initial verify call returned a defer.
349 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
350 the main verify for receipient in uncached-mode.
352 JH/22 Retire historical build files to an "unsupported" subdir. These are
353 defined as "ones for which we have no current evidence of testing".
355 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
356 if present. Previously it was ignored.
358 JH/24 Start using specified-initialisers in C structure init coding. This is
359 a C99 feature (it's 2017, so now considered safe).
361 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
362 if was a fixed-sized field and bitmask ops via macros; it is now more
365 PP/07 GitHub PR 56: Apply MariaDB build fix.
366 Patch provided by Jaroslav Škarvada.
368 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
369 during Coverity cleanups [4.87 JH/47]
370 Diagnosis and fix provided by Michael Fischer v. Mollard.
372 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
373 the right size to place the terminating semicolon on its own folded
374 line, the header hash was calculated to an incorrect value thanks to
375 the (relaxed) space the fold became.
377 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
378 and confused the parent.
380 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
381 which could crash as a result. This could lead to undeliverable messages.
383 JH/28 Logging: "next input sent too soon" now shows where input was truncated
386 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
387 matters on fast-turnover and PID-randomising systems, which were getting
388 out-of-order delivery.
390 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
391 a possibly-overlapping copy. The symptom was that "Remote host closed
392 connection in response to HELO" was logged instead of the actual 4xx
395 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
396 Previously only that bufferd was discarded, resulting in SYMTP command
399 JH/32 DKIM: when a message has multiple signatures matching an identity given
400 in dkim_verify_signers, run the dkim acl once for each. Previously only
401 one run was done. Bug 2189.
403 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
404 "panic the current process" to "deliberately defer". The panic log is
405 still written with the problem list name; the mail and reject logs now
406 get a temp-reject line for the message that was being handled, saying
407 something like "domains check lookup or other defer". The SMTP 451
408 message is still "Temporary local problem".
410 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
411 A crafted sequence of BDAT commands could result in in-use memory beeing
412 freed. CVE-2017-16943.
414 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
415 from SMTP input. Previously it was always done; now only done for DATA
416 and not BDAT commands. CVE-2017-16944.
418 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
419 to the message (such as an overlong header line). Previously this was
420 not done and we did not exit BDAT mode. Followon from the previous item
421 though a different problem.
427 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
428 than -2003 did; needs libidn2 in addition to libidn.
430 JH/02 The path option on a pipe transport is now expanded before use.
432 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
433 Patch provided by "Björn", documentation fix added too.
435 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
436 missing a wire-to-host endian conversion.
438 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
439 close after a BDAT command line could be taken as a following command,
440 giving a synch failure. Fix by only checking for synch immediately
441 before acknowledging the chunk.
443 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
444 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
445 macro. Patches provided by Josh Soref.
447 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
448 Previously we did not; the RFC seems ambiguous and VRFY is not listed
449 by IANA as a service extension. However, John Klensin suggests that we
452 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
453 the dkim code may be unix-mode line endings rather than smtp wire-format
454 CRLF, so prepend a CR to any bare LF.
456 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
457 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
459 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
460 the first were themselves being wrongly included in the feed into dkim
461 processing; with most chunk sizes in use this resulted in an incorrect
462 body hash calculated value.
464 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
465 DKIM signature block, for verification. Although advised against by
466 standards it is specifically not ruled illegal.
468 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
470 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
471 missing a body hash (the bh= tag).
473 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
474 It seems that HAProxy sends the Proxy Protocol information in clear and
475 only then does a TLS startup, so do the same.
477 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
478 TCP connections (such as for Spamd) unless the daemon successfully set
479 Fast Open mode on its listening sockets. This fixes breakage seen on
480 too-old kernels or those not configured for Fast Open, at the cost of
481 requiring both directions being enabled for TFO, and TFO never being used
482 by non-daemon-related Exim processes.
484 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
485 endings, at least on the first header line. Try to canonify any that get
486 past that check, despite the cost.
488 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
489 now limited to an arbitrary five deep, while parsing addresses with the
490 strip_excess_angle_brackets option enabled.
492 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
493 instead leave the unprompted TLS handshake in socket buffer for the
494 TLS library to consume.
496 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
498 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
500 JH/16 Drop variables when they go out of scope. Memory management drops a whole
501 region in one operation, for speed, and this leaves assigned pointers
502 dangling. Add checks run only under the testsuite which checks all
503 variables at a store-reset and panics on a dangling pointer; add code
504 explicitly nulling out all the variables discovered. Fixes one known
505 bug: a transport crash, where a dangling pointer for $sending_ip_address
506 originally assigned in a verify callout, is re-used.
508 PP/06 Drop '.' from @INC in various Perl scripts.
510 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
512 PP/08 Reduce a number of compilation warnings under clang; building with
513 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
514 should be warning-free.
516 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
518 HS/01 Fix portability problems introduced by PP/08 for platforms where
519 realloc(NULL) is not equivalent to malloc() [SunOS et al].
521 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
522 chunk. This allows us to accept broken chunked messages. We need a more
523 general solution here.
525 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
526 already-broken messages in the queue.
528 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
530 JH/19 Fix reference counting bug in routing-generated-address tracking.
536 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
537 supports it and a size is available (ie. the sending peer gave us one).
539 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
540 years of being deprecated). The replacements are the ACLs
541 acl_smtp_mime and acl_not_smtp_mime.
543 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
544 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
545 or even an in-clear connection were permitted. Now, if the host lookup
546 was dnssec and dane was requested then the host is only used if the
547 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
548 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
549 if one fails this test.
550 This means that a poorly-configured remote DNS will make it incommunicado;
551 but it protects against a DNS-interception attack on it.
553 JH/04 Bug 1810: make continued-use of an open smtp transport connection
554 non-noisy when a race steals the message being considered.
556 JH/05 If main configuration option tls_certificate is unset, generate a
557 self-signed certificate for inbound TLS connections.
559 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
560 in rewrites and routers.
562 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
563 and logged a warning sing 4.83; now they are a configuration file error.
565 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
566 (lacking @domain). Apply the same qualification processing as RCPT.
568 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
570 JH/10 Support ${sha256:} applied to a string (as well as the previous
573 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
574 a cutthrough deliver is pending, as we always want to make a connection.
575 This also avoids re-routing the message when later placing the cutthrough
576 connection after a verify cache hit.
577 Do not update it with the verify result either.
579 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
580 when routing results in more than one destination address.
582 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
583 signing (which inhibits the cutthrough capability). Previously only
584 the presence of an option was tested; now an expansion evaluating as
585 empty is permissible (obviously it should depend only on data available
586 when the cutthrough connection is made).
588 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
589 the relevant preceding SMTP command did not note the pipelining mode.
591 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
592 Previously they were not counted.
594 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
595 as one having no matching records. Previously we deferred the message
596 that needed the lookup.
598 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
599 distinguished as "(=".
601 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
602 for missing MX records. Previously it only worked for missing A records.
604 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
606 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
607 after the data-go-ahead and data-ack. Patch from Jason Betts.
609 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
610 even for a "none" policy. Patch from Tony Meyer.
612 JH/22 Fix continued use of a connection for further deliveries. If a port was
613 specified by a router, it must also match for the delivery to be
616 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
617 When one of the recipients of a message was unsuitable for the connection
618 (has no matching addresses), we lost track of needing to mark it
619 deferred. As a result mail would be lost.
621 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
623 JH/25 Decoding ACL controls is now done using a binary search; the source code
624 takes up less space and should be simpler to maintain. Merge the ACL
625 condition decode tables also, with similar effect.
627 JH/26 Fix problem with one_time used on a redirect router which returned the
628 parent address unchanged. A retry would see the parent address marked as
629 delivered, so not attempt the (identical) child. As a result mail would
632 JH/27 Fix a possible security hole, wherein a process operating with the Exim
633 UID can gain a root shell. Credit to http://www.halfdog.net/ for
634 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
637 JH/28 Enable {spool,log} filesystem space and inode checks as default.
638 Main config options check_{log,spool}_{inodes,space} are now
639 100 inodes, 10MB unless set otherwise in the configuration.
641 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
642 Previously it only applied to the main-section connection policy
645 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
647 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
648 by me. Added RFC7919 DH primes as an alternative.
650 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
651 are not in the system include path.
653 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
654 GnuTLS, when a session startup failed (eg because the client disconnected)
655 Exim did stdio operations after fclose. This was exposed by a recent
656 change which nulled out the file handle after the fclose.
658 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
659 signed directly by the cert-signing cert, rather than an intermediate
660 OCSP-signing cert. This is the model used by LetsEncrypt.
662 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
664 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
665 an incoming connection.
667 HS/02 Bug 1802: Do not half-close the connection after sending a request
670 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
671 fallback to "prime256v1".
673 JH/34 SECURITY: Use proper copy of DATA command in error message.
674 Could leak key material. Remotely exploitable. CVE-2016-9963.
680 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
681 and 3.4.4 - once the server is enabled to respond to an OCSP request
682 it does even when not requested, resulting in a stapling non-aware
683 client dropping the TLS connection.
685 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
686 support variable-length bit vectors. No functional change.
688 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
689 The I= interface field on outgoing lines is now after the H= remote
690 host field, same as incoming lines. There is a separate
691 outgoing_interface log selector which allows you to disable the
694 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
695 If not running log_selector +smtp_connection the mainlog would be held
696 open indefinitely after a "too many connections" event, including to a
697 deleted file after a log rotate. Leave the per net connection logging
698 leaving it open for efficiency as that will be quickly detected by the
699 check on the next write.
701 HS/01 Bug 1671: Fix post transport crash.
702 Processing the wait-<transport> messages could crash the delivery
703 process if the message IDs didn't exist for some reason. When
704 using 'split_spool_directory=yes' the construction of the spool
705 file name failed already, exposing the same netto behaviour.
707 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
708 mime_regex ACL conditions.
710 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
711 to DSN fail messages (bounces): remote IP, remote greeting, remote response
712 to HELO, local diagnostic string.
714 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
715 log line to debug. Even when configured with a tls authenticator many
716 client connections are expected to not authenticate in this way, so
717 an authenticate fail is not an error.
719 HS/02 Add the Exim version string to the process info. This way exiwhat
720 gives some more detail about the running daemon.
722 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
723 matter for fast-change records such as DNSBLs.
725 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
726 retry record, even if constant. There may be multiple transports with
727 different interface settings and the retry behaviour needs to be kept
730 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
732 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
734 JH/10 Bug 840: fix log_defer_output option of pipe transport
736 JH/11 Bug 830: use same host for all RCPTS of a message, even under
737 hosts_randomize. This matters a lot when combined with mua_wrapper.
739 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
740 ${quote_pgsql:<string>} operator.
742 JH/13 Bug 1708: avoid misaligned access in cached lookup.
744 JH/14 Change header file name for freeradius-client. Relevant if compiling
745 with Radius support; from the Gentoo tree and checked under Fedora.
747 JH/15 Bug 1712: Introduce $prdr_requested flag variable
749 JH/16 Bug 1714: Permit an empty string as expansion result for transport
750 option transport_filter, meaning no filtering.
752 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
754 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
755 defaults to "*" (all hosts). The variable is now available when not built
756 with TLS, default unset, mainly to enable keeping the testsuite sane.
757 If a server certificate is not supplied (via tls_certificate) an error is
758 logged, and clients will find TLS connections fail on startup. Presumably
759 they will retry in-clear.
760 Packagers of Exim are strongly encouraged to create a server certificate
761 at installation time.
763 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
764 with the $config_file variable.
766 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
767 in transport context, after the attempt, and per-recipient. The latter type
768 is per host attempted. The event data is the error message, and the errno
769 information encodes the lookup type (A vs. MX) used for the (first) host,
770 and the trailing two digits of the smtp 4xx response.
772 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
773 to write to mainlog (or rejectlog, paniclog) in the window between file
774 creation and permissions/ownership being changed. Particularly affects
775 installations where exicyclog is run as root, rather than exim user;
776 result is that the running daemon panics and dies.
778 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
780 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
781 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
782 "pri" and "weight". Note that the previous implicit priority given by the
783 list order is no longer honoured.
785 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
788 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
789 by defining SUPPORT_SOCKS.
791 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
792 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
793 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
794 variables are renamed to proxy_{local,external}_{address,port}.
796 JH/27 Move Internationalisation support from Experimental to mainline, enabled
797 for a build by defining SUPPORT_I18N
799 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
800 of the query string, and make ${quote_redis:} do that quoting.
802 JH/29 Move Events support from Experimental to mainline, enabled by default
803 and removable for a build by defining DISABLE_EVENT.
805 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
807 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
808 cached by the daemon.
810 JH/32 Move Redis support from Experimental to mainline, enabled for a build
811 by defining LOOKUP_REDIS. The libhiredis library is required.
813 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
814 keys are given for lookup.
816 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
817 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
818 only supported when built with TLS support. The PolarSSL SHA routines
819 are still used when the TLS library is too old for convenient support.
821 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
822 openssl_options), for security. OpenSSL forces this from version 1.1.0
823 server-side so match that on older versions.
825 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
826 allocation for $value could be released as the expansion processing
827 concluded, but leaving the global pointer active for it.
829 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
830 and to use the domains and local_parts ACL conditions.
832 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
833 incorrectly not doubled on cutthrough transmission, hence seen as a
834 body-termination at the receiving system - resulting in truncated mails.
835 Commonly the sender saw a TCP-level error, and retransmitted the message
836 via the normal store-and-forward channel. This could result in duplicates
837 received - but deduplicating mailstores were liable to retain only the
838 initial truncated version.
840 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
842 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
844 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
845 we're in there, support oversigning also; bug 1309.
847 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
849 HS/04 Add support for keep_environment and add_environment options.
851 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
852 either intentional arithmetic overflow during PRNG, or testing config-
855 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
856 delivery resulted in actual delivery. Cancel cutthrough before DATA
859 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
860 rejects a recipient: pass the reject to the originator.
862 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
863 Many were false-positives and ignorable, but it's worth fixing the
866 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
867 for the new environment-manipulation done at startup. Move the routines
868 from being local to tls.c to being global via the os.c file.
870 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
871 an extract embedded as result-arg for a map, the first arg for extract
872 is unavailable so we cannot tell if this is a numbered or keyed
873 extraction. Accept either.
879 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
882 JH/02 The smtp transport option "multi_domain" is now expanded.
884 JH/03 The smtp transport now requests PRDR by default, if the server offers
887 JH/04 Certificate name checking on server certificates, when exim is a client,
888 is now done by default. The transport option tls_verify_cert_hostnames
889 can be used to disable this per-host. The build option
890 EXPERIMENTAL_CERTNAMES is withdrawn.
892 JH/05 The value of the tls_verify_certificates smtp transport and main options
893 default to the word "system" to access the system default CA bundle.
894 For GnuTLS, only version 3.0.20 or later.
896 JH/06 Verification of the server certificate for a TLS connection is now tried
897 (but not required) by default. The verification status is now logged by
898 default, for both outbound TLS and client-certificate supplying inbound
901 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
904 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
905 Status Notification (bounce) messages are now MIME format per RFC 3464.
906 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
907 under the control of the dsn_advertise_hosts option, and routers may
908 have a dsn_lasthop option.
910 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
911 default, modifiable by a malware= option. The list separator for
912 the options can now be changed in the usual way. Bug 68.
914 JH/10 The smtp_receive_timeout main option is now expanded before use.
916 JH/11 The incoming_interface log option now also enables logging of the
917 local interface on delivery outgoing connections.
919 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
920 if the interface and destination host and port all match.
922 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
925 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
926 Patch from Andrew Lewis.
928 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
929 now supports optional time-restrictions, weighting, and priority
930 modifiers per server. Patch originally by <rommer@active.by>.
932 JH/16 The spamd_address main option now supports a mixed list of local
933 and remote servers. Remote servers can be IPv6 addresses, and
934 specify a port-range.
936 JH/17 Bug 68: The spamd_address main option now supports an optional
937 timeout value per server.
939 JH/18 Bug 1581: Router and transport options headers_add/remove can
940 now have the list separator specified.
942 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
945 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
948 JH/21 Support for the A6 type of dns record is withdrawn.
950 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
951 rather than the verbs used.
953 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
954 from 255 to 1024 chars.
956 JH/24 Verification callouts now attempt to use TLS by default.
958 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
959 are generic router options now. The defaults didn't change.
961 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
962 Original patch from Alexander Shikoff, worked over by JH.
964 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
965 files if ZCAT_COMMAND is not executable.
967 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
969 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
971 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
972 Normally benign, it bites when the pair was led to by a CNAME;
973 modern usage is to not canonicalize the domain to a CNAME target
974 (and we were inconsistent anyway for A-only vs AAAA+A).
976 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
978 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
979 when evaluating $sender_host_dnssec.
981 JH/31 Check the HELO verification lookup for DNSSEC, adding new
982 $sender_helo_dnssec variable.
984 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
986 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
988 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
990 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
991 documented as working, but never had. Support all but $spam_report.
993 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
994 added for tls authenticator.
996 HS/03 Add perl_taintmode main config option
1002 TL/01 When running the test suite, the README says that variables such as
1003 no_msglog_check are global and can be placed anywhere in a specific
1004 test's script, however it was observed that placement needed to be near
1005 the beginning for it to behave that way. Changed the runtest perl
1006 script to read through the entire script once to detect and set these
1007 variables, reset to the beginning of the script, and then run through
1008 the script parsing/test process like normal.
1010 TL/02 The BSD's have an arc4random API. One of the functions to induce
1011 adding randomness was arc4random_stir(), but it has been removed in
1012 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
1013 function when detected.
1015 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
1016 cause callback expansion.
1018 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
1019 syntax errors in an expansion can be treated as a string instead of
1020 logging or causing an error, due to the internal use of bool_lax
1021 instead of bool when processing it.
1023 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
1024 server certificates when making smtp deliveries.
1026 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
1028 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
1030 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
1032 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
1033 Merged patch from Sebastian Wiedenroth.
1035 JH/05 Fix results-pipe from transport process. Several recipients, combined
1036 with certificate use, exposed issues where response data items split
1037 over buffer boundaries were not parsed properly. This eventually
1038 resulted in duplicates being sent. This issue only became common enough
1039 to notice due to the introduction of connection certificate information,
1040 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1042 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1043 size buffer was used, resulting in syntax errors when an expansion
1046 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1047 version 3.3.6 or later.
1049 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1050 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1051 both become "event_action", the variables become $event_name, $event_data
1052 and $event_defer_errno. There is a new variable $verify_mode, usable in
1053 routers, transports and related events. The tls:cert event is now also
1054 raised for inbound connections, if the main configuration event_action
1057 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1058 early OCSP support, but no stapling (appears to be less than 1.0.0).
1060 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1061 server certificate names available under the smtp transport option
1062 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1065 JH/10 Time-related extraction expansions from certificates now use the main
1066 option "timezone" setting for output formatting, and are consistent
1067 between OpenSSL and GnuTLS compilations. Bug 1541.
1069 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1070 encoded parameter in the incoming message. Bug 1558.
1072 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1073 include certificate info, eximon was claiming there were spoolfile
1076 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1078 JH/14 Log delivery-related information more consistently, using the sequence
1079 "H=<name> [<ip>]" wherever possible.
1081 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1082 are problematic for Debian distribution, omit them from the release
1085 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1087 JH/16 Fix string representation of time values on 64bit time_t architectures.
1090 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1091 output list separator was used.
1096 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1097 checkers that were complaining about end of non-void function with no
1100 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1101 This was a regression introduced in 4.83 by another bugfix.
1103 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1105 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1106 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1112 TF/01 Correctly close the server side of TLS when forking for delivery.
1114 When a message was received over SMTP with TLS, Exim failed to clear up
1115 the incoming connection properly after forking off the child process to
1116 deliver the message. In some situations the subsequent outgoing
1117 delivery connection happened to have the same fd number as the incoming
1118 connection previously had. Exim would try to use TLS and fail, logging
1119 a "Bad file descriptor" error.
1121 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1122 utilities have not been installed.
1124 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1125 temporary space as the ACL may create new global variables.
1127 TL/01 LDAP support uses per connection or global context settings, depending
1128 upon the detected version of the libraries at build time.
1130 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1131 to extract and use the src ip:port in logging and expansions as if it
1132 were a direct connection from the outside internet. PPv2 support was
1133 updated based on HAProxy spec change in May 2014.
1135 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1137 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1138 Properly escape header and check for NULL return.
1140 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1143 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1145 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1146 characters in header names, implemented as a verify condition.
1147 Contributed by Michael Fischer v. Mollard.
1149 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1150 results permerror and temperror. Previous values are deprecated but
1151 still accepted. In a future release, err_perm and err_temp will be
1152 completely removed, which will be a backward incompatibility if the
1153 ACL tests for either of these two old results. Patch contributed by
1154 user bes-internal on the mailing list.
1156 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1158 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1159 selectors, in both main and reject logs.
1161 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1164 JH/07 Add malware type "sock" for talking to simple daemon.
1166 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1168 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1169 routers/transports under cutthrough routing.
1171 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1172 numbers. Touch up "bool" conditional to keep the same definition.
1174 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1176 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1178 JH/12 Expand items in router/transport headers_add or headers_remove lists
1179 individually rather than the list as a whole. Bug 1452.
1181 Required for reasonable handling of multiple headers_ options when
1182 they may be empty; requires that headers_remove items with embedded
1183 colons must have them doubled (or the list-separator changed).
1185 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1186 view the policy declared in the DMARC record. Currently, $dmarc_status
1187 is a combined value of both the record presence and the result of the
1190 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1192 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1193 dnslookup router and the smtp transport (applying to the forward
1196 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1197 of ldap servers used for a specific lookup. Patch provided by Heiko
1200 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1201 New variable $lookup_dnssec_authenticated for observability.
1203 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1204 Patch submitted by Lars Timman.
1206 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1208 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1209 Requires trusted mode and valid format message id, aborts otherwise.
1210 Patch contributed by Heiko Schlichting.
1212 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1213 certextract with support for various fields. Bug 1358.
1215 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1216 is requested by default, modifiable by smtp transport option
1219 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1220 operate on certificate variables to give certificate fingerprints
1221 Also new ${sha256:cert_variable}.
1223 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1225 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1226 Christian Aistleitner.
1228 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1230 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1231 file. Patch from Wolfgang Breyha.
1233 JH/25 Expand the coverage of the delivery $host and $host_address to
1234 client authenticators run in verify callout. Bug 1476.
1236 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1237 align with daemon_smtp_ports. Bug 72.
1239 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1240 support and error reporting did not work properly.
1242 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1243 and is readable. Patch from Andrew Colin Kissa.
1245 TL/14 Enhance documentation of ${run expansion and how it parses the
1246 commandline after expansion, particularly in the case when an
1247 unquoted variable expansion results in an empty value.
1249 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1251 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1252 by renaming away from it. Addresses GH issue 15, reported by
1255 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1257 TL/15 SECURITY: prevent double expansion in math comparison functions
1258 (can expand unsanitized data). Not remotely exploitable.
1265 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1267 PP/02 Make -n do something, by making it not do something.
1268 When combined with -bP, the name of an option is not output.
1270 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1273 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1274 $sender_host_name and config options to manage this, and basic check
1277 PP/05 DSCP support for outbound connections and control modifier for inbound.
1279 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1280 (Only plugin which currently uses this is kerberos4, which nobody should
1281 be using, but we should make it available and other future plugins might
1282 conceivably use it, even though it would break NAT; stuff *should* be
1283 using channel bindings instead).
1285 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1286 name; added for Sendmail compatibility; requires admin caller.
1287 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1288 just ignore it); requires trusted caller.
1289 Also parse but ignore: -Ac -Am -X<logfile>
1292 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1294 TL/02 Add +smtp_confirmation as a default logging option.
1296 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1297 Patch by Magnus Holmgren from 2007-02-20.
1299 TL/04 Bugzilla 1281 - Spec typo.
1300 Bugzilla 1283 - Spec typo.
1301 Bugzilla 1290 - Spec grammar fixes.
1303 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1305 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1307 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1308 mailing list by by Dmitry Isaikin.
1310 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1312 JH/02 Support "G" suffix to numbers in ${if comparisons.
1314 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1316 NM/01 Bugzilla 1197 - Spec typo
1317 Bugzilla 1196 - Spec examples corrections
1319 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1321 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1322 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1325 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1326 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1328 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1329 "acl {{name}{arg}...}", and optional args on acl condition
1332 JH/05 Permit multiple router/transport headers_add/remove lines.
1334 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1336 JH/07 Avoid using a waiting database for a single-message-only transport.
1337 Performance patch from Paul Fisher. Bugzilla 1262.
1339 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1342 JH/09 Add $headers_added variable, with content from use of ACL modifier
1343 add_header (but not yet added to the message). Bugzilla 199.
1345 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1346 Pulled from Bugzilla 817 by Wolfgang Breyha.
1348 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1350 (nb: this is the same fix as in Exim 4.80.1)
1352 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1355 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1358 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1360 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1361 advertise SMTP AUTH mechanism to us, instead of a generic
1362 protocol violation error. Also, make Exim more robust to bad
1363 data from the Dovecot auth socket.
1365 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1367 When a queue runner is handling a message, Exim first routes the
1368 recipient addresses, during which it prunes them based on the retry
1369 hints database. After that it attempts to deliver the message to
1370 any remaining recipients. It then updates the hints database using
1373 So if a recipient address works intermittently, it can get repeatedly
1374 deferred at routing time. The retry hints record remains fresh so the
1375 address never reaches the final cutoff time.
1377 This is a fairly common occurrence when a user is bumping up against
1378 their storage quota. Exim had some logic in its local delivery code
1379 to deal with this. However it did not apply to per-recipient defers
1380 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1382 This change adds a proper retry rule check during routing so that the
1383 final cutoff time is checked against the message's age. We only do
1384 this check if there is an address retry record and there is not a
1385 domain retry record; this implies that previous attempts to handle
1386 the address had the retry_use_local_parts option turned on. We use
1387 this as an approximation for the destination being like a local
1388 delivery, as in LMTP.
1390 I suspect this new check makes the old local delivery cutoff check
1391 redundant, but I have not verified this so I left the code in place.
1393 TF/02 Correct gecos expansion when From: is a prefix of the username.
1395 Test 0254 submits a message to Exim with the header
1399 When I ran the test suite under the user fanf2, Exim expanded
1400 the header to contain my full name, whereas it should have added
1401 a Resent-Sender: header. It erroneously treats any prefix of the
1402 username as equal to the username.
1404 This change corrects that bug.
1406 GF/01 DCC debug and logging tidyup
1407 Error conditions log to paniclog rather than rejectlog.
1408 Debug lines prefixed by "DCC: " to remove any ambiguity.
1410 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1412 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1413 Bug spotted by Jeremy Harris; was flawed since initial commit.
1414 Would have resulted in OCSP responses post-SNI triggering an Exim
1415 NULL dereference and crash.
1417 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1419 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1420 Bug detection, analysis and fix by Samuel Thibault.
1421 Bugzilla 1331, Debian bug #698092.
1423 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1425 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1426 Server implementation by Todd Lyons, client by JH.
1427 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1428 config variable "prdr_enable" controls whether the server
1429 advertises the facility. If the client requests PRDR a new
1430 acl_data_smtp_prdr ACL is called once for each recipient, after
1431 the body content is received and before the acl_smtp_data ACL.
1432 The client is controlled by both of: a hosts_try_prdr option
1433 on the smtp transport, and the server advertisement.
1434 Default client logging of deliveries and rejections involving
1435 PRDR are flagged with the string "PRDR".
1437 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1438 fclose(). Diagnosis by Todd Lyons.
1440 PP/17 Update configure.default to handle IPv6 localhost better.
1441 Patch by Alain Williams (plus minor tweaks).
1444 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1445 This is now consistent with GnuTLS, and is now documented: the
1446 previous undocumented portable approach to treating the option as
1447 unset was to force an expansion failure. That still works, and
1448 an empty string is now equivalent.
1450 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1451 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1452 not performing validation itself.
1454 PP/20 Added force_command boolean option to pipe transport.
1455 Patch from Nick Koston, of cPanel Inc.
1457 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1460 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1462 PP/21 Fix eximon continuous updating with timestamped log-files.
1463 Broken in a format-string cleanup in 4.80, missed when I repaired the
1464 other false fix of the same issue.
1465 Report and fix from Heiko Schlichting.
1468 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1469 Report from Prashanth Katuri.
1471 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1472 It's SecureTransport, so affects any MacOS clients which use the
1473 system-integrated TLS libraries, including email clients.
1475 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1476 using a MIME ACL for non-SMTP local injection.
1477 Report and assistance in diagnosis by Warren Baker.
1479 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1481 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1483 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1484 last id that failed so it may be referenced in subsequent ACL's.
1486 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1489 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1490 ldap library initialization, allowing self-signed CA's to be
1491 used. Also properly sets require_cert option later in code by
1492 using NULL (global ldap config) instead of ldap handle (per
1493 session). Bug diagnosis and testing by alxgomz.
1495 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1496 the src/util/ subdirectory.
1498 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1499 renamed to Transport Post Delivery Action by Jeremy Harris, as
1502 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1503 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1504 redis_servers = needs to be configured which will be used by the redis
1505 lookup. Patch from Warren Baker, of The Packet Hub.
1507 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1509 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1510 hostname or reverse DNS when processing a host list. Used suggestions
1511 from multiple comments on this bug.
1513 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1515 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1516 Missed a few lines, added it to make the runtest require no keyboard
1519 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1520 contains upper case chars. Make router use caseful_local_part.
1522 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1523 support when GnuTLS has been built with p11-kit.
1529 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1531 This, or similar/improved, will also be change PP/11 of 4.82.
1537 PP/01 Handle short writes when writing local log-files.
1538 In practice, only affects FreeBSD (8 onwards).
1539 Bugzilla 1053, with thanks to Dmitry Isaikin.
1541 NM/01 Bugzilla 949 - Documentation tweak
1543 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1546 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1548 PP/02 Implemented gsasl authenticator.
1550 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1552 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1553 `pkg-config foo` for cflags/libs.
1555 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1556 with rest of GSASL and with heimdal_gssapi.
1558 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1559 `pkg-config foo` for cflags/libs for the TLS implementation.
1561 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1562 properties get this fed in as external SSF. A number of robustness
1563 and debugging improvements to the cyrus_sasl authenticator.
1565 PP/08 cyrus_sasl server now expands the server_realm option.
1567 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1568 Patch by Jeremy Harris.
1570 PP/10 Added dbmjz lookup type.
1572 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1574 PP/12 MAIL args handles TAB as well as SP, for better interop with
1575 non-compliant senders.
1576 Analysis and variant patch by Todd Lyons.
1578 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1579 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1580 Patch from Dirk Mueller <dmueller@suse.com>
1582 PP/13 tls_peerdn now print-escaped for spool files.
1583 Observed some $tls_peerdn in wild which contained \n, which resulted
1584 in spool file corruption.
1586 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1587 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1588 or write after TLS renegotiation, which otherwise led to messages
1591 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1592 as a tracking header (ie: a signed header comes before the signature).
1593 Patch from Wolfgang Breyha.
1595 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1596 comma-sep list; embedded commas doubled.
1598 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1600 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1602 Report and patch from Dmitry Banschikov.
1604 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1605 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1606 failures. We appear to now support TLS1.1+ with Exim.
1608 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1609 lets Exim select keys and certificates based upon TLS SNI from client.
1610 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1611 before an outbound SMTP session. New log_selector, +tls_sni.
1613 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1614 NULL dereference. Report and patch from Alun Jones.
1616 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1617 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1618 Not seeing resolver debug output on NetBSD, but suspect this is a
1619 resolver implementation change.
1621 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1622 Left warnings. Added "eximon gdb" invocation mode.
1624 PP/21 Defaulting "accept_8bitmime" to true, not false.
1626 PP/22 Added -bw for inetd wait mode support.
1628 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1629 locate the relevant includes and libraries. Made this the default.
1631 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1632 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1634 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1635 This may cause build issues on older platforms.
1637 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1638 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1639 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1640 Added SNI support via GnuTLS too.
1641 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1643 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1645 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1646 Applied second patch from Janne, implementing suggestion to default
1647 multiple-strings-in-record handling to match SPF spec.
1649 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1651 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1652 read-only, out of scope).
1653 Patch from Wolfgang Breyha, report from Stuart Northfield.
1655 PP/29 Fix three issues highlighted by clang analyser static analysis.
1656 Only crash-plausible issue would require the Cambridge-specific
1657 iplookup router and a misconfiguration.
1658 Report from Marcin Mirosław.
1660 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1662 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1663 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1664 As part of this, removing so much warning spew let me fix some minor
1665 real issues in debug logging.
1667 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1668 assignment on my part. Fixed.
1670 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1671 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1672 Janne Snabb (who went above and beyond: thank you).
1674 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1675 string otherwise requires a connection and a bunch more work and it's
1676 relatively easy to get wrong. Should also expose TLS library linkage
1679 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1680 64-bit ${eval} (JH/03).
1682 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1683 GNU libc to support some of the 64-bit stuff, should not lead to
1684 conflicts. Defined before os.h is pulled in, so if a given platform
1685 needs to override this, it can.
1687 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1688 protection layer was required, which is not implemented.
1689 Bugzilla 1254, patch from Wolfgang Breyha.
1691 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1692 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1693 tls_dhparam take prime identifiers. Also unbreak combination of
1694 OpenSSL+DH_params+TLSSNI.
1696 PP/39 Disable SSLv2 by default in OpenSSL support.
1702 PP/01 Solaris build fix for Oracle's LDAP libraries.
1703 Bugzilla 1109, patch from Stephen Usher.
1705 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1707 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1710 TF/02 Fix a couple more cases where we did not log the error message
1711 when unlink() failed. See also change 4.74-TF/03.
1713 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1714 lock up or crash if it happened to be inside a call to libc when it
1715 got a SIGUSR1 from exiwhat.
1717 The SIGUSR1 handler appends the current process status to the process
1718 log which is later printed by exiwhat. It used to use the general
1719 purpose logging code to do this, but several functions it calls are
1720 not safe for signals.
1722 The new output code in the SIGUSR1 handler is specific to the process
1723 log, and simple enough that it's easy to inspect for signal safety.
1724 Removing some special cases also simplifies the general logging code.
1725 Removing the spurious timestamps from the process log simplifies
1728 TF/04 Improved ratelimit ACL condition.
1730 The /noupdate option has been deprecated in favour of /readonly which
1731 has clearer semantics. The /leaky, /strict, and /readonly update modes
1732 are mutually exclusive. The update mode is no longer included in the
1733 database key; it just determines when the database is updated. (This
1734 means that when you upgrade Exim will forget old rate measurements.)
1736 Exim now checks that the per_* options are used with an update mode that
1737 makes sense for the current ACL. For example, when Exim is processing a
1738 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1739 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1740 must specify per_mail/readonly. If you omit the update mode it defaults to
1741 /leaky where that makes sense (as before) or /readonly where required.
1743 The /noupdate option is now undocumented but still supported for
1744 backwards compatibility. It is equivalent to /readonly except that in
1745 ACLs where /readonly is required you may specify /leaky/noupdate or
1746 /strict/noupdate which are treated the same as /readonly.
1748 A useful new feature is the /count= option. This is a generalization
1749 of the per_byte option, so that you can measure the throughput of other
1750 aggregate values. For example, the per_byte option is now equivalent
1751 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1753 The per_rcpt option has been generalized using the /count= mechanism
1754 (though it's more complicated than the per_byte equivalence). When it is
1755 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1756 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1757 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1758 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1759 Note that using per_rcpt with a non-readonly update mode in more than
1760 one ACL will cause the recipients to be double-counted. (The per_mail
1761 and per_byte options don't have this problem.)
1763 The handling of very low rates has changed slightly. If the computed rate
1764 is less than the event's count (usually one) then this event is the first
1765 after a long gap. In this case the rate is set to the same as this event's
1766 count, so that the first message of a spam run is counted properly.
1768 The major new feature is a mechanism for counting the rate of unique
1769 events. The new per_addr option counts the number of different
1770 recipients that someone has sent messages to in the last time period. It
1771 behaves like per_rcpt if all the recipient addresses are different, but
1772 duplicate recipient addresses do not increase the measured rate. Like
1773 the /count= option this is a general mechanism, so the per_addr option
1774 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1775 example, measure the rate that a client uses different sender addresses
1776 with the options per_mail/unique=$sender_address. There are further
1777 details in the main documentation.
1779 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1781 TF/06 Removed a few PCRE remnants.
1783 TF/07 Automatically extract Exim's version number from tags in the git
1784 repository when doing development or release builds.
1786 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1787 Bugzilla 879. Patch from Paul Fisher.
1789 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1790 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1793 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1795 PP/05 Variable $av_failed, true if the AV scanner deferred.
1796 Bugzilla 1078. Patch from John Horne.
1798 PP/06 Stop make process more reliably on build failure.
1799 Bugzilla 1087. Patch from Heiko Schlittermann.
1801 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1802 Bugzilla 1089. Patch from Heiko Schlittermann.
1804 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1805 Bugzilla 1131. Patch from Holger Weiß.
1807 PP/09 Handle IPv6 addresses with SPF.
1808 Bugzilla 860. Patch from Wolfgang Breyha.
1810 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1812 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1815 PP/11 match_* no longer expand right-hand-side by default.
1816 New compile-time build option, EXPAND_LISTMATCH_RHS.
1817 New expansion conditions, "inlist", "inlisti".
1819 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1821 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1823 PP/14 fix log_write() format string regression from TF/03.
1824 Bugzilla 1152. Patch from Dmitry Isaikin.
1830 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1832 PP/02 Harmonised TLS library version reporting; only show if debugging.
1833 Layout now matches that introduced for other libraries in 4.74 PP/03.
1835 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1837 PP/04 New "dns_use_edns0" global option.
1839 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1842 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1843 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1845 TK/01 Updated PolarSSL code to 0.14.2.
1846 Bugzilla 1097. Patch from Andreas Metzler.
1848 PP/07 Catch divide-by-zero in ${eval:...}.
1849 Fixes bugzilla 1102.
1851 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1854 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1855 format-string attack -- SECURITY: remote arbitrary code execution.
1857 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1858 time unintentionally subject to list matching rules, letting the header
1859 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1860 arbitrary string expansion). This allowed for information disclosure.
1862 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1863 INT_MIN/-1 -- value coerced to INT_MAX.
1869 NM/01 Workaround for PCRE version dependency in version reporting
1872 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1873 This fixes portability to compilers other than gcc, notably
1874 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1876 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1877 makefiles for portability to HP-UX and POSIX correctness.
1879 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1880 Also via indented variable definition in the Makefile.
1881 (Debugging by Oliver Heesakkers).
1883 PP/02 Restore caching of spamd results with expanded spamd_address.
1884 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1886 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1887 Improves build reliability. Fix from: Frank Elsner
1889 NM/02 Fix wide character breakage in the rfc2047 coding
1890 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1892 NM/03 Allow underscore in dnslist lookups
1893 Fixes bug 1026. Patch from Graeme Fowler
1895 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1896 Code patches from Adam Ciarcinski of NetBSD.
1898 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1901 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1902 is logged, to avoid truncation. Patch from John Horne.
1904 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1905 Patch from Jakob Hirsch.
1907 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1908 SQL string expansion failure details.
1909 Patch from Andrey Oktyabrski.
1911 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1912 Patch from Simon Arlott.
1914 PP/09 New lookups functionality failed to compile on old gcc which rejects
1915 extern declarations in function scope.
1916 Patch from Oliver Fleischmann
1918 PP/10 Use sig_atomic_t for flags set from signal handlers.
1919 Check getgroups() return and improve debugging.
1920 Fixed developed for diagnosis in bug 927 (which turned out to be
1923 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1924 Patch from Mark Zealey.
1926 PP/12 Bugzilla 1056: Improved spamd server selection.
1927 Patch from Mark Zealey.
1929 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1930 Based on patch from Heiko Schlittermann.
1932 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1933 Patch from Uwe Doering, sign-off by Michael Haardt.
1935 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1936 variable declaration deep within a block. Bug and patch from
1939 PP/15 lookups-Makefile IRIX compatibility coercion.
1941 PP/16 Make DISABLE_DKIM build knob functional.
1943 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1944 Patch by Simon Arlott
1946 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1947 variable argument macros. Our copy now differs from upstream.
1953 TF/01 Failure to get a lock on a hints database can have serious
1954 consequences so log it to the panic log.
1956 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1957 controlled using the smtp_confirmation log selector.
1959 TF/03 Include the error message when we fail to unlink a spool file.
1961 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1962 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1963 for maintaining out-of-tree patches for some time.
1965 PP/01 Bugzilla 139: Documentation and portability issues.
1966 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1967 Handle per-OS dynamic-module compilation flags.
1969 PP/02 Let /dev/null have normal permissions.
1970 The 4.73 fixes were a little too stringent and complained about the
1971 permissions on /dev/null. Exempt it from some checks.
1972 Reported by Andreas M. Kirchwitz.
1974 PP/03 Report version information for many libraries, including
1975 Exim version information for dynamically loaded libraries. Created
1976 version.h, now support a version extension string for distributors
1977 who patch heavily. Dynamic module ABI change.
1979 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1980 privilege escalation vulnerability whereby the Exim run-time user
1981 can cause root to append content of the attacker's choosing to
1984 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1987 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1988 If dropping privileges for untrusted macros, we disabled normal logging
1989 on the basis that it would fail; for the Exim run-time user, this is not
1990 the case, and it resulted in successful deliveries going unlogged.
1991 Fixed. Reported by Andreas Metzler.
1997 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1998 only prepend for the Resent-* case. Fixes regression introduced in
1999 Exim 4.70 by NM/22 for Bugzilla 607.
2001 PP/02 Include check_rfc2047_length in configure.default because we're seeing
2002 increasing numbers of administrators be bitten by this.
2004 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
2006 PP/03 Bugzilla 994: added openssl_options main configuration option.
2008 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
2010 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
2012 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
2014 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
2015 without a peer certificate, leading to a segfault because of an
2016 assumption that peers always have certificates. Be a little more
2017 paranoid. Problem reported by Martin Tscholak.
2019 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
2020 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
2021 NB: ClamAV planning to remove STREAM in "middle of 2010".
2022 CL also introduces -bmalware, various -d+acl logging additions and
2023 more caution in buffer sizes.
2025 PP/09 Implemented reverse_ip expansion operator.
2027 PP/10 Bugzilla 937: provide a "debug" ACL control.
2029 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
2031 PP/12 Bugzilla 973: Implement --version.
2033 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2035 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2037 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2039 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2040 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2041 ignore trailing whitespace.
2043 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2045 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2048 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2049 Notification from Dr Andrew Aitchison.
2051 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2052 ExtendedDetectionInfo response format.
2053 Notification from John Horne.
2055 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2058 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2059 XSL and documented dependency on system catalogs, with examples of how
2062 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2065 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2066 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2067 configuration file which is writeable by the Exim user or group.
2069 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2070 of configuration files to cover files specified with the -C option if
2071 they are going to be used with root privileges, not just the default
2074 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2075 option (effectively making it always true).
2077 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2078 files to be used while preserving root privileges.
2080 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2081 that rogue child processes cannot use them.
2083 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2084 run-time user, instead of root.
2086 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2087 Exim run-time user without dropping privileges.
2089 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2090 result string, instead of calling string_vformat() twice with the same
2093 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2094 for other users. Others should always drop root privileges if they use
2095 -C on the command line, even for a whitelisted configure file.
2097 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2099 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2105 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2106 $data_path, and $header_path variables; fixed documentation bugs and
2109 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2110 exipick to access non-standard spools, including the "frozen" queue
2113 NM/01 Bugzilla 965: Support mysql stored procedures.
2114 Patch from Alain Williams
2116 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2118 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2119 Patch from Andreas Metzler
2121 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2122 Patch from Kirill Miazine
2124 NM/05 Bugzilla 671: Added umask to procmail example.
2126 JJ/03 installed exipick 20100323.0, fixing doc bug
2128 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2129 directory. Notification and patch from Dan Rosenberg.
2131 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2133 TK/02 Improve log output when DKIM signing operation fails.
2135 MH/01 Treat the transport option dkim_domain as a colon separated
2136 list, not as a single string, and sign the message with each element,
2137 omitting multiple occurences of the same signer.
2139 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2140 Bugzilla 985, 986. Patch by Simon Arlott
2142 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2143 Patch by Simon Arlott
2145 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2146 MBX locking. Notification from Dan Rosenberg.
2152 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2154 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2156 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2158 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2160 NM/04 Bugzilla 915: Flag broken perl installation during build.
2166 TK/01 Added patch by Johannes Berg that expands the main option
2167 "spamd_address" if it starts with a dollar sign.
2169 TK/02 Write list of recipients to X-Envelope-Sender header when building
2170 the mbox-format spool file for content scanning (suggested by Jakob
2173 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2174 (http://www.dcc-servers.net/) support via dccifd. Activated by
2175 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2177 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2178 by Mark Daniel Reidel <mr@df.eu>.
2180 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2181 When building exim an external PCRE library is now needed -
2182 PCRE is a system library on the majority of modern systems.
2183 See entry on PCRE_LIBS in EDITME file.
2185 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2186 conversation. Added nologin parameter to request.
2187 Patch contributed by Kirill Miazine.
2189 TF/01 Do not log submission mode rewrites if they do not change the address.
2191 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2193 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2194 log files in place. Contributed by Roberto Lima.
2196 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2198 TF/03 Bugzilla 615: When checking the local_parts router precondition
2199 after a local_part_suffix or local_part_prefix option, Exim now
2200 does not use the address's named list lookup cache, since this
2201 contains cached lookups for the whole local part.
2203 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2204 Robert Millan. Documentation is in experimental-spec.txt.
2206 TF/04 Bugzilla 668: Fix parallel build (make -j).
2208 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2210 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2211 Patch provided by Jan Srzednicki.
2213 TF/05 Leading white space used to be stripped from $spam_report which
2214 wrecked the formatting. Now it is preserved.
2216 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2217 that they are available at delivery time.
2219 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2221 TF/08 TLS error reporting now respects the incoming_interface and
2222 incoming_port log selectors.
2224 TF/09 Produce a more useful error message if an SMTP transport's hosts
2225 setting expands to an empty string.
2227 NM/06 Bugzilla 744: EXPN did not work under TLS.
2228 Patch provided by Phil Pennock.
2230 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2231 Patch provided by Richard Godbee.
2233 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2234 acl_smtp_notquit, added index entry.
2236 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2237 Patch provided by Eugene Bujak.
2239 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2240 accept(). Patch provided by Maxim Dounin.
2242 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2243 Patch provided by Phil Pennock.
2245 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2247 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2248 Patch provided by Brad "anomie" Jorsch.
2250 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2251 Patch provided by Dean Brooks.
2253 TK/05 Add native DKIM support (does not depend on external libraries).
2255 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2256 Patch provided by Graeme Fowler.
2258 NM/16 Bugzilla 851: Documentation example syntax fix.
2260 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2262 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2265 NM/19 Bugzilla 745: TLS version reporting.
2266 Patch provided by Phil Pennock.
2268 NM/20 Bugzilla 167: bool: condition support.
2269 Patch provided by Phil Pennock.
2271 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2272 clients. Patch provided by Phil Pennock.
2274 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2275 Patch provided by Brad "anomie" Jorsch.
2277 NM/23 Bugzilla 687: Fix misparses in eximstats.
2278 Patch provided by Heiko Schlittermann.
2280 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2281 Patch provided by Heiko Schlittermann.
2283 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2284 plus update to original patch.
2286 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2288 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2289 Patch provided by David Brownlee.
2291 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2293 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2295 NM/30 Bugzilla 888: TLS documentation bugfixes.
2297 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2299 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2300 Unlike the original bugzilla I have changed all shell scripts in src tree.
2302 NM/33 Bugzilla 898: Transport filter timeout fix.
2303 Patch by Todd Rinaldo.
2305 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2306 Patch by Serge Demonchaux.
2308 NM/35 Bugzilla 39: Base64 decode bug fixes.
2309 Patch by Jakob Hirsch.
2311 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2313 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2315 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2317 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2323 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2324 ALT-N's libdkim that I have put here:
2325 http://duncanthrax.net/exim-experimental/
2327 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2328 were called 'true' and it seems that C99 defines that as a reserved
2329 keyword to be used with 'bool' variable types. That means you could
2330 not include C99-style headers which use bools without triggering
2331 build errors in sieve.c.
2333 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2334 as mailq or other aliases. Changed the --help handling significantly
2335 to do whats expected. exim_usage() emits usage/help information.
2337 SC/01 Added the -bylocaldomain option to eximstats.
2339 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2341 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2343 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2349 PH/01 Another patch from the Sieve maintainer.
2351 PH/02 When an IPv6 address is converted to a string for single-key lookup
2352 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2353 dots are used instead of colons so that keys in lsearch files need not
2354 contain colons. This was done some time before quoting was made available
2355 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2356 (notated using the quote facility) so as to distinguish them from IPv4
2357 keys. This meant that lookups for IP addresses in host lists did not work
2358 for iplsearch lookups.
2360 This has been fixed by arranging for IPv6 addresses to be expressed with
2361 colons if the lookup type is iplsearch. This is not incompatible, because
2362 previously such lookups could never work.
2364 The situation is now rather anomalous, since one *can* have colons in
2365 ordinary lsearch keys. However, making the change in all cases is
2366 incompatible and would probably break a number of configurations.
2368 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2371 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2372 conversion specification without a maximum field width, thereby enabling
2373 a rogue spamd server to cause a buffer overflow. While nobody in their
2374 right mind would setup Exim to query an untrusted spamd server, an
2375 attacker that gains access to a server running spamd could potentially
2376 exploit this vulnerability to run arbitrary code as the Exim user.
2378 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2379 $primary_hostname instead of what libspf2 thinks the hosts name is.
2381 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2382 a directory entry by the name of the lookup key. Previously, if a
2383 symlink pointed to a non-existing file or a file in a directory that
2384 Exim lacked permissions to read, a lookup for a key matching that
2385 symlink would fail. Now it is enough that a matching directory entry
2386 exists, symlink or not. (Bugzilla 503.)
2388 PH/03 The body_linecount and body_zerocount variables are now exported in the
2391 PH/04 Added the $dnslist_matched variable.
2393 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2394 This means they are set thereafter only if the connection becomes
2397 PH/06 Added the client_condition to authenticators so that some can be skipped
2398 by clients under certain conditions.
2400 PH/07 The error message for a badly-placed control=no_multiline_responses left
2401 "_responses" off the end of the name.
2403 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2405 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2406 (without spaces) instead of just copying the configuration text.
2408 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2410 PH/11 Added $max_received_linelength.
2412 PH/12 Added +ignore_defer and +include_defer to host lists.
2414 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2415 way in which PCRE > 7.0 is built.
2417 PH/14 Implemented queue_only_load_latch.
2419 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2420 MAIL command. The effect was to mangle the value on 64-bit systems.
2422 PH/16 Another patch from the Sieve maintainer.
2424 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2426 PH/18 If a system quota error occurred while trying to create the file for
2427 a maildir delivery, the message "Mailbox is full" was not appended to the
2428 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2429 only to a quota excession during the actual writing of the file.
2431 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2432 characters?) which causes problems in log lines. The DN values are now
2433 passed through string_printing() before being added to log lines.
2435 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2436 and InterBase are left for another time.)
2438 PH/21 Added message_body_newlines option.
2440 PH/22 Guard against possible overflow in moan_check_errorcopy().
2442 PH/23 POSIX allows open() to be a macro; guard against that.
2444 PH/24 If the recipient of an error message contained an @ in the local part
2445 (suitably quoted, of course), incorrect values were put in $domain and
2446 $local_part during the evaluation of errors_copy.
2452 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2453 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2456 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2457 issue a MAIL command.
2459 PH/02 In an ACL statement such as
2461 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2463 if a client was not listed at all, or was listed with a value other than
2464 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2465 the condition was not true (as it should be), so access was not denied.
2466 The bug was that the ! inversion was incorrectly passed on to the second
2467 item. This has been fixed.
2469 PH/03 Added additional dnslists conditions == and =& which are different from
2470 = and & when the dns lookup returns more than one IP address.
2472 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2473 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2475 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2476 FSYNC, which compiles an option called disable_fsync that allows for
2477 bypassing fsync(). The documentation is heavily laced with warnings.
2479 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2481 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2482 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2483 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2484 including adding "make clean"; (3) Added -fPIC when compiling the test
2485 dynamically loaded module, to get rid of a warning.
2487 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2488 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2489 = 0s. The bug is otherwise harmless.
2491 PH/07 There was a bug in the dovecot authenticator such that the value of
2492 $auth1 could be overwritten, and so not correctly preserved, after a
2493 successful authentication. This usually meant that the value preserved by
2494 the server_setid option was incorrect.
2496 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2498 PH/09 Installed PCRE release 7.0.
2500 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2501 run for batched SMTP input. It is now run at the start of every message
2502 in the batch. While fixing this I discovered that the process information
2503 (output by running exiwhat) was not always getting set for -bs and -bS
2504 input. This is fixed, and it now also says "batched" for BSMTP.
2506 PH/11 Added control=no_pipelining.
2508 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2509 patch, slightly modified), and move the expansion of helo_data till after
2510 the connection is made in the smtp transport (so it can use these
2513 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2515 PH/14 Added log_selector = +pid.
2517 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2519 PH/16 Add ${if forany and ${if forall.
2521 PH/17 Added dsn_from option to vary the From: line in DSNs.
2523 PH/18 Flush SMTP output before performing a callout, unless control =
2524 no_callout_flush is set.
2526 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2527 was true (the default) a successful delivery failed to delete the retry
2528 item, thus causing premature timeout of the address. The bug is now
2531 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2533 PH/21 Long custom messages for fakedefer and fakereject are now split up
2534 into multiline responses in the same way that messages for "deny" and
2535 other ACL rejections are.
2537 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2538 with slight modification.
2540 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2541 draft, changing the syntax and factoring some duplicate code".
2543 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2544 for deliveries of the second and subsequent messages over the same SMTP
2547 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2548 ${reduce, with only minor "tidies".
2550 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2552 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2553 expansion side effects.
2555 PH/27 When a message times out after an over-quota error from an Exim-imposed
2556 quota, the bounce message says "mailbox is full". This message was not
2557 being given when it was a system quota that was exceeded. It now should
2560 MH/03 Made $recipients available in local_scan(). local_scan() already has
2561 better access to the recipient list through recipients_list[], but
2562 $recipients can be useful in postmaster-provided expansion strings.
2564 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2565 in the case of a MAIL command with additional options following the
2566 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2567 were accidentally chopped off.
2569 PH/29 SMTP synchronization checks are implemented when a command is read -
2570 there is a check that no more input is waiting when there shouldn't be
2571 any. However, for some commands, a delay in an ACL can mean that it is
2572 some time before the response is written. In this time, more input might
2573 arrive, invalidly. So now there are extra checks after an ACL has run for
2574 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2575 pipelining has not been advertised.
2577 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2579 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2580 returned in the SMTP response when smtp_return_error_details was false.
2581 This has been fixed.
2583 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2584 instead of the C I/O that was originally supplied, because problems were
2585 reported on Solaris.
2587 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2588 Exim which did not show up earlier: it was assuming that a call to
2589 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2590 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2591 was a macro that became an assignment, so it seemed to work. This has
2592 changed to a proper function call with a void return, hence the compile
2593 error. Exim's code has been fixed.
2595 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2598 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2600 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2602 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2603 because it turns out to be too big or there is a timeout) there is no
2604 "Completed" line in the log. When some messages of this type were
2605 selected by exigrep, they were listed as "not completed". Others were
2606 picked up by some special patterns. I have improved the selection
2607 criteria to be more general.
2609 PH/38 The host_find_failed option in the manualroute router can now be set
2610 to "ignore", to completely ignore a host whose IP address cannot be
2611 found. If all hosts are ignored, the behaviour is controlled by the new
2612 host_all_ignored option.
2614 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2615 homing or because of multiple MX records with /mx) generated more than
2616 one IP address, and the following item turned out to be the local host,
2617 all the secondary addresses of the first item were incorrectly removed
2618 from the list, along with the local host and any following hosts (which
2619 is what is supposed to happen).
2621 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2622 whoever called Exim into the -H file. In the case of the daemon it was
2623 behaving confusingly. When first started, it used values for whoever
2624 started the daemon, but after a SIGHUP it used the Exim user (because it
2625 calls itself on a restart). I have changed the code so that it now always
2628 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2629 message are rejected with the same error (e.g. no authentication or bad
2630 sender address), and a DATA command is nevertheless sent (as can happen
2631 with PIPELINING or a stupid MUA), the error message that was given to the
2632 RCPT commands is included in the rejection of the DATA command. This is
2633 intended to be helpful for MUAs that show only the final error to their
2636 PH/42 Another patch from the Sieve maintainer.
2638 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2639 Eximstats - Fixed some broken HTML links and added missing column headers
2641 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2642 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2644 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2646 PH/43 Yet another patch from the Sieve maintainer.
2648 PH/44 I found a way to check for a TCP/IP connection going away before sending
2649 the response to the final '.' that terminates a message, but only in the
2650 case where the client has not sent further data following the '.'
2651 (unfortunately, this is allowed). However, in many cases there won't be
2652 any further data because there won't be any more messages to send. A call
2653 to select() can be used: if it shows that the input is "ready", there is
2654 either input waiting, or the socket has been closed. An attempt to read
2655 the next input character can distinguish the two cases. Previously, Exim
2656 would have sent an OK response which the client would never have see.
2657 This could lead to message repetition. This fix should cure that, at
2658 least in a lot of common cases.
2660 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2661 advertised in response to EHLO.
2667 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2668 fixed by 4.65/MH/01 (is this a record?) are fixed:
2670 (i) An empty string was always treated as zero by the numeric comparison
2671 operators. This behaviour has been restored.
2673 (ii) It is documented that the numeric comparison operators always treat
2674 their arguments as decimal numbers. This was broken in that numbers
2675 starting with 0 were being interpreted as octal.
2677 While fixing these problems I realized that there was another issue that
2678 hadn't been noticed. Values of message_size_limit (both the global option
2679 and the transport option) were treated as octal if they started with 0.
2680 The documentation was vague. These values are now always treated as
2681 decimal, and I will make that clear in the documentation.
2687 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2688 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2691 MH/01 Don't check that the operands of numeric comparison operators are
2692 integers when their expansion is in "skipping" mode (fixes bug
2693 introduced by 4.64-PH/07).
2695 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2696 child addresses, Exim now panics and dies. Previously, because the count
2697 is held in a short int, deliveries were likely to be lost. As such a
2698 large number of recipients for a single message is ridiculous
2699 (performance will be very, very poor), I have chosen to impose a limit
2700 rather than extend the field.
2706 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2707 leftover -K file (the existence of which was triggered by #402).
2708 While we were at it, introduced process PID as part of the -K
2709 filename. This should rule out race conditions when creating
2712 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2713 processing considerably. Previous code took too long for large mails,
2714 triggering a timeout which in turn triggers #401.
2716 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2717 in the DK code in transports.c. sendfile() is not really portable,
2718 hence the _LINUX specificness.
2720 TF/01 In the add_headers option to the mail command in an Exim filter,
2721 there was a bug that Exim would claim a syntax error in any
2722 header after the first one which had an odd number of characters
2725 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2726 callout verification, Exim cached a "reject" for the entire domain. This
2727 is correct for most verifications, but it is not correct for a recipient
2728 verification with use_sender or use_postmaster set, because in that case
2729 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2730 case of MAIL FROM:<> rejection from other early rejections (e.g.
2731 rejection of HELO). When verifying a recipient using a non-null MAIL
2732 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2733 Whatever the result of the callout, the value of the domain cache is
2734 left unchanged (for any other kind of callout, getting as far as trying
2735 RCPT means that the domain itself is ok).
2737 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2740 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2741 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2742 session, the daemon ignores ECONNECT errors and logs others; it now
2743 ignores EPIPE as well.
2745 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2746 (quoted-printable decoding).
2748 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2749 later the small subsequent patch to fix an introduced bug.
2751 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2753 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2755 PH/08 An error is now given if message_size_limit is specified negative.
2757 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2758 to be given (somewhat) arbitrary names.
2760 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2763 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2764 miscellaneous code fixes
2766 PH/10 Added the log_reject_target ACL modifier to specify where to log
2769 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2770 hostname. This is wrong, because it relates to the incoming message (and
2771 probably the interface on which it is arriving) and not to the outgoing
2772 callout (which could be using a different interface). This has been
2773 changed to use the value of the helo_data option from the smtp transport
2774 instead - this is what is used when a message is actually being sent. If
2775 there is no remote transport (possible with a router that sets up host
2776 addresses), $smtp_active_hostname is used.
2778 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2779 tweaks were necessary in order to get it to work (see also 21 below):
2780 (a) The code assumed that strncpy() returns a negative number on buffer
2781 overflow, which isn't the case. Replaced with Exim's string_format()
2783 (b) There were several signed/unsigned issues. I just did the minimum
2784 hacking in of casts. There is scope for a larger refactoring.
2785 (c) The code used strcasecmp() which is not a standard C function.
2786 Replaced with Exim's strcmpic() function.
2787 (d) The code set only $1; it now sets $auth1 as well.
2788 (e) A simple test gave the error "authentication client didn't specify
2789 service in request". It would seem that Dovecot has changed its
2790 interface. Fortunately there's a specification; I followed it and
2791 changed what the client sends and it appears to be working now.
2793 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2796 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2797 address A is aliased to B and C, where B exists and C does not. Without
2798 -v the output is "A verified" because verification stops after a
2799 successful redirection if more than one address is generated. However,
2800 with -v the child addresses are also verified. Exim was outputting "A
2801 failed to verify" and then showing the successful verification for C,
2802 with its parentage. It now outputs "B failed to verify", showing B's
2803 parentage before showing the successful verification of C.
2805 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2806 look up a TXT record in a specific list after matching in a combined
2809 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2810 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2811 they consult the DNS. I had assumed they would set it the way they
2812 wanted; and indeed my experiments on Linux seem to show that in some
2813 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2814 To be on the safe side, however, I have now made the interface to
2815 host_find_byname() similar to host_find_bydns(), with an argument
2816 containing the DNS resolver options. The host_find_byname() function now
2817 sets these options at its start, just as host_find_bydns() does. The smtp
2818 transport options dns_qualify_single and dns_search_parents are passed to
2819 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2820 of host_find_byname() use the default settings of RES_DEFNAMES
2821 (qualify_single) but not RES_DNSRCH (search_parents).
2823 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2824 spool_read_header() do less string testing, by means of a preliminary
2825 switch on the second character of optional "-foo" lines. (This is
2826 overdue, caused by the large number of possibilities that now exist.
2827 Originally there were few.) While I was there, I also converted the
2828 str(n)cmp tests so they don't re-test the leading "-" and the first
2829 character, in the hope this might squeeze out yet more improvement.
2831 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2832 flag allowing group syntax was set by the header_syntax check but not
2833 turned off, possible causing trouble later; (2) The flag was not being
2834 set at all for the header_verify test, causing "group"-style headers to
2835 be rejected. I have now set it in this case, and also caused header_
2836 verify to ignore an empty address taken from a group. While doing this, I
2837 came across some other cases where the code for allowing group syntax
2838 while scanning a header line wasn't quite right (mostly, not resetting
2839 the flag correctly in the right place). These bugs could have caused
2840 trouble for malformed header lines. I hope it is now all correct.
2842 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2843 with the "reply" argument non-NULL. The code, however (which originally
2844 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2845 but it didn't always do it. This confused somebody who was copying the
2846 code for some other use. I have removed all the tests.
2848 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2849 feature that was used to support insecure browsers during the U.S. crypto
2850 embargo. It requires special client support, and Exim is probably the
2851 only MTA that supported it -- and would never use it because real RSA is
2852 always available. This code has been removed, because it had the bad
2853 effect of slowing Exim down by computing (never used) parameters for the
2854 RSA_EXPORT functionality.
2856 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2857 authenticator to fail if there's a tab character in the incoming data
2858 (there should never be unless someone is messing about, as it's supposed
2859 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2860 if the connection is using TLS or if the remote IP is the same as the
2861 local IP, and the "valid-client-cert option" if a client certificate has
2864 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2865 authenticators. This can be used for authorization after authentication
2866 succeeds. (In the case of plaintext, it servers for both authentication
2869 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2870 if any retry times were supplied.
2872 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2873 connection, where there is no client IP address to check. In this
2874 situation, the verify now always succeeds.
2876 PH/25 Applied John Jetmore's -Mset patch.
2878 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2880 PH/27 In a string expansion for a processed (not raw) header when multiple
2881 headers of the same name were present, leading whitespace was being
2882 removed from all of them, but trailing whitespace was being removed only
2883 from the last one. Now trailing whitespace is removed from each header
2884 before concatenation. Completely empty headers in a concatenation (as
2885 before) are ignored.
2887 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2888 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2890 PH/29 [Removed. This was a change that I later backed out, and forgot to
2891 correct the ChangeLog entry (that I had efficiently created) before
2892 committing the later change.]
2894 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2895 address errors (4xx response to RCPT) over the same connection as other
2896 messages routed to the same hosts. Such deliveries are always "forced",
2897 so retry times are not inspected. This resulted in far too many retries
2898 for the affected addresses. The effect occurred only when there were more
2899 hosts than the hosts_max_try setting in the smtp transport when it had
2900 the 4xx errors. Those hosts that it had tried were not added to the list
2901 of hosts for which the message was waiting, so if all were tried, there
2902 was no problem. Two fixes have been applied:
2904 (i) If there are any address or message errors in an SMTP delivery, none
2905 of the hosts (tried or untried) are now added to the list of hosts
2906 for which the message is waiting, so the message should not be a
2907 candidate for sending over the same connection that was used for a
2908 successful delivery of some other message. This seems entirely
2909 reasonable: after all the message is NOT "waiting for some host".
2910 This is so "obvious" that I'm not sure why it wasn't done
2911 previously. Hope I haven't missed anything, but it can't do any
2912 harm, as the worst effect is to miss an optimization.
2914 (ii) If, despite (i), such a delivery is accidentally attempted, the
2915 routing retry time is respected, so at least it doesn't keep
2916 hammering the server.
2918 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2919 in ${readsocket because some servers need this prod.
2921 PH/32 Added some extra debug output when updating a wait-xxx database.
2923 PH/33 The hint "could be header name not terminated by colon", which has been
2924 given for certain expansion errors for a long time, was not being given
2925 for the ${if def:h_colon_omitted{... case.
2927 PH/34 The spec says: "With one important exception, whenever a domain list is
2928 being scanned, $domain contains the subject domain." There was at least
2929 one case where this was not true.
2931 PH/35 The error "getsockname() failed: connection reset by peer" was being
2932 written to the panic log as well as the main log, but it isn't really
2933 panic-worthy as it just means the connection died rather early on. I have
2934 removed the panic log writing for the ECONNRESET error when getsockname()
2937 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2938 runs only) independently of the message's sender address. This meant
2939 that, if the 4xx error was in fact related to the sender, a different
2940 message to the same recipient with a different sender could confuse
2941 things. In particular, this can happen when sending to a greylisting
2942 server, but other circumstances could also provoke similar problems.
2943 I have changed the default so that the retry time for these errors is now
2944 based a combination of the sender and recipient addresses. This change
2945 can be overridden by setting address_retry_include_sender=false in the
2948 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2949 remote server are returned as part of bounce messages. This was not
2950 happening for LMTP over a pipe (the lmtp transport), but now it is the
2951 same for both kinds of LMTP.
2953 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2954 in header lines that were in fact CNAMEs. This is no longer the case.
2956 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2957 and queue runs started by the daemon processed all messages. This has
2958 been fixed so that -R and -S can now usefully be given with -q<time>.
2960 PH/40 Import PCRE release 6.7 (fixes some bugs).
2962 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2964 PH/42 Give an error if -q is specified more than once.
2966 PH/43 Renamed the variables $interface_address and $interface_port as
2967 $received_ip_address and $received_port, to make it clear that these
2968 values apply to message reception, and not to the outgoing interface when
2969 a message is delivered. (The old names remain recognized, of course.)
2971 PH/44 There was no timeout on the connect() call when using a Unix domain
2972 socket in the ${readsocket expansion. There now is.
2974 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2975 be meaningful with "accept".
2977 SC/01 Eximstats V1.43
2978 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2980 SC/02 Eximstats V1.44
2981 Use a glob alias rather than an array ref in the generated
2982 parser. This improves both readability and performance.
2984 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2985 Collect SpamAssassin and rejection statistics.
2986 Don't display local sender or destination tables unless
2987 there is data to show.
2988 Added average volumes into the top table text output.
2990 SC/04 Eximstats V1.46
2991 Collect data on the number of addresses (recipients)
2992 as well as the number of messages.
2994 SC/05 Eximstats V1.47
2995 Added 'Message too big' to the list of mail rejection
2996 reasons (thanks to Marco Gaiarin).
2998 SC/06 Eximstats V1.48
2999 Mainlog lines which have GMT offsets and are too short to
3000 have a flag are now skipped.
3002 SC/07 Eximstats V1.49 (Alain Williams)
3003 Added the -emptyok flag.
3005 SC/08 Eximstats V1.50
3006 Fixes for obtaining the IP address from reject messages.
3008 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
3009 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
3010 whitespace changes from 4.64-PH/27
3012 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
3015 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
3016 are found, allow negative numbers in numeric criteria)
3018 JJ/06 exipick.20061117.2, added new $message_body_missing variable
3020 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
3021 to match changes made in 4.64-PH/43
3023 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
3025 PH/47 Put in an explicit test for a DNS lookup of an address record where the
3026 "domain" is actually an IP address, and force a failure. This locks out
3027 those revolvers/nameservers that support "A-for-A" lookups, in
3028 contravention of the specifications.
3030 PH/48 When a host name was looked up from an IP address, and the subsequent
3031 forward lookup of the name timed out, the host name was left in
3032 $sender_host_name, contrary to the specification.
3034 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3035 restricted to single-key lookups, Exim was not diagnosing an error if
3036 * or *@ was used with a query-style lookup.
3038 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3040 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3041 long ago, but noone interested enough thought of it. Let's just say that
3042 the "1.1" means that there are some new functions that weren't there at
3043 some point in the past.
3045 PH/51 Error processing for expansion failure of helo_data from an smtp
3046 transport during callout processing was broken.
3048 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3049 tested/used via the -bh/-bhc/-bs options.
3051 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3052 bug, fixed in subsequent PCRE releases).
3054 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3055 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3057 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3063 SC/01 Use a glob alias rather than an array ref in eximstats generated
3064 parser. This improves both readability and performance.
3066 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3067 Don't display local sender or destination tables in eximstats unless
3068 there is data to show.
3069 Added average volumes into the eximstats top table text output.
3071 SC/03 Collect data on the number of addresses (recipients) as well
3072 as the number of messages in eximstats.
3074 TF/01 Correct an error in the documentation for the redirect router. Exim
3075 does (usually) call initgroups() when daemonizing.
3077 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3078 with consistent privilege compared to when running as a daemon.
3080 TF/03 Note in the spec that $authenticated_id is not set for local
3081 submissions from trusted users.
3083 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3084 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3086 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3087 by adding some example configuration directives to the default
3088 configuration file. A little bit of work is required to uncomment the
3089 directives and define how usernames and passwords are checked, but
3090 there is now a framework to start from.
3092 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3093 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3094 without this. I don't know how relevant this is to other LDAP libraries.
3096 PH/02 Add the verb name to the "unknown ACL verb" error.
3098 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3100 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3102 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3103 directory not expanded when it should be if an expanded home directory
3104 was set for the address (which is overridden by the transport).
3106 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3109 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3110 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3111 because it is too late at that time, and has no effect.
3113 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3114 security issue with \' (bugzilla #107). I could not use the
3115 PQescapeStringConn() function, because it needs a PGconn value as one of
3118 PH/08 When testing addresses using -bt, indicate those final addresses that
3119 are duplicates that would not cause an additional delivery. At least one
3120 person was confused, thinking that -bt output corresponded to deliveries.
3121 (Suppressing duplicates isn't a good idea as you lose the information
3122 about possibly different redirections that led to the duplicates.)
3124 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3125 systems where poll() doesn't work, in particular OS X.
3127 PH/10 Added more information to debugging output for retry time not reached.
3129 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3130 operations in malware.c.
3132 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3135 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3136 syslog_duplication set false, log lines that would normally be written
3137 both the the main log and to the reject log were not written to syslog at
3140 PH/14 In the default configuration, change the use of "message" in ACL warn
3141 statements to "add_header".
3143 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3144 not followed by a command (e.g. "seen endif").
3146 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3147 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3150 PH/17 Added extra conditions to the default value of delay_warning_condition
3154 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3155 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3156 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3159 The Auto-Submitted: and various List- headers are standardised, whereas I
3160 don't think Precedence: ever was.
3162 PH/18 Refactored debugging code in route_finduser() to show more information,
3163 in particular, the error code if getpwnam() issues one.
3165 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3166 This is apparently needed in addition to the PH/07 change above to avoid
3167 any possible encoding problems.
3169 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3170 but not after initializing Perl.
3172 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3173 output them only if debugging. By default they are written stderr,
3174 apparently, which is not desirable.
3176 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3179 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3182 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3184 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3185 authenticated or an ident call has been made. Suppress the default
3186 values for $authenticated_id and $authenticated_sender (but permit -oMai
3187 and -oMas) when testing with -bh.
3189 PH/24 Re-jigged the order of the tests in the default configuration so that the
3190 tests for valid domains and recipients precede the DNS black list and CSA
3191 tests, on the grounds that those ones are more expensive.
3193 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3194 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3195 This bug exists in every version of Exim that I still have, right back to
3198 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3199 However, an attempt to turn on case-sensitivity in a regex key by
3200 including (?-i) didn't work because the subject string was already
3201 lowercased, and the effects were non-intuitive. It turns out that a
3202 one-line patch can be used to allow (?-i) to work as expected.
3208 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3209 other effects) broke the use of negated acl sub-conditions.
3211 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3214 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3215 "Deny" causes Exim to reject the incoming connection with a 554 error.
3216 Unfortunately, if there is a major crisis, such as a disk failure,
3217 tcp-wrappers gives "deny", whereas what one would like would be some
3218 kind of temporary error. A kludge has been added to help with this.
3219 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3220 554 error is used if errno is still zero or contains ENOENT (which occurs
3221 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3224 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3226 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3227 errors. Otherwise a message that provokes a temporary error (when other
3228 messages do not) can cause a whole host to time out.
3230 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3231 addresses were routed directly to files or pipes from a redirect router.
3232 File deliveries just didn't batch; pipe deliveries might have suffered
3235 PH/06 A failure to get a lock for a hints database would erroneously always say
3236 "Failed to get write lock", even when it was really a read lock.
3238 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3239 of 0600. This has been changed to use the value of the lockfile_mode
3240 option (which defaults to 0600).
3242 PH/08 Applied small patch from the Sieve maintainer.
3244 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3245 folder from quota calculations, a direct delivery into this folder messed
3246 up the contents of the maildirsize file. This was because the regex was
3247 used only to exclude .Trash (or whatever) when the size of the mailbox
3248 was calculated. There was no check that a delivery was happening into an
3249 excluded directory. This bug has been fixed by ignoring all quota
3250 processing for deliveries into excluded directories.
3252 PH/10 Added the maildirfolder_create_regex option to appendfile.
3258 PH/01 The code for finding all the local interface addresses on a FreeBSD
3259 system running IPv6 was broken. This may well have applied to all BSD
3260 systems, as well as to others that have similar system calls. The broken
3261 code found IPv4 interfaces correctly, but gave incorrect values for the
3262 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3263 that it would not match correctly against @[] and not recognize the IPv6
3266 PH/02 The ipliteral router was not recognizing addresses of the form user@
3267 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3269 PH/03 Added disable_ipv6.
3271 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3272 decoded form, because it is most often used to construct To: headers
3273 lines in autoreplies, and the decoded form may well be syntactically
3274 invalid. However, $reply_address has leading white space removed, and all
3275 newlines turned into spaces so that the autoreply transport does not
3278 PH/05 If group was specified without a user on a router, and no group or user
3279 was specified on a transport, the group from the router was ignored.
3281 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3282 for visible compile-time settings that can be used to change these
3283 numbers, for those that want even more. Backwards compatibility with old
3284 spool files has been maintained. However, going back to a previous Exim
3285 release will lost any variables that are in spool files.
3287 PH/07 Two small changes when running in the test harness: increase delay when
3288 passing a TCP/IP connection to a new process, in case the original
3289 process has to generate a bounce, and remove special handling of
3290 127.0.0.2 (sic), which is no longer necessary.
3292 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3293 be the same on different OS.
3295 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3298 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3299 whether --show-vars was specified or not
3301 JJ/02 exipick: Added support for new ACL variable spool format introduced
3304 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3305 syntactically invalid From: or Reply-to: line, and a filter used this to
3306 generate an autoreply, and therefore failed to obtain an address for the
3307 autoreply, Exim could try to deliver to a non-existent relative file
3308 name, causing unrelated and misleading errors. What now happens is that
3309 it logs this as a hard delivery error, but does not attempt to create a
3312 PH/11 The exinext utility has a -C option for testing purposes, but although
3313 the given file was scanned by exinext itself; it wasn't being passed on
3314 when Exim was called.
3316 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3317 an end-of-file indication when reading a command response.
3319 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3320 compiled. In many other places in Exim, IPv6 addresses are always
3321 recognized, so I have changed this. It also means that IPv4 domain
3322 literals of the form [IPV4:n.n.n.n] are now always recognized.
3324 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3325 used if the router is not running as root, for example, when verifying at
3326 ACL time, or when using -bh. The debugging output from this situation was
3327 non-existent - all you got was a failure to exec. I have made two
3330 (a) Failures to set uid/gid, the current directory, or a process leader
3331 in a subprocess such as that created by queryprogram now generate
3332 suitable debugging output when -d is set.
3334 (b) The queryprogram router detects when it is not running as root,
3335 outputs suitable debugging information if -d is set, and then runs
3336 the subprocess without attempting to change uid/gid.
3338 PH/15 Minor change to Makefile for building test_host (undocumented testing
3341 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3342 additional section of a DNS packet that returns MX or SRV records.
3343 Instead, it always explicitly searches for A/AAAA records. This avoids
3344 major problems that occur when a DNS server includes only records of one
3345 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3346 fixed another bug: if SRV records were looked up and the corresponding
3347 address records were *not* found in the additional section, the port
3348 values from the SRV records were lost.
3350 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3351 using the correct key (the original address) when searching the retry
3352 rules in order to find which one to use for generating the retry hint.
3354 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3355 adding the default one. Similarly, if it contains a Reply-To: header, the
3356 errors_reply_to option, if set, is not used.
3358 PH/19 When calculating a retry time, Exim used to measure the "time since
3359 failure" by looking at the "first failed" field in the retry record. Now
3360 it does not use this if it is later than than the arrival time of the
3361 message. Instead it uses the arrival time. This makes for better
3362 behaviour in cases where some deliveries succeed, thus re-setting the
3363 "first failed" field. An example is a quota failure for a huge message
3364 when small messages continue to be delivered. Without this change, the
3365 "time since failure" will always be short, possible causing more frequent
3366 delivery attempts for the huge message than are intended.
3367 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3369 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3370 $1, $2, $3) because the numerical variables can be reset during some
3371 expansion items (e.g. "match"), thereby losing the authentication data.
3373 PH/21 Make -bV show the size of off_t variables so that the test suite can
3374 decide whether to run tests for quotas > 2G.
3376 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3377 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3378 filecount value is greater than 2G or if a quota value is greater than 2G
3379 on a system where the size of off_t is not greater than 4, a panic error
3382 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3383 never match. The debug and -bh output now contains an explicit error
3384 message indicating a malformed IPv4 address or mask.
3386 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3387 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3388 PH/23 above applies.
3390 PH/25 Do not write to syslog when running in the test harness. The only
3391 occasion when this arises is a failure to open the main or panic logs
3392 (for which there is an explicit test).
3394 PH/26 Added the /no_tell option to "control=freeze".
3396 PH/27 If a host name lookup failed very early in a connection, for example, if
3397 the IP address matched host_lookup and the reverse lookup yielded a name
3398 that did not have a forward lookup, an error message of the form "no IP
3399 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3400 could be logged. Now it outputs the IP address instead of "NULL".
3402 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3403 allows the sender and the authenticated sender to be set when
3404 submitting a message from within Exim. Since child_open_exim() is
3405 documented for local_scan(), the new function should be too.
3407 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3408 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3409 results in an empty string is now treated as unset.
3411 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3413 PH/31 Added sender_verify_fail logging option.
3415 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3416 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3417 tidied the source and removed it altogether.
3419 PH/33 When a queue run was abandoned because the load average was too high, a
3420 log line was always written; now it is written only if the queue_run log
3421 selector is set. In addition, the log line for abandonment now contains
3422 information about the queue run such as the pid. This is always present
3423 in "start" and "stop" lines but was omitted from the "abandon" line.
3425 PH/34 Omit spaces between a header name and the colon in the error message that
3426 is given when verify = headers_syntax fails (if there are lots of them,
3427 the message gets confusing).
3429 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3430 names, as there are now some PTR records that contain slashes. This check
3431 is only to protect against broken name servers that fall over on strange
3432 characters, so the fact that it applies to all lookups doesn't matter.
3434 PH/36 Now that the new test suite is complete, we can remove some of the
3435 special code in Exim that was needed for the old test suite. For example,
3436 sorting DNS records because real resolvers return them in an arbitrary
3437 order. The new test suite's fake resolver always returns records in the
3440 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3441 bounces) except when queue_only is set, to avoid logging races between
3442 the different processes.
3444 PH/38 Panic-die if .include specifies a non-absolute path.
3446 PH/39 A tweak to the "H" retry rule from its user.
3448 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3449 a label. They prevented compilation on older perls.
3451 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3452 a warning to be raised on newish perls.
3454 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3455 on queue. Changes to match documented behaviour of showing count of
3456 messages matching specified criteria.
3458 PH/40 Changed the default ident timeout from 30s to 5s.
3460 PH/41 Added support for the use of login_cap features, on those BSD systems
3461 that have them, for controlling the resources used by pipe deliveries.
3463 PH/42 The content-scanning code uses fopen() to create files in which to put
3464 message data. Previously it was not paying any attention to the mode of
3465 the files. Exim runs with umask(0) because the rest of the code creates
3466 files with open(), and sets the required mode explicitly. Thus, these
3467 files were ending up world-writeable. This was not a big issue, because,
3468 being within the spool directory, they were not world-accessible. I have
3469 created a function called modefopen, which takes an additional mode
3470 argument. It sets umask(777), creates the file, chmods it to the required
3471 mode, then resets the umask. All the relevant calls to fopen() in the
3472 content scanning code have been changed to use this function.
3474 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3475 to 24 hours. This avoids potential overflow problems when processing G
3476 and H retry rules. I suspect nobody ever tinkers with this value.
3478 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3480 PH/45 When the plaintext authenticator is running as a client, the server's
3481 challenges are checked to ensure they are valid base64 strings. By
3482 default, the authentication attempt is cancelled if an invalid string is
3483 received. Setting client_ignore_invalid_base64 true ignores these errors.
3484 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3485 they are received. Thus, the responses can be made to depend on the
3486 challenges. If an invalid string is ignored, an empty string is placed in
3489 PH/46 Messages that are created by the autoreply transport now contains a
3490 References: header, in accordance with RFCs 2822 and 3834.
3492 PH/47 Added authenticated_sender_force to the smtp transport.
3494 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3496 PH/49 Installed latest patch from the Sieve maintainer.
3498 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3499 was also set, the appendfile transport was unnecessarily scanning a
3500 directory of message files (e.g. for maildir delivery) to find the count
3501 of files (along with the size), even though it did not need this
3502 information. It now does the scan only if it needs to find either the
3503 size of the count of files.
3505 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3507 PH/52 Two bugs concerned with error handling when the smtp transport is
3510 (i) Exim was not creating retry information for temporary errors given
3511 for individual recipients after the DATA command when the smtp transport
3512 was used in LMTP mode. This meant that they could be retried too
3513 frequently, and not timed out correctly.
3515 (ii) Exim was setting the flag that allows error details to be returned
3516 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3517 recipients that were returned after the DATA command.
3519 PH/53 This is related to PH/52, but is more general: for any failing address,
3520 when detailed error information was permitted to be returned to the
3521 sender, but the error was temporary, then after the final timeout, only
3522 "retry timeout exceeded" was returned. Now it returns the full error as
3523 well as "retry timeout exceeded".
3525 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3526 do this, and (what is worse) MTAs that accept it.
3528 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3529 will now be deprecated.
3531 PH/56 New os.c-cygwin from the Cygwin maintainer.
3533 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3534 formats (previously only available in exim formats via -bpr, -bpru,
3535 and -bpra. Now also available in native and exiqgrep formats)
3537 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3538 with very large, slow to parse queues
3540 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3542 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3544 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3545 responses to authentication challenges, though it was showing the
3546 challenges; (ii) I've removed the CR characters from the debug output for
3549 PH/58 Allow for the insertion of a newline as well as a space when a string
3550 is turned into more than one encoded-word during RFC 2047 encoding. The
3551 Sieve code now uses this.
3553 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3554 data_4xx, lost_connection, tls_required.
3556 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3557 message was being sent as an SMTP response.
3559 PH/61 Add -l and -k options to exicyclog.
3561 PH/62 When verifying, if an address was redirected to one new address, so that
3562 verification continued, and the new address failed or deferred after
3563 having set something in $address_data, the value of $address_data was not
3564 passed back to the ACL. This was different to the case when no
3565 redirection occurred. The value is now passed back in both cases.
3567 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3568 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3569 use login_cap.h, so on its own it isn't the distinguishing feature. The
3570 new name refers directly to the setclassresources() function.
3572 PH/65 Added configuration files for NetBSD3.
3574 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3576 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3577 is preferred over IPv4.
3579 PH/68 The bounce_return_message and bounce_return_body options were not being
3580 honoured for bounces generated during the reception of non-SMTP messages.
3581 In particular, this applied to messages rejected by the ACL. This bug has
3582 been fixed. However, if bounce_return_message is true and bounce_return_
3583 body is false, the headers that are returned for a non-SMTP message
3584 include only those that have been read before the error was detected.
3585 (In the case of an ACL rejection, they have all been read.)
3587 PH/69 The HTML version of the specification is now built in a directory called
3588 spec_html instead of spec.html, because the latter looks like a path with
3589 a MIME-type, and this confuses some software.
3591 PH/70 Catch two compiler warnings in sieve.c.
3593 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3594 function verify_get_ident() calls ip_connect() to connect a socket, but
3595 if the "connect()" function timed out, ip_connect() used to close the
3596 socket. However, verify_get_ident() also closes the socket later, and in
3597 between Exim writes to the log, which may get opened at this point. When
3598 the socket was closed in ip_connect(), the log could get the same file
3599 descriptor number as the socket. This naturally causes chaos. The fix is
3600 not to close the socket in ip_connect(); the socket should be closed by
3601 the function that creates it. There was only one place in the code where
3602 this was missing, in the iplookup router, which I don't think anybody now
3603 uses, but I've fixed it anyway.
3605 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3606 well as to direct DNS lookups. Otherwise the handling of names in host
3607 lists is inconsistent and therefore confusing.
3613 PH/01 Two changes to the default runtime configuration:
3615 (1) Move the checks for relay_from_hosts and authenticated clients from
3616 after to before the (commented out) DNS black list checks.
3618 (2) Add control=submission to the relay_from_hosts and authenticated
3619 clients checks, on the grounds that messages accepted by these
3620 statements are most likely to be submissions.
3622 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3624 (1) Generate an error if the third argument for the ${prvs expansion is
3627 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3630 (3) Reset the variables that are obtained from the first argument of
3631 ${prvscheck and used in the second argument before leaving the code,
3632 because their memory is reclaimed, so using them afterwards may do
3635 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3636 one (it's much easier than Tom thought :-).
3638 (5) Because of (4), we can now allow for the use of $prvscheck_result
3639 inside the third argument.
3641 PH/03 For some reason, the default setting of PATH when running a command from
3642 a pipe transport was just "/usr/bin". I have changed it to
3645 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3646 anything to be listed in the output from -bV.
3648 PH/05 When a filter generated an autoreply, the entire To: header line was
3649 quoted in the delivery log line, like this:
3651 => >A.N.Other <ano@some.domain> <original@ddress> ...
3653 This has been changed so that it extracts the operative address. There
3654 may be more than one such address. If so, they are comma-separated, like
3657 => >ano@some.domain,ona@other.domain <original@ddress> ...
3659 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3660 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3661 not being looked up in the rDNS to get a host name, Exim was showing the
3662 IP address twice in Received: lines, even though the IP addresses were
3663 identical. For example:
3665 Received: from [1.2.3.4] (helo=[1.2.3.4])
3667 However, if the real host name was known, it was omitting the HELO data
3668 if it matched the actual IP address. This has been tidied up so that it
3669 doesn't show the same IP address twice.
3671 PH/07 When both +timestamp and +memory debugging was on, the value given by
3672 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3673 called by the debug printing, thereby overwriting the timestamp buffer.
3674 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3676 PH/08 When the original message was included in an autoreply transport, it
3677 always said "this is a copy of the message, including all the headers",
3678 even if body_only or headers_only was set. It now gives an appropriate
3681 PH/09 Applied a patch from the Sieve maintainer which:
3683 o fixes some comments
3684 o adds the (disabled) notify extension core
3685 o adds some debug output for the result of if/elsif tests
3686 o points to the current vacation draft in the documentation
3687 and documents the missing references header update
3691 o fixes a bug in processing the envelope test (when testing
3692 multiple envelope elements, the last element determined the
3695 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3696 Electronic Mail") by including:
3698 Auto-submitted: auto-generated
3700 in the messages that it generates (bounce messages and others, such as
3701 warnings). In the case of bounce messages for non-SMTP messages, there was
3702 also a typo: it was using "Auto_submitted" (underscore instead of
3703 hyphen). Since every message generated by Exim is necessarily in response
3704 to another message, thes have all been changed to:
3706 Auto-Submitted: auto-replied
3708 in accordance with these statements in the RFC:
3710 The auto-replied keyword:
3712 - SHOULD be used on messages sent in direct response to another
3713 message by an automatic process,
3715 - MUST NOT be used on manually-generated messages,
3717 - MAY be used on Delivery Status Notifications (DSNs) and Message
3718 Disposition Notifications (MDNs),
3720 - MUST NOT be used on messages generated by automatic or periodic
3721 processes, except for messages which are automatic responses to
3724 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3725 to the default Received: header definition.
3727 PH/12 Added log selector acl_warn_skipped (default on).
3729 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3730 variables because (a) they are in the wrong storage pool and (b) even if
3731 they were copied, it wouldn't work properly because of the caching.
3733 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3734 checking when decoding. Apparently there are clients that generate
3735 overlong encoded strings. Why am I not surprised?
3737 PH/15 If the first argument of "${if match_address" was not empty, but did not
3738 contain an "@" character, Exim crashed. Now it writes a panic log message
3739 and treats the condition as false.
3741 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3743 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3744 extension "envelope-auth". The code is finished and in agreement with
3745 other implementations, but there is no documentation so far and in fact,
3746 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3747 not changing the active code.
3749 Print executed "if" and "elsif" statements when debugging is used. This
3750 helps a great deal to understand what a filter does.
3752 Document more things not specified clearly in RFC3028. I had all this
3753 sorted out, when out of a sudden new issues came to my mind. Oops."
3755 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3758 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3759 canonical form (as documented). However, after a host name lookup from
3760 the IP address, check_host() was doing a simple string comparison with
3761 addresses acquired from the DNS when checking that the found name did
3762 have the original IP as one of its addresses. Since any found IPv6
3763 addresses are likely to be in abbreviated form, the comparison could
3764 fail. Luckily, there already exists a function for doing the comparison
3765 by converting both addresses to binary, so now that is used instead of
3766 the text comparison.
3768 PH/20 There was another similar case to PH/19, when a complete host name was
3769 given in a host list; looking up its IP address could give an abbreviated
3770 form, whereas the current host's name might or might not be abbreviated.
3771 The same fix has been applied.
3777 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3778 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3781 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3782 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3784 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3786 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3787 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3788 Also, for "Auto-Submitted", treat anything other than "no" as
3789 non-personal, in accordance with RFC 3834. (Previously it treated
3790 anything starting "auto-" as non-personal.)
3792 TF/01 The control=submission/name=... option had a problem with syntax
3793 errors if the name included a slash character. The /name= option
3794 now slurps the rest of the string, so it can include any characters
3795 but it must come last in the list of options (after /sender_retain
3798 PH/05 Some modifications to the interface to the fake nameserver for the new
3806 TK/01 Added the "success_on_redirect" address verification option. See
3807 NewStuff for rationale and an example.
3809 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3811 PH/02 Patch to exigrep to allow it to work on syslog lines.
3813 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3814 fread() to skip over the body file's header line, because in Cygwin the
3815 header line is locked and is inaccessible.
3817 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3818 co-exist for some time) to make it clear that it is the Exim ID that is
3819 referenced, not the Message-ID: header line.
3821 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3822 string_format() function, because snprintf() does not exist on all
3825 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3826 ${stat: expansion item.
3828 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3829 protocol synchronization error", to keep the pedants happy.
3831 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3832 well as for IRIX systems, when gcc is being used. See the host.c source
3835 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3837 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3840 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3841 transport and to the smtp transport in LMTP mode.
3843 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3845 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3846 run a filter in a subprocess. This could lead to confusion in subsequent
3847 lookups in the parent process. There should also be a search_tidyup() at
3848 the end of the subprocess.
3850 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3851 only if the host matched helo_try_verify_hosts, which caused the
3852 verification to occur when the EHLO/HELO command was issued. The ACL just
3853 tested the remembered result. Now, if a previous verification attempt has
3854 not happened, "verify = helo" does it there and then.
3856 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3858 TK/03 Fix log output including CR from clamd.
3860 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3861 exist provoked a memory error which could cause a segfault.
3863 PH/15 Installed PCRE 6.2
3865 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3867 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3868 of the problem. Specifically, suggested +O2 rather than +O1 for the
3871 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3873 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3874 record, the port was not correctly logged when the outgoing_port log
3875 selector was set (it logged the transort's default port).
3877 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3878 fallback_hosts, and "hosts" in the smtp transport.
3880 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3881 host errors such as "Connection refused".
3883 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3884 authentication with radiusclient 0.4.9:
3886 - Error returned from rc_read_config was caught wrongly
3887 - Username/password not passed on to radius server due to wrong length.
3889 The presumption is that some radiusclient API changes for 4.51/PH/17
3890 were not taken care of correctly. The code is still untested by me (my
3891 Linux distribution still has 0.3.2 of radiusclient), but it was
3892 contributed by a Radius user.
3894 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3895 expanding the "port" option of the smtp transport.
3897 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3898 while reading a MIME header. Thanks to Tom Hughes for a patch.
3900 PH/24 Include config.h inside local_scan.h so that configuration settings are
3903 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3904 that in an ACL for RCPT (for example), you can examine exactly what was
3907 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3908 commands, but it was not correctly comparing the address with the actual
3909 client host address. Thus, it would show the EHLO address in Received:
3910 header lines when this was not necessary.
3912 PH/27 Added the % operator to ${eval:}.
3914 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3915 it should be ignoring failures (because with -C, for example, it has lost
3916 privilege). It wasn't ignoring creation failures other than "already
3919 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3922 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3923 because an input line was too long, either on its own, or by virtue of
3924 too many continuations, the temporary file was not being removed, and the
3925 return code was incorrect.
3927 PH/31 Missing "BOOL" in function definition in filtertest.c.
3929 PH/32 Applied Sieve patches from the maintainer.
3931 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3933 PH/33 Added "verify = not_blind".
3935 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3936 Local/Makefile (with some defaults set). These are used in built scripts
3937 such as exicyclog, but they have never been used in the exim_install
3938 script (though there are many overriding facilities there). I have
3939 arranged that the exim_install script now takes note of these two
3942 PH/35 Installed configuration files for Dragonfly.
3944 PH/36 When a locally submitted message by a trusted user did not contain a
3945 From: header, and the sender address was obtained from -f or from an SMTP
3946 MAIL command, and the trusted user did not use -F to supply a sender
3947 name, $originator_name was incorrectly used when constructing a From:
3948 header. Furthermore, $originator_name was used for submission mode
3949 messages from external hosts without From: headers in a similar way,
3950 which is clearly wrong.
3952 PH/37 Added control=suppress_local_fixups.
3954 PH/38 When log_selector = +received_sender was set, and the addition of the
3955 sender made the log line's construction buffer exactly full, or one byte
3956 less than full, an overflow happened when the terminating "\n" was
3959 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3960 when the result of a list match is failure because a DNS lookup failed.
3962 PH/40 RM_COMMAND is now used in the building process.
3964 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3965 the "build-* directories that it finds.
3967 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3968 address in a domain literal was a prefix of an interface address.
3970 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3971 when verifying a sender address, unless rewrite_headers is false.
3973 PH/44 Wrote a long comment about why errors_to addresses are verified as
3974 recipients, not senders.
3976 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3977 the ratelimit ACL was added.
3979 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3981 PH/46 Added extra information about PostgreSQL errors to the error string.
3983 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3984 avoiding the need to install special zones in a real server. This is
3985 backwards compatible; if it can't find the fake resolver, it drops back.
3986 Thus, both old and new test suites can be run.
3988 TF/02 Added util/ratelimit.pl
3990 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3991 clock is set back in time.
3993 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3994 Candler <B.Candler@pobox.com>.
3996 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3997 OK for addresses that are the result of redirections.
3999 PH/48 A number of further additions for the benefit of the new test suite,
4000 including a fake gethostbyname() that interfaces to the fake DNS resolver
4003 TF/06 The fix for widen_domains has also been applied to qualify_single and
4004 search_parents which are the other dnslookup options that can cause
4007 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
4010 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
4012 TF/07 Exim produced the error message "an SRV record indicated no SMTP
4013 service" if it encountered an MX record with an empty target hostname.
4014 The message is now "an MX or SRV record indicated no SMTP service".
4016 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
4017 if the DNS of the sending site is misconfigured. This is quite a
4018 common situation. This change restores the behaviour of treating a
4019 helo verification defer as a failure.
4021 PH/51 If self=fail was set on a router, the bounce message did not include the
4022 actual error message.
4028 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
4030 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
4031 command itself ended in error, the underlying message about the transport
4032 filter timeout was being overwritten with the pipe command error. Now the
4033 underlying error message should be appended to the second error message.
4035 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4037 PH/02 Reduce the amount of output that "make" produces by default. Full output
4038 can still be requested.
4040 PH/03 The warning log line about a condition test deferring for a "warn" verb
4041 was being output only once per connection, rather than after each
4042 occurrence (because it was using the same function as for successful
4043 "warn" verbs). This seems wrong, so I have changed it.
4045 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4046 it should not have, which might have caused a crash in the right
4047 circumstances, but probably never did.
4049 PH/04 Installed a modified version of Tony Finch's patch to make submission
4050 mode fix the return path as well as the Sender: header line, and to
4051 add a /name= option so that you can make the user's friendly name appear
4054 TF/03 Added the control = fakedefer ACL modifier.
4056 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4057 Mark Lowes for thorough testing.
4059 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4061 TK/03 Merged latest SRS patch from Miles Wilton.
4063 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4064 with the definition in sysexits.h (which is #included earlier).
4065 Fortunately, Exim does not actually use EX_OK. The code used to try to
4066 preserve the sysexits.h value, by assuming that macro definitions were
4067 scanned for macro replacements. I have been disabused of this notion,
4068 so now the code just undefines EX_OK before #including unistd.h.
4070 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4071 in the smtp transport. When a block could not be written in a single
4072 write() function, the timeout was being re-applied to each part-write.
4073 This seems wrong - if the receiver was accepting one byte at a time it
4074 would take for ever. The timeout is now adjusted when this happens. It
4075 doesn't have to be particularly precise.
4077 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4078 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4080 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4081 without a domain if the check to <postmaster@domain> fails.
4083 SC/01 Eximstats: added -xls and the ability to specify output files
4084 (patch written by Frank Heydlauf).
4086 SC/02 Eximstats: use FileHandles for outputting results.
4088 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4090 SC/04 Eximstats: fixed display of large numbers with -nvr option
4092 SC/05 Eximstats: fixed merging of reports with empty tables.
4094 SC/06 Eximstats: added the -include_original_destination flag
4096 SC/07 Eximstats: removed tabs and trailing whitespace.
4098 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4100 TK/06 MBOX spool code: Add real "From " MBOX separator line
4101 so the .eml file is really in mbox format (even though
4102 most programs do not really care). Patch from Alex Miller.
4104 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4105 The latter is generated from $received_to and is only set if the
4106 message has one envelope recipient. SA can use these headers,
4107 obviously out-of-the-box. Patch from Alex Miller.
4109 PH/08 The ${def test on a variable was returning false if the variable's
4110 value was "0", contrary to what the specification has always said!
4111 The result should be true unless the variable is empty.
4113 PH/09 The syntax error of a character other than { following "${if
4114 def:variable_name" (after optional whitespace) was not being diagnosed.
4115 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4116 accidental colon was present, for example, could give incorrect results.
4118 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4119 result is used (not including appendfile, where other changes are about
4122 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4123 This involved changing a lot of size variables from int to off_t. It
4124 should work with maildirs and everything.
4126 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4127 spamd dying while we are connected to it.
4129 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4132 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4133 function for BDB 4.3.
4135 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4137 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4138 the sources for PCRE have been re-arranged and more files are now
4141 PH/15 The code I had for printing potentially long long variables in PH/11
4142 above was not the best (it lost precision). The length of off_t variables
4143 is now inspected at build time, and an appropriate printing format (%ld
4144 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4145 to be "long long int" or "long int". This is needed for the internal
4146 formatting function string_vformat().
4148 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4149 the configuration file to be ":syslog", then the script "guesses" where
4150 the logs files are, rather than using the compiled in default. In our
4151 case the guess is not the same as the compiled default, so the script
4152 suddenly stopped working when I started to use syslog. The patch checks
4153 to see if log_file_path is "". If so, it attempts to read it from exim
4154 with no configuration file to get the compiled in version, before it
4155 falls back to the previous guessing code."
4157 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4158 implementing BATV in an Exim configuration. See NewStuff for the gory
4161 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4162 Makefile that are specific to HP-UX.
4164 PH/18 If the "use_postmaster" option was set for a recipient callout together
4165 with the "random" option, the postmaster address was used as the MAIL
4166 FROM address for the random test, but not for the subsequent recipient
4167 test. It is now used for both.
4169 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4170 patch removes a few documentation additions to RFC 3028, because the
4171 latest draft now contains them. It adds the new en;ascii-case comparator
4172 and a new error check for 8bit text in MIME parts. Comparator and
4173 require names are now matched exactly. I enabled the subaddress
4174 extension, but it is not well tested yet (read: it works for me)."
4176 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4177 rework some of the code of TK/09 above to avoid the hardwired use of
4178 "%lld" and "long long". Replaced the call to snprintf() with a call to
4181 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4182 records point to non-existent hosts", "retry timeout exceeded", and
4183 "retry time not reached for any host after a long failure period".
4185 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4186 experimental DomainKeys support:
4188 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4189 (2) On an error such as an illegally used "control", the wrong name for
4190 the control was given.
4192 These problems did NOT occur unless DomainKeys support was compiled.
4194 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4196 PH/24 Added ${if match_ip condition.
4198 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4199 databases so that it will be absolutely obvious if a crash occurs in the
4200 DB library. This is a regular occurrence (often caused by mis-matched
4203 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4204 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4205 picked up on a user's system that detects such things. There doesn't seem
4206 to be a gcc warning option for this - only an attribute that has to be
4207 put on the function's prototype. It seems that in Fedora Core 4 they have
4208 set this on a number of new functions. No doubt there will be more in due
4211 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4212 specify a transport. However, if an address that was verified by such a
4213 router was the subject of a callout, Exim crashed because it tried to
4214 read the rcpt_include_affixes from the non-existent transport. Now it
4215 just assumes that the setting of that option is false. This bug was
4216 introduced by 4.51/PH/31.
4218 PH/28 Changed -d+all to exclude +memory, because that information is very
4219 rarely of interest, but it makes the output a lot bigger. People tend to
4220 do -d+all out of habit.
4222 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4223 code in os-type was giving problems when libc.so lives in lib64, like on
4226 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4227 aren't the modern standard, and it seems that some systems' include files
4228 don't always have them. Exim was already checking for some of the newer
4229 ones like T_AAAA, and defining it itself. I've added checks for all the
4230 record types that Exim uses.
4232 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4233 not automatically generating a new one, as it is supposed to. This
4234 prevented TLS from working. If the file did exist, but contained invalid
4235 data, a new version was generated, as expected. It was only the case of a
4236 non-existent file that was broken.
4238 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4239 with a change in libdomainkeys > 0.64.
4241 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4242 from DNS. If the selector record carries the flag, it now has
4243 precedence over the domain-wide flag.
4245 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4247 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4248 the use of a port name that isn't defined in /etc/services) occurred, the
4249 message was deferred as in a normal delivery, and thus remained on the
4250 spool, instead of being failed because of the mua_wrapper setting. This
4251 is now fixed, and I tidied up some of the mua_wrapper messages at the
4254 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4255 the messages in a hash of arrays rather than using individual hashes.
4256 This is a bit cleaner and results in dramatic memory savings, albeit
4257 at a slight CPU cost.
4259 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4260 as requested by Marc Sherman.
4262 SC/10 Eximstats: added histograms for user specified patterns as requested
4265 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4267 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4268 fopen() in the content-scanning modules that did not already have it.
4274 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4275 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4277 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4279 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4281 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4282 to test Sieve filters that use "vacation".
4284 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4285 that changes the way the GnuTLS parameters are stored in the cache file.
4286 The new format can be generated externally. For backward compatibility,
4287 if the data in the cache doesn't make sense, Exim assumes it has read an
4288 old-format file, and it generates new data and writes a new file. This
4289 means that you can't go back to an older release without removing the
4292 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4293 work if there are any delivery delays because "one_time" forces the
4294 parent to be marked "delivered", so its unseen clone is never tried
4295 again. For this reason, Exim now forbids the simultaneous setting of
4298 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4299 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4300 entry if you want to know the details. The fix, however, neglected to
4301 consider the case when local delivery batching is involved. The test for
4302 "previously delivered" was not happening when checking to see if an
4303 address could be batched with a previous (undelivered) one; under
4304 certain circumstances this could lead to multiple deliveries to the same
4307 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4308 in its include files, and this causes problems building Exim.
4310 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4311 header_syntax) but Exim was just ignoring anything given after a slash.
4312 In particular, this caused confusion with an attempt to use "verify =
4313 reverse_host_lookup/defer_ok". An error is now given when options are
4314 supplied for verify items that do not have them. (Maybe reverse_host_
4315 lookup should have a defer_ok option, but that's a different point.)
4317 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4318 defined by RFC 821) to 2048, because there were problems with some AUTH
4319 commands, and RFC 1869 says the size should be increased for extended
4320 SMTP commands that take arguments.
4322 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4325 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4326 "unknown" error; now it says that the functionality isn't in the binary.
4328 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4329 an address' error message when a string expansion fails (syntax or
4330 whatever). Otherwise the password may appear in the log. Following change
4331 PH/42 below, there is no longer a chance of it appearing in a bounce
4334 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4336 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4337 of its addresses was ever tried. (Bugzilla bug #2.)
4339 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4340 the result incorrectly in the debug output. (It correctly added a newline
4341 to what was transported.)
4343 TF/01 Added $received_time.
4345 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4346 commented out examples of how to interface to a virus scanner and to
4347 SpamAssassin. Also added commented examples of av_scanner and
4348 spamd_address settings.
4350 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4351 and controls are allowed in which ACLs. There were a couple of minor
4352 errors. Some of the entries in the conditions table (which is a table of
4353 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4354 negation of where the condition IS allowed.
4356 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4358 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4359 header file does not have a version number, so I've had to invent a new
4360 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4361 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4362 radiusclient), but it was contributed by a Radius user.
4364 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4365 files or directories, for OpenSSL.
4367 PH/19 When an Exim process that is running as root has to create an Exim log
4368 file, it does so in a subprocess that runs as exim:exim so as to get the
4369 ownership right at creation (otherwise, other Exim processes might see
4370 the file with the wrong ownership). There was no test for failure of this
4371 fork() call, which would lead to the process getting stuck as it waited
4372 for a non-existent subprocess. Forks do occasionally fail when resources
4373 run out. I reviewed all the other calls to fork(); they all seem to check
4376 PH/20 When checking for unexpected SMTP input at connect time (before writing
4377 the banner), Exim was not dealing correctly with a non-positive return
4378 from the read() function. If the client had disconnected by this time,
4379 the result was a log entry for a synchronization error with an empty
4380 string after "input=" when read() returned zero. If read() returned -1
4381 (an event I could not check), uninitialized data bytes were printed.
4382 There were reports of junk text (parts of files, etc) appearing after
4385 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4387 PH/22 Added support for macro redefinition, and (re)definition in between
4388 driver and ACL definitions.
4390 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4391 forgetting to use the resulting value; it was using the unexpanded value.
4393 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4394 hadn't been configured. The fix is from Juergen Kreileder, who
4395 understands it better than I do:
4397 "Here's what I see happening with three configured cyrus_sasl
4398 authenticators configured (plain, login, cram-md5):
4400 On startup auth_cyrus_sasl_init() gets called for each of these.
4401 This means three calls to sasl_listmech() without a specified mech_list.
4402 => SASL tests which mechs of all available mechs actually work
4403 => three warnings about OTP not working
4404 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4406 With the patch, sasl_listmech() also gets called three times. But now
4407 SASL's mech_list option is set to the server_mech specified in the the
4408 authenticator. Or in other words, the answer from sasl_listmech()
4409 gets limited to just the mech you're testing for (which is different
4411 => the return list contains just 'plain' or 'login', 'cram-md5' or
4412 nothing depending on the value of ob->server_mech.
4414 I've just tested the patch: Authentication still works fine,
4415 unavailable mechs specified in the exim configuration are still
4416 caught, and the auth.log warnings about OTP are gone."
4418 PH/25 When debugging is enabled, the contents of the command line are added
4419 to the debugging output, even when log_selector=+arguments is not
4422 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4423 answer is "GNU", and only if the return is "GNU/something" is the answer
4426 PH/27 $acl_verify_message is now set immediately after the failure of a
4427 verification in an ACL, and so is available in subsequent modifiers. In
4428 particular, the message can be preserved by coding like this:
4430 warn !verify = sender
4431 set acl_m0 = $acl_verify_message
4433 Previously, $acl_verify_message was set only while expanding "message"
4434 and "log_message" when a very denied access.
4436 PH/28 Modified OS/os.c-Linux with
4438 -#ifndef OS_LOAD_AVERAGE
4439 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4441 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4442 nomenclature these days.)
4444 PH/29 Installed patch from the Sieve maintainer that adds the options
4445 sieve_useraddress and sieve_subaddress to the redirect router.
4447 PH/30 In these circumstances:
4448 . Two addresses routed to the same list of hosts;
4449 . First host does not offer TLS;
4450 . First host accepts first address;
4451 . First host gives temporary error to second address;
4452 . Second host offers TLS and a TLS session is established;
4453 . Second host accepts second address.
4454 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4455 and peerdn, if requested) that were in fact used only for the second
4458 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4459 attention to any local part prefix or suffix that was matched by the
4460 router that accepted the address. It now behaves in the same way as it
4461 does for delivery: the affixes are removed from the local part unless
4462 rcpt_include_affixes is set on the transport.
4464 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4465 timeout during the DATA phase of an incoming message.
4467 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4468 applied a patch sanctioned by the Sieve maintainer.
4470 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4471 the uid or gid is negative. A case of a negative gid caused this to be
4472 noticed. The fix allows for either to be negative.
4474 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4475 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4478 PH/36 The change PH/12 above was broken. Fixed it.
4480 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4481 the grounds that routing the same address twice would always produce the
4482 same answer. This might have been true once, but it is certainly no
4483 longer true now. Routing a child address may depend on the previous
4484 routing that produced that child. Some complicated redirection strategies
4485 went wrong when messages had multiple recipients, and made Exim's
4486 behaviour dependent on the order in which the addresses were given.
4488 I have moved the duplicate checking until after the routing is complete.
4489 Exim scans the addresses that are assigned to local and remote
4490 transports, and removes any duplicates. This means that more work will be
4491 done, as duplicates will always all be routed, but duplicates are
4492 presumably rare, so I don't expect this is of any significance.
4494 For deliveries to pipes, files, and autoreplies, the duplicate checking
4495 still happens during the routing process, since they are not going to be
4498 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4499 It corrects a timeout issue with spamd. This is Ian's comment: "The
4500 background is that sometimes spamd either never reads data from a
4501 connection it has accepted, or it never writes response data. The exiscan
4502 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4503 blindly assumes that writes won't block so it may never time out."
4505 PH/39 Allow G after quota size as well as K and M.
4507 PH/40 The value set for $authenticated_id in an authenticator may not contain
4508 binary zeroes or newlines because the value is written to log lines and
4509 to spool files. There was no check on this. Now the value is run through
4510 the string_printing() function so that such characters are converted to
4511 printable escape sequences.
4513 PH/41 $message_linecount is a new variable that contains the total number of
4514 lines in the message. Compare $body_linecount, which is the count for the
4517 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4518 bounce and delay warning messages, except in certain special cases, which
4521 (a) An SMTP error message from a remote host;
4522 (b) A message specified in a :fail: redirection;
4523 (c) A message specified in a "fail" command in a system filter;
4524 (d) A message specified in a FAIL return from the queryprogram router;
4525 (e) A message specified by the cannot_route_message router option.
4527 In these cases only, Exim does include the error details in bounce and
4528 warning messages. There are also a few cases where bland messages such
4529 as "unrouteable address" or "local delivery error" are given.
4531 PH/43 $value is now also set for the "else" part of a ${run expansion.
4533 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4534 being worked on, but at least Exim now implements the latest version to
4537 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4538 process to complete was treated as a delivery failure, a timeout while
4539 writing the message to the pipe was logged, but erroneously treated as a
4540 successful delivery. Such timeouts include transport filter timeouts. For
4541 consistency with the overall process timeout, these timeouts are now
4542 treated as errors, giving rise to delivery failures by default. However,
4543 there is now a new Boolean option for the pipe transport called
4544 timeout_defer, which, if set TRUE, converts the failures into defers for
4545 both kinds of timeout. A transport filter timeout is now identified in
4548 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4549 systems where "make" and "gmake" are different, calling "gmake" at top
4550 level broke things. I've arranged for the value of $(MAKE) to be passed
4551 from the Makefile to this script so that it can call the same version of
4555 A note about Exim versions 4.44 and 4.50
4556 ----------------------------------------
4558 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4559 changes of various kinds. As a consequence, a big documentation update was
4560 needed. This delayed the release for rather longer than seemed good, especially
4561 in the light of a couple of (minor) security issues. Therefore, the changes
4562 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4563 release. So 4.44 and 4.50 are in effect two different branches that both start
4566 I have left the 4.50 change log unchanged; it contains all the changes since
4567 4.43. The change log for 4.44 is below; many of its items are identical to
4568 those for 4.50. This seems to be the most sensible way to preserve the
4569 historical information.
4575 1. Minor wording change to the doc/README.SIEVE file.
4577 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4578 computation of the current number of files was incorrect.
4580 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4581 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4584 4. Give more explanation in the error message when the command for a transport
4585 filter fails to execute.
4587 5. There are several places where Exim runs a non-Exim command in a
4588 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4589 was being done only for the command run by the queryprogram router. It is
4590 now done for all such subprocesses. The other cases are: ${run, transport
4591 filters, and the commands run by the lmtp and pipe transports.
4593 6. Added CONFIGURE_GROUP build-time option.
4595 7. Some older OS have a limit of 256 on the maximum number of file
4596 descriptors. Exim was using setrlimit() to set 1000 as a large value
4597 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4598 systems. I've change it so that if it can't get 1000, it tries for 256.
4600 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4601 was an oversight, and furthermore, ever since the addition of extra
4602 controls (e.g. 4.43/32), the checks on when to allow different forms of
4603 "control" were broken. There should now be diagnostics for all cases when a
4604 control that does not make sense is encountered.
4606 9. Added the /retain_sender option to "control=submission".
4608 10. $recipients is now available in the predata ACL (oversight).
4610 11. Tidy the search cache before the fork to do a delivery from a message
4611 received from the command line. Otherwise the child will trigger a lookup
4612 failure and thereby defer the delivery if it tries to use (for example) a
4613 cached ldap connection that the parent has called unbind on.
4615 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4616 of $address_data from the recipient verification was clobbered by the
4617 sender verification.
4619 13. The value of address_data from a sender verification is now available in
4620 $sender_address_data in subsequent conditions in the ACL statement.
4622 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4624 15. Added a new option "connect=<time>" to callout options, to set a different
4627 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4628 was its contents. (It was OK if the option was not defined at all.)
4630 17. A "Completed" log line is now written for messages that are removed from
4631 the spool by the -Mrm option.
4633 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4634 information about exactly what failed.
4636 19. Added -dd to debug only the daemon process.
4638 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4639 handles timeouts, both on the server side and network timeouts. Renamed the
4640 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4642 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4643 It is now set to "smtps".
4645 22. $host_address is now set to the target address during the checking of
4646 ignore_target_hosts.
4648 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4649 being passed; this would have caused $sender_host_name to have been used if
4650 matching the list had actually called for a host name (not very likely,
4651 since this list is usually IP addresses). A host name is now passed as
4654 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4655 code that specifies a non-restarting handler (typically sigaction() in
4656 modern systems) in an attempt to fix a rare and obscure crash bug.
4658 25. Narrowed the window for a race in the daemon that could cause it to ignore
4659 SIGCHLD signals. This is not a major problem, because they are used only to
4660 wake it up if nothing else does.
4662 26. A malformed maildirsize file could cause Exim to calculate negative values
4663 for the mailbox size or file count. Odd effects could occur as a result.
4664 The maildirsize information is now recalculated if the size or filecount
4667 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4668 support for a long time. Removed HAVE_SYS_VFS_H.
4670 28. Installed the latest version of exipick from John Jetmore.
4672 29. In an address list, if the pattern was not a regular expression, an empty
4673 subject address (from a bounce message) matched only if the pattern was an
4674 empty string. Non-empty patterns were not even tested. This was the wrong
4675 because it is perfectly reasonable to use an empty address as part of a
4676 database query. An empty address is now tested by patterns that are
4677 lookups. However, all the other forms of pattern expect the subject to
4678 contain a local part and a domain, and therefore, for them, an empty
4679 address still always fails if the pattern is not itself empty.
4681 30. Exim went into a mad DNS loop when attempting to do a callout where the
4682 host was specified on an smtp transport, and looking it up yielded more
4683 than one IP address.
4685 31. Re-factored the code for checking spool and log partition space into a
4686 function that finds that data and another that does the check. The former
4687 is then used to implement four new variables: $spool_space, $log_space,
4688 $spool_inodes, and $log_inodes.
4690 32. The RFC2047 encoding function was originally intended for short strings
4691 such as real names; it was not keeping to the 75-character limit for
4692 encoded words that the RFC imposes. It now respects the limit, and
4693 generates multiple encoded words if necessary. To be on the safe side, I
4694 have increased the buffer size for the ${rfc2047: expansion operator from
4697 33. It is now permitted to omit both strings after an "if" condition; if the
4698 condition is true, the result is "true". As before, when the second string
4699 is omitted, a false condition yields an empty string. This makes it less
4700 cumbersome to write custom ACL and router conditions.
4702 34. Failure to deliver a bounce message always caused it to be frozen, even if
4703 there was an errors_to setting on the router. The errors_to setting is now
4706 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4707 canonical form (fully expanded) before being placed in
4708 $sender_host_address.
4710 36. The table in the code that translates DNS record types into text (T_A to
4711 "A" for instance) was missing entries for NS and CNAME. It is just possible
4712 that this could have caused confusion if both these types were looked up
4713 for the same domain, because the text type is used as part of Exim's
4714 per-process caching. But the chance of anyone hitting this buglet seems
4717 37. The dnsdb lookup has been extended in a number of ways.
4719 (1) There is a new type, "zns", which walks up the domain tree until it
4720 finds some nameserver records. It should be used with care.
4722 (2) There is a new type, "mxh", which is like "mx" except that it returns
4723 just the host names, not the priorities.
4725 (3) It is now possible to give a list of domains (or IP addresses) to be
4726 looked up. The behaviour when one of the lookups defers can be
4727 controlled by a keyword.
4729 (4) It is now possible to specify the separator character for use when
4730 multiple records are returned.
4732 38. The dnslists ACL condition has been extended: it is now possible to supply
4733 a list of IP addresses and/or domains to be looked up in a particular DNS
4736 39. Added log_selector=+queue_time_overall.
4738 40. When running the queue in the test harness, wait just a tad after forking a
4739 delivery process, to get repeatability of debugging output.
4741 41. Include certificate and key file names in error message when GnuTLS fails
4742 to set them up, because the GnuTLS error message doesn't include the name
4743 of the failing file when there is a problem reading it.
4745 42. Allow both -bf and -bF in the same test run.
4747 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4749 44. The "Exiscan patch" is now merged into the mainline Exim source.
4751 45. Sometimes the final signoff response after QUIT could fail to get
4752 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4753 < 0 before doing a fflush(). This bug looks as though it goes back to the
4754 introduction of TLS in release 3.20, but "sometimes" must have been rare
4755 because the tests only now provoked it.
4757 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4758 (this can affect the format of dates).
4760 47. exim_tidydb, when checking for the continued existence of a message for
4761 which it has found a message-specific retry record, was not finding
4762 messages that were in split spool directories. Consequently, it was
4763 deleting retry records that should have stayed in existence.
4765 48. Steve fixed some bugs in eximstats.
4767 49. The SPA authentication driver was not abandoning authentication and moving
4768 on to the next authenticator when an expansion was forced to fail,
4769 contradicting the general specification for all authenticators. Instead it
4770 was generating a temporary error. It now behaves as specified.
4772 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4773 (the order specifies the preference for clients). The order is now AES256,
4774 AES128, 3DES, ARCFOUR128.
4776 51. Small patch to Sieve code - explicitly set From: when generating an
4779 52. Exim crashed if a remote delivery caused a very long error message to be
4780 recorded - for instance if somebody sent an entire SpamAssassin report back
4781 as a large number of 550 error lines. This bug was coincidentally fixed by
4782 increasing the size of one of Exim's internal buffers (big_buffer) that
4783 happened as part of the Exiscan merge. However, to be on the safe side, I
4784 have made the code more robust (and fixed the comments that describe what
4787 53. Now that there can be additional text after "Completed" in log lines (if
4788 the queue_time_overall log selector is set), a one-byte patch to exigrep
4789 was needed to allow it to recognize "Completed" as not the last thing in
4792 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4793 patch that reportedly fixes this has been added. I am not expert enough to
4794 create a test for it. This is what the patch creator wrote:
4796 "I found a little strange behaviour of ldap code when working with
4797 Windows 2003 AD Domain, where users was placed in more than one
4798 Organization Units. When I tried to give exim partial DN, the exit code
4799 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4800 But simultaneously result of request was absolutely normal ldap result,
4801 so I produce this patch..."
4803 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4804 REFERENCE, so I have modified the code to exclude the patch when that macro
4807 55. Some experimental protocols are using DNS PTR records for new purposes. The
4808 keys for these records are domain names, not reversed IP addresses. The
4809 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4810 leaves it alone. Component reversal etc. now happens only for IP addresses.
4813 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4815 57. Double the size of the debug message buffer (to 2048) so that more of very
4816 long debug lines gets shown.
4818 58. The exicyclog utility now does better if the number of log files to keep
4819 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4821 59. Two changes related to the smtp_active_hostname option:
4823 (1) $smtp_active_hostname is now available as a variable.
4824 (2) The default for smtp_banner uses $smtp_active_hostname instead
4825 of $primary_hostname.
4827 60. The host_aton() function is supposed to be passed a string that is known
4828 to be a valid IP address. However, in the case of IPv6 addresses, it was
4829 not checking this. This is a hostage to fortune. Exim now panics and dies
4830 if the condition is not met. A case was found where this could be provoked
4831 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4832 components; fortuitously, this particular loophole had already been fixed
4833 by change 4.50/55 above.
4835 If there are any other similar loopholes, the new check in host_aton()
4836 itself should stop them being exploited. The report I received stated that
4837 data on the command line could provoke the exploit when Exim was running as
4838 exim, but did not say which command line option was involved. All I could
4839 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4840 running as the user.
4843 61. There was a buffer overflow vulnerability in the SPA authentication code
4844 (which came originally from the Samba project). I have added a test to the
4845 spa_base64_to_bits() function which I hope fixes it.
4848 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4849 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4851 63. The daemon start-up calls getloadavg() while still root for those OS that
4852 need the first call to be done as root, but it missed one case: when
4853 deliver_queue_load_max is set with deliver_drop_privilege. This is
4854 necessary for the benefit of the queue runner, because there is no re-exec
4855 when deliver_drop_privilege is set.
4857 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4858 This has been fixed.
4860 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4861 was in use, was not putting the data itself into the right store pool;
4862 consequently, it could be overwritten for a subsequent message in the same
4863 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4866 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4868 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4869 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4870 the function were treating the return as a boolean value, which happened to
4871 work because 0=false and not-0=true, but is not correct code.
4873 68. The host_aton() function was not handling scoped IPv6 addresses (those
4874 with, for example, "%eth0" on the end) correctly.
4876 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4877 negated items (that is, ~something) in unsigned ints. Some compilers
4878 apparently mutter when there is no cast.
4880 70. If an address verification called from an ACL failed, and did not produce a
4881 user-specific message (i.e. there was only a "system" message), nothing was
4882 put in $acl_verify_message. In this situation, it now puts the system
4885 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4886 session; change 4.31/43 added the unwanted input to the log line - except
4887 that it did not do this in the start of session case. It now does.
4889 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4890 This is wrong and can cause the other end to generate a synchronization
4891 error if it is another Exim or anything else that does the synchronization
4892 check. A QUIT command is no longer sent after a timeout.
4894 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4895 during host lookups.
4897 74. The defer_ok option of callout verification was not working if it was used
4898 when verifying addresses in header lines, that is, for this case:
4900 verify = header_sender/callout=defer_ok
4902 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4903 those file descriptors could be used for SMTP connections. If anything
4904 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4905 could be sent to the SMTP client, causing chaos. The daemon now opens
4906 stdin, stdout, and stderr to /dev/null when it puts itself into the
4909 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4910 log by default. The user can override this with suitable Perl magic.
4912 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4913 the log message when discard triggers, was not working for the DATA ACL or
4914 for the non-SMTP ACL.
4916 78. Error message wording change in sieve.c.
4918 79. If smtp_accept_max_per_host was set, the number of connections could be
4919 restricted to fewer than expected, because the daemon was trying to set up
4920 a new connection before checking whether the processes handling previous
4921 connections had finished. The check for completed processes is now done
4922 earlier. On busy systems, this bug wouldn't be noticed because something
4923 else would have woken the daemon, and it would have reaped the completed
4926 80. If a message was submitted locally by a user whose login name contained one
4927 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4928 It caused a spool format error. I have fixed the spool reading code. A
4929 related problem was that the "from" clause in the Received: line became
4930 illegal because of the space(s). It is now covered by ${quote_local_part.
4932 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4935 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4937 83. Patch from Sieve maintainer to fix unterminated string problem in
4938 "vacation" handling.
4940 84. Some minor changes to the Linux configuration files to help with other
4941 OS variants using glibc.
4943 85. One more patch for Sieve to update vacation handling to latest spec.
4946 ----------------------------------------------------
4947 See the note above about the 4.44 and 4.50 releases.
4948 ----------------------------------------------------
4954 1. Change 4.43/35 introduced a bug that caused file counts to be
4955 incorrectly computed when quota_filecount was set in an appendfile
4958 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4959 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4962 3. Give more explanation in the error message when the command for a transport
4963 filter fails to execute.
4965 4. There are several places where Exim runs a non-Exim command in a
4966 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4967 was being done only for the command run by the queryprogram router. It is
4968 now done for all such subprocesses. The other cases are: ${run, transport
4969 filters, and the commands run by the lmtp and pipe transports.
4971 5. Some older OS have a limit of 256 on the maximum number of file
4972 descriptors. Exim was using setrlimit() to set 1000 as a large value
4973 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4974 systems. I've change it so that if it can't get 1000, it tries for 256.
4976 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4977 was an oversight, and furthermore, ever since the addition of extra
4978 controls (e.g. 4.43/32), the checks on when to allow different forms of
4979 "control" were broken. There should now be diagnostics for all cases when a
4980 control that does not make sense is encountered.
4982 7. $recipients is now available in the predata ACL (oversight).
4984 8. Tidy the search cache before the fork to do a delivery from a message
4985 received from the command line. Otherwise the child will trigger a lookup
4986 failure and thereby defer the delivery if it tries to use (for example) a
4987 cached ldap connection that the parent has called unbind on.
4989 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4990 of $address_data from the recipient verification was clobbered by the
4991 sender verification.
4993 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4994 was its contents. (It was OK if the option was not defined at all.)
4996 11. A "Completed" log line is now written for messages that are removed from
4997 the spool by the -Mrm option.
4999 12. $host_address is now set to the target address during the checking of
5000 ignore_target_hosts.
5002 13. When checking ignore_target_hosts for an ipliteral router, no host name was
5003 being passed; this would have caused $sender_host_name to have been used if
5004 matching the list had actually called for a host name (not very likely,
5005 since this list is usually IP addresses). A host name is now passed as
5008 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5009 code that specifies a non-restarting handler (typically sigaction() in
5010 modern systems) in an attempt to fix a rare and obscure crash bug.
5012 15. Narrowed the window for a race in the daemon that could cause it to ignore
5013 SIGCHLD signals. This is not a major problem, because they are used only to
5014 wake it up if nothing else does.
5016 16. A malformed maildirsize file could cause Exim to calculate negative values
5017 for the mailbox size or file count. Odd effects could occur as a result.
5018 The maildirsize information is now recalculated if the size or filecount
5021 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5022 support for a long time. Removed HAVE_SYS_VFS_H.
5024 18. Updated exipick to current release from John Jetmore.
5026 19. Allow an empty sender to be matched against a lookup in an address list.
5027 Previously the only cases considered were a regular expression, or an
5030 20. Exim went into a mad DNS lookup loop when doing a callout where the
5031 host was specified on the transport, if the DNS lookup yielded more than
5034 21. The RFC2047 encoding function was originally intended for short strings
5035 such as real names; it was not keeping to the 75-character limit for
5036 encoded words that the RFC imposes. It now respects the limit, and
5037 generates multiple encoded words if necessary. To be on the safe side, I
5038 have increased the buffer size for the ${rfc2047: expansion operator from
5041 22. Failure to deliver a bounce message always caused it to be frozen, even if
5042 there was an errors_to setting on the router. The errors_to setting is now
5045 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5046 canonical form (fully expanded) before being placed in
5047 $sender_host_address.
5049 24. Updated eximstats to version 1.33
5051 25. Include certificate and key file names in error message when GnuTLS fails
5052 to set them up, because the GnuTLS error message doesn't include the name
5053 of the failing file when there is a problem reading it.
5055 26. Expand error message when OpenSSL has problems setting up cert/key files.
5058 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5059 (this can affect the format of dates).
5061 28. exim_tidydb, when checking for the continued existence of a message for
5062 which it has found a message-specific retry record, was not finding
5063 messages that were in split spool directories. Consequently, it was
5064 deleting retry records that should have stayed in existence.
5066 29. eximstats updated to version 1.35
5067 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5068 1.35 - bugfix such that pie charts by volume are generated correctly
5070 30. The SPA authentication driver was not abandoning authentication and moving
5071 on to the next authenticator when an expansion was forced to fail,
5072 contradicting the general specification for all authenticators. Instead it
5073 was generating a temporary error. It now behaves as specified.
5075 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5076 (the order specifies the preference for clients). The order is now AES256,
5077 AES128, 3DES, ARCFOUR128.
5079 31. Small patch to Sieve code - explicitly set From: when generating an
5082 32. Exim crashed if a remote delivery caused a very long error message to be
5083 recorded - for instance if somebody sent an entire SpamAssassin report back
5084 as a large number of 550 error lines. This bug was coincidentally fixed by
5085 increasing the size of one of Exim's internal buffers (big_buffer) that
5086 happened as part of the Exiscan merge. However, to be on the safe side, I
5087 have made the code more robust (and fixed the comments that describe what
5090 33. Some experimental protocols are using DNS PTR records for new purposes. The
5091 keys for these records are domain names, not reversed IP addresses. The
5092 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5093 leaves it alone. Component reversal etc. now happens only for IP addresses.
5096 34. The host_aton() function is supposed to be passed a string that is known
5097 to be a valid IP address. However, in the case of IPv6 addresses, it was
5098 not checking this. This is a hostage to fortune. Exim now panics and dies
5099 if the condition is not met. A case was found where this could be provoked
5100 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5101 components; fortuitously, this particular loophole had already been fixed
5102 by change 4.50/55 or 4.44/33 above.
5104 If there are any other similar loopholes, the new check in host_aton()
5105 itself should stop them being exploited. The report I received stated that
5106 data on the command line could provoke the exploit when Exim was running as
5107 exim, but did not say which command line option was involved. All I could
5108 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5109 running as the user.
5112 35. There was a buffer overflow vulnerability in the SPA authentication code
5113 (which came originally from the Samba project). I have added a test to the
5114 spa_base64_to_bits() function which I hope fixes it.
5117 36. The daemon start-up calls getloadavg() while still root for those OS that
5118 need the first call to be done as root, but it missed one case: when
5119 deliver_queue_load_max is set with deliver_drop_privilege. This is
5120 necessary for the benefit of the queue runner, because there is no re-exec
5121 when deliver_drop_privilege is set.
5123 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5124 was in use, was not putting the data itself into the right store pool;
5125 consequently, it could be overwritten for a subsequent message in the same
5126 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5129 38. Sometimes the final signoff response after QUIT could fail to get
5130 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5131 < 0 before doing a fflush(). This bug looks as though it goes back to the
5132 introduction of TLS in release 3.20, but "sometimes" must have been rare
5133 because the tests only now provoked it.
5139 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5140 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5141 Now it returns an integer. A number of calls were still expecting a T/F
5142 return. Fortuitously, in all cases, the tests worked in OK situations,
5143 which is the norm. However, things would have gone wrong on any write
5144 failures on the smtp file descriptor. This function is used when sending
5145 messages over SMTP and also when doing verify callouts.
5147 2. When Exim is called to do synchronous delivery of a locally submitted
5148 message (the -odf or -odi options), it no longer closes stderr before doing
5151 3. Implemented the mua_wrapper option.
5153 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5155 5. Implemented the functions header_remove(), header_testname(),
5156 header_add_at_position(), and receive_remove_recipient(), and exported them
5159 6. If an ACL "warn" statement specified the addition of headers, Exim already
5160 inserted X-ACL-Warn: at the start if there was no header name. However, it
5161 was not making this test for the second and subsequent header lines if
5162 there were newlines in the string. This meant that an invalid header could
5163 be inserted if Exim was badly configured.
5165 7. Allow an ACL "warn" statement to add header lines at the start or after all
5166 the Received: headers, as well as at the end.
5168 8. Added the rcpt_4xx retry error code.
5170 9. Added postmaster_mailfrom=xxx to callout verification option.
5172 10. Added mailfrom=xxxx to the callout verification option, for verify=
5175 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5176 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5178 12. Inserted some casts to stop certain compilers warning when using pointer
5179 differences as field lengths or precisions in printf-type calls (mostly
5180 affecting debugging statements).
5182 13. Added optional readline() support for -be (dynamically loaded).
5184 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5185 same clock tick as a message's arrival, so that its received time was the
5186 same as the "first fail" time on the retry record, and that message
5187 remained on the queue past the ultimate address timeout, every queue runner
5188 would try a delivery (because it was past the ultimate address timeout) but
5189 after another failure, the ultimate address timeout, which should have then
5190 bounced the address, did not kick in. This was a "< instead of <=" error;
5191 in most cases the first failure would have been in the next clock tick
5192 after the received time, and all would be well.
5194 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5195 being recognized when the domain list was tested by the match_domain
5196 condition in an expansion string.
5198 16. Added the ${str2b64: operator.
5200 17. Exim was always calling setrlimit() to set a large limit for the number of
5201 processes, without checking whether the existing limit was already
5202 adequate. (It did check for the limit on file descriptors.) Furthermore,
5203 errors from getrlimit() and setrlimit() were being ignored. Now they are
5204 logged to the main and panic logs, but Exim does carry on, to try to do its
5205 job under whatever limits there are.
5207 18. Imported PCRE 5.0.
5209 19. Trivial typo in log message " temporarily refused connection" (the leading
5212 20. If the log selector return_path_on_delivery was set and an address was
5213 redirected to /dev/null, the delivery process crashed because it assumed
5214 that a return path would always be set for a "successful" delivery. In this
5215 case, the whole delivery is bypassed as an optimization, and therefore no
5218 21. Internal re-arrangement: the function for sending a challenge and reading
5219 a response while authentication was assuming a zero-terminated challenge
5220 string. It's now changed to take a pointer and a length, to allow for
5221 binary data in such strings.
5223 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5225 23. Exim was not respecting finduser_retries when seeking the login of the
5226 uid under which it was called; it was always trying 10 times. (The default
5227 setting of finduser_retries is zero.) Also, it was sleeping after the final
5228 failure, which is pointless.
5230 24. Implemented tls_on_connect_ports.
5232 25. Implemented acl_smtp_predata.
5234 26. If the domain in control=submission is set empty, Exim assumes that the
5235 authenticated id is a complete email address when it generates From: or
5236 Sender: header lines.
5238 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5239 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5240 chown and chgrp in /bin and hostname in /usr/bin.
5242 28. Exim was keeping the "process log" file open after each use, just as it
5243 does for the main log. This opens the possibility of it remaining open for
5244 long periods when the USR1 signal hits a daemon. Occasional processlog
5245 errors were reported, that could have been caused by this. Anyway, it seems
5246 much more sensible not to leave this file open at all, so that is what now
5249 29. The long-running daemon process does not normally write to the log once it
5250 has entered its main loop, and it closes the log before doing so. This is
5251 so that log files can straightforwardly be renamed and moved. However,
5252 there are a couple of unusual error situations where the daemon does write
5253 log entries, and I had neglected to close the log afterwards.
5255 30. The text of an SMTP error response that was received during a remote
5256 delivery was being truncated at 512 bytes. This is too short for some of
5257 the long messages that one sometimes sees. I've increased the limit to
5260 31. It is now possible to make retry rules that apply only when a message has a
5261 specific sender, in particular, an empty sender.
5263 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5264 it possible to be selective about when SMTP synchronization is enforced.
5266 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5268 32. Implemented hosts_connection_nolog.
5270 33. Added an ACL for QUIT.
5272 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5275 35. Added mailbox_size and mailbox_filecount to appendfile.
5277 36. Added control = no_multiline_responses to ACLs.
5279 37. There was a bug in the logic of the code that waits for the clock to tick
5280 in the case where the clock went backwards by a substantial amount such
5281 that the microsecond fraction of "now" was more than the microsecond
5282 fraction of "then" (but the whole seconds number was less).
5284 38. Added support for the libradius Radius client library this is found on
5285 FreeBSD (previously only the radiusclient library was supported).
5291 1. When certain lookups returned multiple values in the form name=value, the
5292 quoting of the values was not always being done properly. Specifically:
5293 (a) If the value started with a double quote, but contained no whitespace,
5295 (b) If the value contained whitespace other than a space character (i.e.
5296 tabs or newlines or carriage returns) it was not quoted.
5297 This fix has been applied to the mysql and pgsql lookups by writing a
5298 separate quoting function and calling it from the lookup code. The fix
5299 should probably also be applied to nisplus, ibase and oracle lookups, but
5300 since I cannot test any of those, I have not disturbed their existing code.
5302 2. A hit in the callout cache for a specific address caused a log line with no
5303 reason for rejecting RCPT. Now it says "Previous (cached) callout
5304 verification failure".
5306 3. There was an off-by-one bug in the queryprogram router. An over-long
5307 return line was truncated at 256 instead of 255 characters, thereby
5308 overflowing its buffer with the terminating zero. As well as fixing this, I
5309 have increased the buffer size to 1024 (and made a note to document this).
5311 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5312 when Exim is waiting for an SMTP response from a remote server, Exim
5313 restarts its select() call on the socket, thereby resetting its timeout.
5314 This is not a problem when such interrupts are rare. Somebody set up a cron
5315 job to run exiwhat every 2 minutes, which is less than the normal select()
5316 timeout (5 or 10 minutes). This meant that the select() timeout never
5317 kicked in because it was always reset. I have fixed this by comparing the
5318 time when an interrupt arrives with the time at the start of the first call
5319 to select(). If more time than the timeout has elapsed, the interrupt is
5320 treated as a timeout.
5322 5. Some internal re-factoring in preparation for the addition of Sieve
5323 extensions (by MH). In particular, the "personal" test is moved to a
5324 separate function, and given an option for scanning Cc: and Bcc: (which is
5325 not set for Exim filters).
5327 6. When Exim created an email address using the login of the caller as the
5328 local part (e.g. when creating a From: or Sender: header line), it was not
5329 quoting the local part when it contained special characters such as @.
5331 7. Installed new OpenBSD configuration files.
5333 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5334 try to make them clearer.
5336 9. Callout options, other than the timeout value, were being ignored when
5337 verifying sender addresses in header lines. For example, when using
5339 verify = header_sender/callout=no_cache
5341 the cache was (incorrectly) being used.
5343 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5344 only the Cygwin environment.
5346 11. When return_path_on_delivery was set as a log selector, if different remote
5347 addresses in the same message used different return paths and parallel
5348 remote delivery occurred, the wrong values would sometimes be logged.
5349 (Whenever a remote delivery process finished, the return path value from
5350 the most recently started remote delivery process was logged.)
5352 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5353 lines when AUTH and/or TLS are in use. This is the "received protocol"
5354 field. Exim used to use "asmtp" for authenticated SMTP, without any
5355 indication (in the protocol name) for TLS use. Now it follows the RFC and
5356 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5357 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5358 names appear in log lines as well as in Received: header lines.
5360 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5361 extensions, and comparison tests, and to fix some bugs.
5363 14. Changes to the "personal" filter test:
5365 (1) The test was buggy in that it was just doing the equivalent of
5366 "contains" tests on header lines. For example, if a user's address was
5367 anne@some.where, the "personal" test would incorrectly be true for
5369 To: susanne@some.where
5371 This test is now done by extracting each address from the header in turn,
5372 and checking the entire address. Other tests that are part of "personal"
5373 are now done using regular expressions (for example, to check local parts
5374 of addresses in From: header lines).
5376 (2) The list of non-personal local parts in From: addresses has been
5377 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5378 taken from the Sieve specification recommendations.
5380 (3) If the message contains any header line starting with "List-" it is
5381 treated as non-personal.
5383 (4) The test for "circular" in the Subject: header line has been removed
5384 because it now seems ill-conceived.
5386 15. Minor typos in src/EDITME comments corrected.
5388 16. Installed latest exipick from John Jetmore.
5390 17. If headers_add on a router specified a text string that was too long for
5391 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5392 of string_sprintf() is now avoided.
5394 18. $message_body_size was not set (it was always zero) when running the DATA
5395 ACL and the local_scan() function.
5397 19. For the "mail" command in an Exim filter, no default was being set for
5398 the once_repeat time, causing a random time value to be used if "once" was
5399 specified. (If the value happened to be <= 0, no repeat happened.) The
5400 default is now 0s, meaning "never repeat". The "vacation" command was OK
5401 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5402 (I found it when inspecting the code).
5404 20. There is now an overall timeout for performing a callout verification. It
5405 defaults to 4 times the callout timeout, which applies to individual SMTP
5406 commands during the callout. The overall timeout applies when there is more
5407 than one host that can be tried. The timeout is checked before trying the
5408 next host. This prevents very long delays if there are a large number of
5409 hosts and all are timing out (e.g. when the network connections are timing
5410 out). The value of the overall timeout can be changed by specifying an
5411 additional sub-option for "callout", called "maxwait". For example:
5413 verify = sender/callout=5s,maxwait=20s
5415 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5416 to the end before writing, but this should make it even safer).
5418 22. Exim was forgetting that it had advertised PIPELINING for the second and
5419 subsequent messages on an SMTP connection. It was also not resetting its
5420 memory on STARTTLS and an internal HELO.
5422 23. When Exim logs an SMTP synchronization error within a session, it now
5423 records whether PIPELINING has been advertised or not.
5425 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5426 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5427 rather than long int.
5429 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5431 26. Added the never_mail option to autoreply.
5437 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5438 crash if the getsockname() call failed; this can happen if a connection is
5439 closed very soon after it is established. The problem was simply in the
5440 order in which certain operations were done, causing Exim to try to write
5441 to the SMTP stream before it had set up the file descriptor. The bug has
5442 been fixed by making things happen in the correct order.
5448 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5449 before the connection was closed, thus losing the rejection response.
5451 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5452 some early Solaris releases, but causes trouble in current releases where
5453 socklen_t is defined.
5455 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5458 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5461 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5462 that releasing the top of it at the end releases what was used for sub-
5463 expansions (unless the block got too big). However, discard this block if
5464 the first thing is a variable or header, so that we can use its block when
5465 it is dynamic (useful for very large $message_headers, for example).
5467 6. Lookups now cache *every* query, not just the most recent. A new, separate
5468 store pool is used for this. It can be recovered when all lookup caches are
5469 flushed. Lookups now release memory at the end of their result strings.
5470 This has involved some general refactoring of the lookup sources.
5472 7. Some code has been added to the store_xxx() functions to reduce the amount
5473 of flapping under certain conditions.
5475 8. log_incoming_interface used to affect only the <= reception log lines. Now
5476 it causes the local interface and port to be added to several more SMTP log
5477 lines, for example "SMTP connection from", and rejection lines.
5479 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5481 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5483 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5484 could be overwritten at the end of the current message (or the start of a
5485 new message if it was set in a HELO ACL). The value is now preserved for
5486 the duration of the SMTP connection.
5488 12. If a transport had a headers_rewrite setting, and a matching header line
5489 contained an unqualified address, that address was qualified, even if it
5490 did not match any rewriting rules. The underlying bug was that the values
5491 of the flags that permit the existence of unqualified sender and recipient
5492 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5493 non-local messages, and by -bnq for local messages) were not being
5494 preserved with the message after it was received.
5496 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5497 "next input=" as part of the text comprising the host identity instead of
5498 the correct text. The code was using the same buffer for two different
5499 strings. However, depending on which order the printing function evaluated
5500 its arguments, the bug did not always show up. Under Linux, for example, my
5501 test suite worked just fine.
5503 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5504 doesn't work with some older versions of Perl. It has been changed to "my",
5505 which in any case is probably the better facility to use.
5507 15. A really picky compiler found some instances of statements for creating
5508 error messages that either had too many or two few arguments for the format
5511 16. The size of the buffer for calls to the DNS resolver has been increased
5512 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5513 for addresses that have a lot of PTR records. This alleviates a problem; it
5514 does not fully solve it.
5516 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5517 buffer now truncates the list and logs the incident, which is the same
5518 action as happens when Exim is looking up a host name and its aliases.
5519 Previously in this situation something unpredictable would happen;
5520 sometimes it was "internal error: store_reset failed".
5522 18. If a server dropped the connection unexpectedly when an Exim client was
5523 using GnuTLS and trying to read a response, the client delivery process
5524 crashed while trying to generate an error log message.
5526 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5527 string, for example:
5529 warn message = H1: something\nH2: something
5531 the text was added as a single header line from Exim's point of view
5532 though it ended up OK in the delivered message. However, searching for the
5533 second and subsequent header lines using $h_h2: did not work. This has been
5534 fixed. Similarly, if a system filter added multiple headers in this way,
5535 the routers could not see them.
5537 20. Expanded the error message when iplsearch is called with an invalid key to
5538 suggest using net-iplsearch in a host list.
5540 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5541 ACLs are no longer actually imposed (and a message to that effect is
5544 22. If a "gecos" field in a passwd entry contained escaped characters, in
5545 particular, if it contained a \" sequence, Exim got it wrong when building
5546 a From: or a Sender: header from that name. A second bug also caused
5547 incorrect handling when an unquoted " was present following a character
5548 that needed quoting.
5550 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5551 was not being matched caselessly.
5553 24. Arranged for all hyphens in the exim.8 source to be escaped with
5556 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5557 properly. Recipient callout cache records were still being keyed to include
5558 the sender, even when use_sender was set false. This led to far more
5559 callouts that were necessary. The sender is no longer included in the key
5560 when use_sender is false.
5562 26. Added "control = submission" modifier to ACLs.
5564 27. Added the ${base62d: operator to decode base 62 numbers.
5566 28. dnsdb lookups can now access SRV records.
5568 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5569 the configuration file.
5571 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5572 (-v) mode. This makes the output for a verbose queue run more intelligible.
5574 31. Added a use_postmaster feature to recipient callouts.
5576 32. Added the $body_zerocount variable, containing the number of binary zero
5577 bytes in the message body.
5579 33. The time of last modification of the "new" subdirectory is now used as the
5580 "mailbox time last read" when there is a quota error for a maildir
5583 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5585 35. Added +ignore_unknown as a special item in host lists.
5587 36. Code for decoding IPv6 addresses in host lists is now included, even if
5588 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5589 address was recognized as an IP address, but was then not correctly decoded
5590 into binary, causing unexpected and incorrect effects when compared with
5597 1. Very minor rewording of debugging text in manualroute to say "list of
5598 hosts" instead of "hostlist".
5600 2. If verify=header_syntax was set, and a header line with an unqualified
5601 address (no domain) and a large number of spaces between the end of the
5602 name and the colon was received, the reception process suffered a buffer
5603 overflow, and (when I tested it) crashed. This was caused by some obsolete
5604 code that should have been removed. The fix is to remove it!
5606 3. When running in the test harness, delay a bit after writing a bounce
5607 message to get a bit more predictability in the log output.
5609 4. Added a call to search_tidyup() just before forking a reception process. In
5610 theory, someone could use a lookup in the expansion of smtp_accept_max_
5611 per_host which, without the tidyup, could leave open a database connection.
5613 5. Added the variables $recipient_data and $sender_data which get set from a
5614 lookup success in an ACL "recipients" or "senders" condition, or a router
5615 "senders" option, similar to $domain_data and $local_part_data.
5617 6. Moved the writing of debug_print from before to after the "senders" test
5620 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5621 problems for message scanning, either using a data ACL, or using
5622 local_scan() because the Received: header was not generated till after they
5623 were called (in order to set the time as the time of reception completion).
5624 I have revised the way this works. The header is now generated after the
5625 body is received, but before the ACL or local_scan() are called. After they
5626 are run, the timestamp in the header is updated.
5632 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5633 before starting a queue runner without re-exec. This happened only when
5634 deliver_drop_privilege was set or when the Exim user was set to root. The
5635 effect of the bug was that timeouts during subsequent deliveries caused
5636 crashes instead of being properly handled. The handler is now left at its
5637 default (and expected) setting.
5639 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5640 message, again when deliver_drop_privilege is set or Exim is run as root.
5641 The bug described in (1) was not present in this case, but the tidying up
5642 of the other signals was missing. I have made the two cases consistent.
5644 3. The ignore_target_hosts setting on a manualroute router was being ignored
5645 for hosts that were looked up using the /MX notation.
5647 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5650 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5651 operated on the sender address. After changing the $sender_address to <>
5652 for the sender address verify, Exim was re-instated it as the original
5653 (before rewriting) address, but remembering that it had rewritten it, so it
5654 wasn't rewriting it again. This bug also had the effect of breaking the
5655 sender address verification caching when the sender address was rewritten.
5657 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5658 This has been changed so that if the ip literal address matches
5659 ignore_target_hosts, the router declines.
5661 7. Added expansion conditions match_domain, match_address, and match_local_
5662 part (NOT match_host).
5664 8. The placeholder for the Received: header didn't have a length field set.
5666 9. Added code to Exim itself and to exim_lock to test for a specific race
5667 condition that could lead to file corruption when using MBX delivery. The
5668 issue is with the lockfile that is created in /tmp. If this file is removed
5669 after a process has opened it but before that process has acquired a lock,
5670 there is the potential for a second process to recreate the file and also
5671 acquire a lock. This could lead to two Exim processes writing to the file
5672 at the same time. The added code performs the same test as UW imapd; it
5673 checks after acquiring the lock that its file descriptor still refers to
5674 the same named file.
5676 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5677 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5680 11. Added the smtp_active_hostname option. If used, this will typically be made
5681 to depend on the incoming interface address. Because $interface_address is
5682 not set up until the daemon has forked a reception process, error responses
5683 that can happen earlier (such as "too many connections") no longer contain
5686 12. If an expansion in a condition on a "warn" statement fails because a lookup
5687 defers, the "warn" statement is abandoned, and the next ACL statement is
5688 processed. Previously this caused the whole ACL to be aborted.
5690 13. Added the iplsearch lookup type.
5692 14. Added ident_timeout as a log selector.
5694 15. Added tls_certificate_verified as a log selector.
5696 16. Added a global option tls_require_ciphers (compare the smtp transport
5697 option of the same name). This controls incoming TLS connections.
5699 17. I finally figured out how to make tls_require_ciphers do a similar thing
5700 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5701 before starting the TLS session.
5703 18. Tabs are now shown as \t in -bP output.
5705 19. If the log selector return_path_on_delivery was set, Exim crashed when
5706 bouncing a message because it had too many Received: header lines.
5708 20. If two routers both had headers_remove settings, and the first one included
5709 a superfluous trailing colon, the final name in the first list and the
5710 first name in the second list were incorrectly joined into one item (with a
5711 colon in the middle).
5717 1. Added -C and -D options to the exinext utility, mainly to make it easier
5718 to include in the automated testing, but these could be helpful when
5719 multiple configurations are in use.
5721 2. The exinext utility was not formatting the output nicely when there was
5722 an alternate port involved in the retry record key, nor when there was a
5723 message id as well (for retries that were specific to a specific message
5724 and a specific host). It was also confused by IPv6 addresses, because of
5725 the additional colons they contain. I have fixed the IPv4 problem, and
5726 patched it up to do a reasonable job for IPv6.
5728 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5729 delivery, the log line now contains "pipelined" if PIPELINING was used.
5731 4. An SMTP transport process used to panic and die if the bind() call to set
5732 an explicit outgoing interface failed. This has been changed; it is now
5733 treated in the same way as a connect() failure.
5735 5. A reference to $sender_host_name in the part of a conditional expansion
5736 that was being skipped was still causing a DNS lookup. This no longer
5739 6. The def: expansion condition was not recognizing references to header lines
5740 that used bh_ and bheader_.
5742 7. Added the _cache feature to named lists.
5744 8. The code for checking quota_filecount in the appendfile transport was
5745 allowing one more file than it should have been.
5747 9. For compatibility with Sendmail, the command line option
5755 and sets the incoming protocol and host name (for trusted callers). The
5756 host name and its colon can be omitted when only the protocol is to be set.
5757 Note the Exim already has two private options, -pd and -ps, that refer to
5758 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5759 "s", but I don't think that's a major issue.
5761 10. A number of refactoring changes to the code, none of which should affect
5764 (a) The number of logging options was getting close to filling up the
5765 32-bit word that was used as a bit map. I have split them into two classes:
5766 those that are passed in the argument to log_write(), and those that are
5767 only ever tested independently outside of that function. These are now in
5768 separate 32-bit words, so there is plenty of room for expansion again.
5769 There is no change in the user interface or the logging behaviour.
5771 (b) When building, for example, log lines, the code previously used a
5772 macro that called string_cat() twice, in order to add two strings. This is
5773 not really sufficiently general. Furthermore, there was one instance where
5774 it was actually wrong because one of the argument was used twice, and in
5775 one call a function was used. (As it happened, calling the function twice
5776 did not affect the overall behaviour.) The macro has been replaced by a
5777 function that can join an arbitrary number of extra strings onto a growing
5780 (c) The code for expansion conditions now uses a table and a binary chop
5781 instead of a serial search (which was left over from when there were very
5782 few conditions). Also, it now recognizes conditions like "pam" even when
5783 the relevant support is not compiled in: a suitably worded error message is
5784 given if an attempt is made to use such a condition.
5786 11. Added ${time_interval:xxxxx}.
5788 12. A bug was causing one of the ddress fields not to be passed back correctly
5789 from remote delivery subprocesses. The field in question was not being
5790 subsequently used, so this caused to problems in practice.
5792 13. Added new log selectors queue_time and deliver_time.
5794 14. Might have fixed a bug in maildirsizefile handling that threw up
5795 "unexpected character" debug warnings, and recalculated the data
5796 unnecessarily. In any case, I expanded the warning message to give more
5799 15. Added the message "Restricted characters in address" to the statements in
5800 the default ACL that block characters like @ and % in local parts.
5802 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5803 Three changes have been made:
5805 (a) There was a serious bug; a negative response to MAIL caused the whole
5806 recipient domain to be cached as invalid, thereby blocking all messages
5807 to all local parts at the same domain, from all senders. This bug has
5808 been fixed. The domain is no longer cached after a negative response to
5809 MAIL if the sender used is not empty.
5811 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5814 (c) A new callout option, "use_sender" has been added for people who want
5815 the modified behaviour.
5821 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5824 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5825 indeed breaks things for older releases.
5827 3. Added additional logging to the case where there is a problem reading data
5828 from a filter that is running in a subprocess using a pipe, in order to
5829 try to track down a specific problem.
5831 4. Testing facility fudge: when running in the test harness and attempting
5832 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5833 getting "No route to host". Convert this to a timeout.
5835 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5838 6. Some OS don't have socklen_t but use size_t instead. This affects the
5839 fifth argument of getsockopt() amongst other things. This is now
5840 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5841 can be set for individual OS. I have set it for SunOS5, OSF1, and
5842 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5843 some earlier ones do not.
5845 7. Change 4.30/15 was not doing the test caselessly.
5847 8. The standard form for an IPv6 address literal was being rejected by address
5848 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5849 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5850 this, as well as the form without the "IPv6" on the front (but only when
5851 address literals are enabled, of course).
5853 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5855 10. Exim crashed if a message with an empty sender address specified by -f
5856 encountered a router with an errors_to setting. This could be provoked only
5857 by a command such as
5861 where an empty string was supplied; "<>" did not hit this bug.
5863 11. Installed PCRE release 4.5.
5865 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5866 remained set. It is now erased.
5868 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5869 times from message ids (which are base 36 rather than the normal 62).
5871 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5872 were being counted as actual protocol errors, and logged if the log
5873 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5874 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5875 and DATA following a set of rejected RCPTs do not count as protocol errors.
5876 In other words, Exim assumes they were pipelined, though this may not
5877 actually be the case. Of course, in all cases the client gets an
5878 appropriate error code.
5880 15. If a lookup fails in an ACL condition, a message about the failure may
5881 be available; it is used if testing the ACL cannot continue, because most
5882 such messages specify what the cause of the deferral is. However, some
5883 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5884 that caused an old message to be retained and used if a later statement
5885 caused a defer, replacing the real cause of the deferral.
5887 16. If an IP address had so many PTR records that the DNS lookup buffer
5888 was not large enough to hold them, Exim could crash while trying to process
5889 the truncated data. It now detects and logs this case.
5891 17. Further to 4.21/58, another change has been made: if (and only if) the
5892 first line of a message (the first header line) ends with CRLF, a bare LF
5893 in a subsequent header line has a space inserted after it, so as not to
5894 terminate the header.
5896 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5897 unnecessarily, used atoi() instead of strtol(), and didn't check the
5898 termination when getting file sizes from file names by regex.
5900 19. Completely re-implemented the support for maildirsize files, in the light
5901 of a number of problems with the previous contributed implementation
5902 (4.30/29). In particular:
5904 . If the quota is zero, the maildirsize file is maintained, but no quota is
5907 . If the maildir directory does not exist, it is created before any attempt
5908 to write a maildirsize file.
5910 . The quota value in the file is just a cache; if the quota is changed in
5911 the transport, the new value overrides.
5913 . A regular expression is available for excluding directories from the
5916 20. The autoreply transport checks the characters in options that define the
5917 message's headers; it allows continued headers, but it was checking with
5918 isspace() after an embedded newline instead of explicitly looking for a
5921 21. If all the "regular" hosts to which an address was routed had passed their
5922 expiry times, and had not reached their retry times, the address was
5923 bounced, even if fallback hosts were defined. Now Exim should go on to try
5926 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5927 equivalent code in the SMTP transport. Some hosts send humungous responses
5928 to HELO/EHLO, more than 1024 it seems.
5930 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5931 gives compiler warnings in some environments. I've now done it "properly",
5934 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5935 (because of problems with the built-in one) was declared to return uschar *
5936 instead of char *, causing compiler failure.
5938 25. Fixed a file descriptor leak when processing alias/forward files.
5940 26. Fixed a minor format string issue in dbfn.c.
5942 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5944 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5945 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5948 29. When a sender address is verified, it is cached, to save repeating the test
5949 when there is more than one recipient in a message. However, when the
5950 verification involves a callout, it is possible for different callout
5951 options to be set for different recipients. It is too complicated to keep
5952 track of this in the cache, so now Exim always runs a verification when a
5953 callout is required, relying on the callout cache for the optimization.
5954 The overhead is duplication of the address routing, but this should not be
5957 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5958 to be verified with callout=postmaster, and the main callout worked but the
5959 postmaster check failed, the verification correctly failed. However, if a
5960 subsequent RCPT command asked for sender verification *without* the
5961 postmaster check, incorrect caching caused this verification also to fail,
5964 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5965 it was not caching the DNS options (qualify_single, search_parents) that
5966 were used when the lookup failed. A subsequent lookup with different
5967 options therefore always gave the same answer, though there were cases
5968 where it should not have. (Example: a "domains = !$mx_any" option on a
5969 dnslookup router: the "domains" option is always processed without any
5970 widening, but the router might have qualify_single set.) Now Exim uses the
5971 cached value only when the same options are set.
5973 32. Added John Jetmore's "exipick" utility to the distribution.
5975 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5976 than a timeout (e.g. a certificate is required, and is not provided), an
5977 Exim server now closes the connection immediately. Previously it waited for
5978 the client to close - but if the client is SSL, it seems that they each
5979 wait for each other, leading to a delay before one of them times out.
5981 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5982 maintained 0.8.x compatibility because I don't think many are using it, and
5983 it is clearly obsolete.
5985 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5988 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5989 client certificate was expired. A simple patch fixes this, though I don't
5990 understand the full logic of why the verify callback is called multiple
5993 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5994 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5995 which causes problems with some clients (such as the Certicom SSL Plus
5996 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5997 disables the coutermeasure allowing Eudora to connect."
5999 38. Exim was not checking that a write() to a log file succeeded. This could
6000 lead to Bad Things if a log got too big, in particular if it hit a file
6001 size limit. Exim now panics and dies if it cannot write to a log file, just
6002 as it does if it cannot open a log file.
6004 39. Modified OS/Makefile-Linux so that it now contains
6006 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
6008 The two -D definitions ensure that Exim is compiled with large file
6009 support, which makes it possible to handle log files that are bigger than
6012 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
6013 instance) a domain was checked against a named list that involved a lookup,
6014 causing $domain_data to be set, then another domain was checked against the
6015 same list, then the first domain was re-checked, the value of $domain_data
6016 after the final check could be wrong. In particular, if the second check
6017 failed, it could be set empty. This bug probably also applied to
6020 41. The strip_trailing_dot option was not being applied to the address given
6021 with the -f command-line option.
6023 42. The code for reading a message's header from the spool was incrementing
6024 $received_count, but never initializing it. This meant that the value was
6025 incorrect (doubled) while delivering a message in the same process in which
6026 it was received. In the most common configuration of Exim, this never
6027 happens - a fresh exec is done - but it can happen when
6028 deliver_drop_privilege is set.
6030 43. When Exim logs an SMTP synchronization error - client data sent too soon -
6031 it now includes up to 150 characters of the unexpected data in the log
6034 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6035 and building data strings. The size of both of these buffers was 10 000
6036 bytes - far larger than anybody would *ever* want, thought I. Needless to
6037 say, somebody hit the limit. I have increased the maximum line length to
6038 20 000 and the maximum data length of concatenated lines to 100 000. I have
6039 also fixed two bugs, because there was no checking on these buffers. Tsk,
6040 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6041 buffer is too small.
6043 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6044 lsearch lookups. Now it does.
6046 46. When parsing a route_list item in a manualroute router, a fixed-length
6047 buffer was used for the list of hosts. I made this 1024 bytes long,
6048 thinking that nobody would ever have a list of hosts that long. Wrong.
6049 Somebody had a whole pile of complicated expansion conditions, and the
6050 string was silently truncated, leading to an expansion error. It turns out
6051 that it is easier to change to an unlimited length (owing to other changes
6052 that have happened since this code was originally written) than to build
6053 structure for giving a limitation error. The length of the item that
6054 expands into the list of hosts is now unlimited.
6056 47. The lsearch lookup could not handle data where the length of text line was
6057 more than 4095 characters. Such lines were truncated, leading to shortened
6058 data being returned. It should now handle lines of any length.
6060 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6061 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6064 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6065 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6066 no longer happens in comment lines. A list of replacements is now placed
6067 at the head of all of the source files, except those whose only change is
6068 to replace PERL_COMMAND in the very first #! line.
6070 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6071 messages on the queue, with a bottom-up merge sort, using code contributed
6072 by Michael Haardt. This should make operations like -bp somewhat faster on
6073 large queues. It won't affect queue runners, except when queue_run_in_order
6076 51. Installed eximstats 1.31 in the distribution.
6078 52. Added support for SRV lookups to the dnslookup router.
6080 53. If an ACL referred to $message_body or $message_body_end, the value was not
6081 reset for any messages that followed in the same SMTP session.
6083 54. The store-handling optimization for building very long strings was not
6084 differentiating between the different store pools. I don't think this
6085 actually made any difference in practice, but I've tidied it.
6087 55. While running the routers to verify a sender address, $sender_address
6088 was still set to the sender address. This is wrong, because when routing to
6089 send a bounce to the sender, it would be empty. Therefore, I have changed
6090 it so that, while verifying a sender address, $sender_address is set to <>.
6091 (There is no change to what happens when verifying a recipient address.)
6093 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6094 target A or AAAA records (if not already returned) without resetting the
6095 qualify_single or search_parents options of the DNS resolver. These are
6096 inappropriate in this case because the targets of MX and SRV records must
6097 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6098 target that, when qualified, matched something in the local domain. These
6099 two options are now turned off when doing these lookups.
6101 57. It seems that at least some releases of Reiserfs (which does not have the
6102 concept of a fixed number of inodes) returns zero and not -1 for the
6103 number of available inodes. This interacted badly with check_spool_inodes,
6104 which assumed that -1 was the "no such thing" setting. What I have done is
6105 to check that the total number of inodes is greater than zero before doing
6106 the test of how many are available.
6108 58. When a "warn" ACL statement has a log_message modifier, the message is
6109 remembered, and not repeated. This is to avoid a lot of repetition when a
6110 message has many recipients that cause the same warning to be written.
6111 However, Exim was preserving the list of already written lines for an
6112 entire SMTP session, which doesn't seem right. The memory is now reset if a
6113 new message is started.
6115 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6116 debugging output unless log_rewrite was also set.
6118 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6119 of (int)(handle) when we know that handle contains (void *)(-1).
6121 61. The Exim daemon panic-logs an error return when it closes the incoming
6122 connection. However "connection reset by peer" seems to be common, and
6123 isn't really an error worthy of noting specially, so that particular error
6126 62. When Exim is trying to find all the local interfaces, it used to panic and
6127 die if the ioctl to get the interface flags failed. However, it seems that
6128 on at least one OS (Solaris 9) it is possible to have an interface that is
6129 included in the list of interfaces, but for which you get a failure error
6130 for this call. This happens when the interface is not "plumbed" into a
6131 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6132 failure of the "get flags" call assumes that the interface is down.
6134 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6135 makes life easier for people who are doing arithmetic on fields extracted
6136 from dates, where you often get leading zeros that should not be
6137 interpreted as octal.
6139 64. Added qualify_domain to the redirect router, to override the global
6142 65. If a pathologically long header line contained very many addresses (the
6143 report of this problem mentioned 10 000) and each of them was rewritten,
6144 Exim could use up a very large amount of memory. (It kept on making new
6145 copies of the header line as it rewrote, and never released the old ones.)
6146 At the expense of a bit more processing, the header rewriting function has
6147 been changed so that it no longer eats memory in this way.
6149 66. The generation of the Received: header has been moved from the time that a
6150 message starts to be received, to the time that it finishes. The timestamp
6151 in the Received: header should now be very close to that of the <= log
6152 line. There are two side-effects of this change:
6154 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6155 logged header lines no longer include the local Received: line, because
6156 it has not yet been created. The same applies to a copy of the message
6157 that is returned to a non-SMTP sender when a message is rejected.
6159 (b) When a filter file is tested using -bf, no additional Received: header
6160 is added to the test message. After some thought, I decided that this
6163 This change does not affect the value of $received_for. It is still set
6164 after address rewriting, but before local_scan() is called.
6166 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6168 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6169 gave an unhelpful panic error message, and a defer error. I have managed to
6170 change this behaviour so that it now rejects any supplied certificate,
6171 which seems right, as the list of acceptable certificates is empty.
6173 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6174 gave an unhelpful defer error. I have not managed to make this reject any
6175 supplied certificates, but the error message it gives is "no certificate
6176 supplied", which is not helpful.
6178 70. exigrep's output now also includes lines that are not associated with any
6179 message, but which match the given pattern. Implemented by a patch from
6180 Martin Sluka, which also tidied up the Perl a bit.
6182 71. Recipient callout verification, like sender verification, was using <> in
6183 the MAIL FROM command. This isn't really the right thing, since the actual
6184 sender may affect whether the remote host accepts the recipient or not. I
6185 have changed it to use the actual sender in the callout; this means that
6186 the cache record is now keyed on a recipient/sender pair, not just the
6187 recipient address. There doesn't seem to be a real danger of callout loops,
6188 since a callout by the remote host to check the sender would use <>.
6189 [SEE ABOVE: changed after hitting problems.]
6191 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6192 temporary errors. However, in the case of such a code being given after
6193 the end of a data transmission (i.e. after ".") Exim was failing to write
6194 a retry record for the message. (Yes, there was some broken host that was
6195 actually sending 8xx at this point.)
6197 73. An unknown lookup type in a host list could cause Exim to panic-die when
6198 the list was checked. (An example that provoked this was putting <; in the
6199 middle of a list instead of at the start.) If this happened during a DATA
6200 ACL check, a -D file could be left lying around. This kind of configuration
6201 error no longer causes Exim to die; instead it causes a defer error. The
6202 incident is still logged to the main and panic logs.
6204 74. Buglet left over from Exim 3 conversion. The message "too many messages
6205 in one connection" was written to the rejectlog but not the mainlog, except
6206 when address rewriting (yes!) was being logged.
6208 75. Added write_rejectlog option.
6210 76. When a system filter was run not as root (that is, when system_filter_user
6211 was set), the values of the $n variables were not being returned to the
6212 main process; thus, they were not subsequently available in the $sn
6215 77. Added +return_path_on_delivery log selector.
6217 78. A connection timeout was being treated differently from recipients deferred
6218 when testing hosts_max_try with a message that was older than the host's
6219 retry timeout. (The host should not be counted, thus allowing all hosts to
6220 be tried at least once before bouncing.) This may have been the cause of an
6221 occasionally reported bug whereby a message would remain on the queue
6222 longer than the retry timeout, but would be bounced if a delivery was
6223 forced. I say "may" because I never totally pinned down the problem;
6224 setting up timeout/retry tests is difficult. See also the next item.
6226 79. The ultimate address timeout was not being applied to errors that involved
6227 a combination of host plus message (for example, a timeout on a MAIL
6228 command). When an address resolved to a number of possible hosts, and they
6229 were not all tried for each delivery (e.g. because of hosts_max_try), a
6230 message could remain on the queue longer than the retry timeout.
6232 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6235 81. Fixed an obscure SMTP outgoing bug which required at least the following
6236 conditions: (a) there was another message waiting for the same server;
6237 (b) the server returned 5xx to all RCPT commands in the first message so
6238 that the message was not completed; (c) the server dropped the connection
6239 or gave a negative response to the RSET that Exim sends to abort the
6240 transaction. The observed case was a dropped connection after DATA that had
6241 been sent in pipelining mode. That is, the server had advertised PIPELINING
6242 but was not implementing it correctly. The effect of the bug was incorrect
6243 behaviour, such as trying another host, and this could lead to a crash.
6249 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6250 and daemon.c were passed as pointers to ints; they should have been
6251 pointers to socklen_t variables (which are typically unsigned ints).
6253 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6256 3. Fixed a really odd bug that affected only the testing scheme; patching a
6257 certain fixed string in the binary changed the value of another string that
6258 happened to be identical to the end of the original first string.
6260 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6261 name", it returns that address as the IP address. On some operating
6262 systems (e.g. Solaris), it also passes back the IP address string as the
6263 "host name". However, on others (e.g. Linux), it passes back an empty
6264 string. Exim wasn't checking for this, and was changing the host name to an
6265 empty string, assuming it had been canonicalized.
6267 5. Although rare, it is permitted to have more than one PTR record for a given
6268 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6269 all the names associated with an address, because they do in Solaris.
6270 However, it seems that they do not in Linux for data that comes from the
6271 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6272 I found this out when I moved to a new Linux workstation and tried to run
6273 the Exim test suite.
6275 To get round this problem I have changed the code so that it now does its
6276 own call to the DNS to look up PTR records when searching for a host name.
6277 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6278 addresses that are only in /etc/hosts are still found.
6280 This behaviour is, however, controlled by an option called host_lookup_
6281 order, which defaults to "bydns:byaddr". If people want to use the other
6282 order, or indeed, just use one or the other means of lookup, they can
6283 specify it in this variable.
6285 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6286 some operating systems, this comes back from gethostbyaddr() as an empty
6287 string, and this is what Exim used to test for. However, it seems that in
6288 other systems, "." is yielded. Exim now tests for this case too.
6290 7. The values of check_spool_space and check_log_space are now held internally
6291 as a number of kilobytes instead of an absolute number of bytes. If a
6292 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6293 kilobyte. This means that much larger values can be stored.
6295 8. Exim monitor: an attempt to get the action menu when not actually pointing
6296 at a message produces an empty menu entitled "No message selected". This
6297 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6298 no entries in it ("Shell widget menu has zero width and/or height"). So I
6299 have added a single, blank menu entry in this case.
6301 9. Added ${quote_local_part.
6303 10. MIME decoding is now applied to the contents of Subject: header lines when
6306 11. Now that a reference to $sender_host_address automatically causes a reverse
6307 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6308 host lookup before query-style lookups in lists that might use this
6309 variable. This has therefore been abolished, and the "net-" prefix is no
6310 longer necessary for query-style lookups.
6312 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6313 have been a typo for LFLAGS, so it has been changed.
6315 13. The install script calls Exim with "-C /dev/null" in order to find the
6316 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6317 to be output. However, since Exim outputs its version number before the
6318 error, it didn't break the script. It just looked ugly. I fixed this by
6319 always allowing "-C /dev/null" if the caller is root.
6321 14. Ignore overlarge ACL variable number when reading spool file - insurance
6322 against a later release with more variables having written the file.
6324 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6325 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6326 this, as well as the form without the "IPv6" on the front.
6328 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6329 OS/Makefile-Darwin file.
6331 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6332 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6334 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6335 variable. However, some OS use ints for inodes. I've added cast to long int
6336 to get rid of the compiler warning.
6338 19. I had forgotten to lock out "/../" in configuration file names when
6339 ALT_CONFIG_PREFIX was set.
6341 20. Routers used for verification do not need to specify transports. However,
6342 if such a router generated a host list, and callout was configured, Exim
6343 crashed, because it could not find a port number from the (non-existent)
6344 transport. It now assumes port 25 in this circumstance.
6346 21. Added the -t option to exigrep.
6348 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6349 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6350 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6351 them is set, LOOKUP_LSEARCH is forced.
6353 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6355 24. Added sender and host information to the "rejected by local_scan()" log
6356 line; previously there was no indication of these.
6358 25. Added .include_if_exists.
6360 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6361 Linux. It turns out that not all file systems support this. Apparently some
6362 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6363 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6364 not supported on the file descriptor, is now ignored when Exim is trying to
6365 sync a directory. This applies only to Linux.
6367 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6369 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6370 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6373 29. Added support for maildirsize files from supplied patch (modified a bit).
6375 30. The use of :fail: followed by an empty string could lead Exim to respond to
6376 sender verification failures with (e.g.):
6378 550 Verification failed for <xxx>
6379 550 Sender verify failed
6381 where the first response line was missing the '-' that indicates it is not
6382 the final line of the response.
6384 31. The loop for finding the name of the user that called Exim had a hardwired
6385 limit of 10; it now uses the value of finduser_retries, which is used for
6386 all other user lookups.
6388 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6391 33. Exim was neglecting to zero errno before one call of strtol() when
6392 expanding a string and expecting an integer value. On some systems this
6393 resulted in spurious "integer overflow" errors. Also, it was casting the
6394 result into an int without checking.
6396 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6397 did not work. The code looks as if it has *never* worked, though it appears
6398 to have been documented since at least release 1.62. I have made it work.
6400 35. The "timeout_DNS" error in retry rules, also documented since at least
6401 1.62, also never worked. As it isn't clear exactly what this means, and
6402 clearly it isn't a major issue, I have abolished the feature by treating it
6403 as "timeout", and writing a warning to the main and panic logs.
6405 36. The display of retry rules for -brt wasn't always showing the error code
6408 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6409 timeout_connect_A, timeout_connect_MX.
6411 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6412 to the empty sender.
6414 39. The daemon was not analysing the content of -oX till after it had closed
6415 stderr and disconnected from the controlling terminal. This meant that any
6416 syntax errors were only noted on the panic log, and the return code from
6417 the command was 0. By re-arranging the code a little, I've made the
6418 decoding happen first, so such errors now appear on stderr, and the return
6419 code is 1. However, the actual setting up of the sockets still happens in
6420 the disconnected process, so errors there are still only recorded on the
6423 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6424 connections (as happens on some IP stacks) was logged at start up time as
6425 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6426 it from "IPv6 and IPv4", which means that two separate sockets are being
6429 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6430 says whether AF_INET or AF_INET6 was passed as an argument.
6432 42. Exiwhat output was messed up when time zones were included in log
6435 43. Exiwhat now gives more information about the daemon's listening ports,
6436 and whether -tls-on-connect was used.
6438 44. The "port" option of the smtp transport is now expanded.
6440 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6441 silently ignored. Now an error message is written to the main and panic
6444 46. There's a new ACL modifier called "logwrite" which writes to a log file
6445 as soon as it is encountered.
6447 47. Added $local_user_uid and $local_user_gid at routing time.
6449 48. Exim crashed when trying to verify a sender address that was being
6452 49. Exim was recognizing only a space character after ".include". It now also
6453 recognizes a tab character.
6455 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6456 extracting the relevant information from the specification. The man page no
6457 longer contains scrambled data for the -d option, and I've added a section
6458 at the front about calling Exim under different names.
6460 51. Added "extra_headers" argument to the "mail" command in filter files.
6462 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6465 53. Installed eximstats 1.29.
6467 54. Added transport_filter_timeout as a generic transport option.
6469 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6470 Cc: header lines. This was required by RFC 822, but it not required by RFC
6473 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6474 incoming messages that did not have them. Now it does so only if the
6475 message originates locally, that is, if there is no associated remote host
6476 address. When Resent- header lines are present, this applies to the Resent-
6477 lines rather than the non-Resent- lines.
6479 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6480 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6482 58. Messages for configuration errors now include the name of the main
6483 configuration file - useful now that there may be more than one file in a
6484 list (.included file names were always shown).
6486 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6487 for those rare installations that do not start the daemon as root or run it
6488 setuid root. I've cut out the call to initgroups() if the daemon is not
6491 60. The Exim user and group can now be bound into the binary as text strings
6492 that are looked up at the start of Exim's processing.
6494 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6496 62. Added $mailstore_basename variable.
6498 63. Installed patch to sieve.c from Michael Haardt.
6500 64. When Exim failed to open the panic log after failing to open the main log,
6501 the original message it was trying to log was written to stderr and debug
6502 output, but if they were not available (the usual case in production), it
6503 was lost. Now it is written to syslog before the two lines that record the
6504 failures to open the logs.
6506 65. Users' Exim filters run in subprocesses under the user's uid. It is
6507 possible for a "deliver" command or an alias in a "personal" command to
6508 provoke an address rewrite. If logging of address rewriting is configured,
6509 this fails because the process is not running as root or exim. There may be
6510 a better way of dealing with this, but for the moment (because 4.30 needs
6511 to be released), I have disabled address rewrite logging when running a
6512 filter in a non-root, non-exim process.
6518 1. The buildconfig auxiliary program wasn't quoting the value set for
6519 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6520 not defined. This bug was masked in 4.22 by the effect that was fixed in
6523 2. Some messages that were rejected after a message id was allocated were
6524 shown as "incomplete" by exigrep. It no longer does this for messages that
6525 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6527 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6528 have allow_domain_literals set, the ID did not get logged in the <= line.
6529 Domain literals are now always recognized in Message-ID: header lines.
6531 4. The first argument for a ${extract expansion item is the key name or field
6532 number. Leading and trailing spaces in this item were not being ignored,
6533 causing some misleading effects.
6535 5. When deliver_drop_privilege was set, single queue runner processes started
6536 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6537 same command in the process it spins off) were not dropping privilege.
6539 6. When the daemon running as "exim" started a queue runner, it always
6540 re-executed Exim in the spun-off process. This is a waste of effort when
6541 deliver_drop_privilege is set. The new process now just calls the
6542 queue-runner function directly.
6548 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6551 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6552 ignored. Though the use of -oP was forcing the writing of a pid file, it
6553 was always written to the default place.
6555 3. If the message "no IP address found for host xxxx" is generated during
6556 incoming verification, it is now followed by identification of the incoming
6557 connection (so you can more easily find what provoked it).
6559 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6561 5. Added some features to "harden" Exim a bit more against certain attacks:
6563 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6564 be put in Local/Makefile. This is like the never_users runtime option,
6565 but it cannot be overridden. The default setting is "root".
6567 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6568 prefix string with which any file named in a -C command line option
6571 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6572 is retained for -C and -D only if the caller of Exim is root. Without
6573 it, the exim user may also use -C and -D and retain privilege.
6575 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6576 command line option is disabled.
6578 6. Macro names set by the -D option must start with an upper case letter, just
6579 like macro names defined in the configuration file.
6581 7. Added "dereference=" facility to LDAP.
6583 8. Two instances of the typo "uknown" in the source files are fixed.
6585 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6586 the Configure-Makefile script screwed up while processing it.
6588 10. Incorporated PCRE 4.4.
6590 11. The SMTP synchronization check was not operating right at the start of an
6591 SMTP session. For example, it could not catch a HELO sent before the client
6592 waited for the greeting. There is now a check for outstanding input at the
6593 point when the greeting is written. Because of the duplex, asynchronous
6594 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6595 way, but not yet received, when the check is performed.
6597 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6598 on TCP/IP sockets, because this apparently causes some broken clients to
6601 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6602 unchanged) from the Cygwin maintainer.
6604 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6605 was supported instead of testing for mbx. Effectively a typo.
6607 15. The spa authenticator server code was not checking that the input it
6608 received was valid base64.
6610 16. The debug output line for the "set" modifier in ACLs was not showing the
6611 name of the variable that was being set.
6613 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6615 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6616 lookup on its own. Something else was needed to trigger the lookup. For
6617 example, a match in host_lookup or the need for a host name in a host list.
6618 Now, if $sender_host_name is referenced and the host name has not yet been
6619 looked up, a lookup is performed. If the lookup fails, the variable remains
6620 empty, and $host_lookup_failed is set to "1".
6622 19. Added "eqi" as a case-independent comparison operator.
6624 20. The saslauthd authentication condition could segfault if neither service
6625 nor realm was specified.
6627 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6628 error message that was logged was misleading, and incoming SMTP
6629 connections were dropped. The message is now more accurate, and temporary
6630 errors are given to SMTP connections.
6632 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6633 (except RSET) with a 421 error, until QUIT is received. However, it was
6634 failing to send a response to QUIT.
6636 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6637 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6638 ACL is now run after the helo_try_verify_hosts code.
6640 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6641 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6642 case-independent, so other case variants are also recognized). Apparently
6643 some systems use these upper case variants.
6645 25. If more than two messages were waiting for the same host, and a transport
6646 filter was specified for the transport, Exim sent two messages over the
6647 same TCP/IP connection, and then failed with "socket operation on non-
6648 socket" when it tried to send the third.
6650 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6652 27. The extern definition of crypt16() in expand.c was not being excluded when
6653 the OS had its own crypt16() function.
6655 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6656 as a preferred synonym for return_size_limit, both as an option and as an
6659 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6661 30. Changed the default configuration ACL to relax the local part checking rule
6662 for addresses that are not in any local domains. For these addresses,
6663 slashes and pipe symbols are allowed within local parts, but the sequence
6664 /../ is explicitly forbidden.
6666 31. SPA server authentication was not clearing the challenge buffer before
6669 32. log_message in a "warn" ACL statement was writing to the reject log as
6670 well as to the main log, which contradicts the documentation and doesn't
6671 seem right (because no rejection is happening). So I have stopped it.
6673 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6674 However, I am unable to do any testing of this.
6676 34. Fixed an infelicity in the appendfile transport. When checking directories
6677 for a mailbox, to see if any needed to be created, it was accidentally
6678 using path names with one or more superfluous leading slashes; tracing
6679 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6681 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6682 contents are added to the log line that is written for every discarded
6683 recipient. (Previously a log_message setting was ignored.)
6685 36. The ${quote: operator now quotes the string if it is empty.
6687 37. The install script runs exim in order to find its version number. If for
6688 some reason other than non-existence or emptiness, which it checks, it
6689 could not run './exim', it was installing it with an empty version number,
6690 i.e. as "exim-". This error state is now caught, and the installation is
6693 38. An argument was missing from the function that creates an error message
6694 when Exim fails to connect to the socket for saslauthd authentication.
6695 This could cause Exim to crash, or give a corrupted message.
6697 39. Added isip, isip4, and isip6 to ${if conditions.
6699 40. The ACL variables $acl_xx are now saved with the message, and can be
6700 accessed later in routers, transports, and filters.
6702 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6703 strings in the file are not string-expanded.
6705 42. If a MAIL command specified a SIZE value that was too large to fit into an
6706 int variable, the check against message_size_limit failed. Such values are
6707 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6708 day this will have to be increased, but I don't think I want to be around
6709 when emails are that large.
6716 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6717 iconv() is not standard in FreeBSD.
6719 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6720 IPv6 enabled. The observed symptom was a segmentation fault on return from
6721 the function os_common_find_running_interfaces() in src/os.c.
6723 3. In the check_special_case() function in daemon.c I had used "errno" as an
6724 argument name, which causes warnings on some systems. This was basically a
6725 typo, since it was named "eno" in the comments!
6727 4. The code that waits for the clock to tick (at a resolution of some fraction
6728 of a second) so as to ensure message-id uniqueness was always waiting for
6729 at least one whole tick, when it could have waited for less. [This is
6730 almost certainly not relevant at current processor speeds, where it is
6731 unlikely to ever wait at all. But we try to future-proof.]
6733 5. The function that sleeps for a time interval that includes fractions of a
6734 second contained a race. It did not block SIGALRM between setting the
6735 timer, and suspending (a couple of lines later). If the interval was short
6736 and the sigsuspend() was delayed until after it had expired, the suspension
6737 never ended. On busy systems this could lead to processes getting stuck for
6740 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6741 process, before it forks any delivery processes. The open lookup caching
6742 mechanism meant that the open file or database connection was passed into
6743 the delivery process. The problem was that delivery processes always tidy
6744 up cached lookup data. This could cause a problem for the next delivery
6745 process started by the queue runner, because the external queue runner
6746 process does not know about the closure. So the next delivery process
6747 still has data in the lookup cache. In the case of a file lookup, there was
6748 no problem because closing a file descriptor in a subprocess doesn't affect
6749 the parent. However, if the lookup was caching a connection to a database,
6750 the connection was closed, and the second delivery process was likely to
6751 see errors such as "PGSQL: query failed: server closed the connection
6752 unexpectedly". The problem has been fixed by closing all cached lookups
6753 in a queue runner before running a delivery process.
6755 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6756 seem to have the "const" qualifier which it has on other OS. I've
6759 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6760 *of the same type* (client or server) with the same public name that an
6761 error should be diagnosed.
6763 9. When Exim looked up a host name for an IP address, but failed to find the
6764 original IP address when looking up the host name (a safety check), it
6765 output the message "<ip address> does not match any IP for NULL", which was
6766 confusing, to say the least. The bug was that the host name should have
6767 appeared instead of "NULL".
6769 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6770 user that is built into the binary, and the -C or -D options is used, root
6771 privilege is dropped before the configuration file is read. In addition,
6772 logging is switched to stderr instead of the normal log files. If the
6773 configuration then re-defines the Exim user, the unprivileged environment
6774 is probably not what is expected, so Exim logs a panic warning message (but
6777 However, if deliver_drop_privilege is set, the unprivileged state may well
6778 be exactly what is intended, so the warning has been cut out in that case,
6779 and Exim is allowed to try to write to its normal log files.
6785 1. smtp_return_error_details was not giving details for temporary sender
6786 or receiver verification errors.
6788 2. Diagnose a configuration error if two authenticators have the same public
6791 3. Exim used not to create the message log file for a message until the first
6792 delivery attempt. This could be confusing when incoming messages were held
6793 for policy or load reasons. The message log file is now created at the time
6794 the message is received, and an initial "Received" line is written to it.
6796 4. The automatically generated man page for command line options had a minor
6797 bug that caused no ill effects; however, a more serious problem was that
6798 the procedure for building the man page automatically didn't always
6799 operate. Consequently, release 4.20 contains an out-of-date version. This
6800 shouldn't happen again.
6802 5. When building Exim with embedded Perl support, the script that builds the
6803 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6804 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6806 6. The freeze_tell option was not being used for messages that were frozen on
6807 arrival, either by an ACL or by local_scan().
6809 7. Added the smtp_incomplete_transaction log selector.
6811 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6812 was accepting AUTH without a new EHLO.
6814 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6815 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6818 10. Logging of TCP/IP connections (when configured) now happens in the main
6819 daemon process instead of the child process, so that the TCP/IP connection
6820 count is more accurate (but it can never be perfect).
6822 11. The use of "drop" in a nested ACL was not being handled correctly in the
6823 outer ACL. Now, if condition failure induced by the nested "drop" causes
6824 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6825 or "require"), the connection is dropped.
6827 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6828 that yield "discard" can now be used with an "accept" or a "discard" verb,
6829 but an error is generated for any others (because I can't see a useful way
6830 to define what should happen).
6832 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6833 are now processed in the same way as lines in the Exim configuration file.
6834 In particular, continuation lines are supported.
6836 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6838 15. Added -ti meaning -t -i.
6840 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6841 domains, and warn by logging if others are found.
6843 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6844 returned from GIFCONF when Exim is trying to find the list of interfaces on
6845 a host. The code in os.c has been modified to copy each ifreq to an aligned
6846 structure in all cases.
6848 Also, in some cases, the returned ifreq's were being copied to a 'struct
6849 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6850 means the last couple of bytes of an IPv6 address could be chopped if the
6851 ifreq contained only a normal sockaddr (14 bytes storage).
6853 18. Named domain lists were not supported in the hosts_treat_as_local option.
6854 An entry such as +xxxx was not recognized, and was treated as a literal
6857 19. Ensure that header lines added by a DATA ACL are included in the reject log
6858 if the ACL subsequently rejects the message.
6860 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6861 MD5 (which is deprecated).
6863 21. When testing a filter file using -bf, Exim was writing a message when it
6864 took the sender from a "From " line in the message, but it was not doing so
6865 when it took $return_path from a Return-Path: header line. It now does.
6867 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6868 with a valid header line field name (a series of printing characters
6869 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6871 23. Changed "disc" in the source to "disk" to conform to the documentation and
6872 the book and for uniformity.
6874 24. Ignore Sendmail's -Ooption=value command line item.
6876 25. When execve() failed while trying to run a command in a pipe transport,
6877 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6878 could be confused with a return value of 69 from the command itself. This
6879 has been changed to 127, the value the shell returns if it is asked to run
6880 a non-existent command. The wording for the related log line suggests a
6881 non-existent command as the problem.
6883 26. If received_header_text expands to an empty string, do not add a Received:
6884 header line to the message. (Well, it adds a token one on the spool, but
6885 marks it "old" so that it doesn't get used or transmitted.)
6887 27. Installed eximstats 1.28 (addition of -nt option).
6889 28. There was no check for failure on the call to getsockname() in the daemon
6890 code. This can fail if there is a shortage of resources on the system, with
6891 ENOMEM, for example. A temporary error is now given on failure.
6893 29. Contrary to the C standard, it seems that in some environments, the
6894 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6895 program. Exim now does this explicitly; it affects the formatting of
6896 timestamps using strftime().
6898 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6899 complaints. I've now initialized all the variables, to avoid this.
6901 32. Header lines added by a system filter were not being "seen" during
6902 transport-time rewrites.
6904 33. The info_callback() function passed to OpenSSL is set up with type void
6905 (*)(SSL *, int, int), as described somewhere. However, when calling the
6906 function (actually a macro) that sets it up, the type void(*)() is
6907 expected. I've put in a cast to prevent warnings from picky compilers.
6909 34. If a DNS black list lookup found a CNAME record, but there were no A
6910 records associated with the domain it pointed at, Exim crashed.
6912 35. If a DNS black list lookup returned more than one A record, Exim ignored
6913 all but the first. It now scans all returned addresses if a particular IP
6914 value is being sought. In this situation, the contents of the
6915 $dnslist_value variable are a list of all the addresses, separated by a
6918 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6919 to accept a host name and all its aliases if the forward lookup for any of
6920 them yielded the IP address of the incoming connection. Now it accepts only
6921 those names whose forward lookup yields the correct IP address. Any other
6922 names are discarded. This closes a loophole whereby a rogue DNS
6923 administrator could create reverse DNS records to break through a
6924 wildcarded host restriction in an ACL.
6926 37. If a user filter or a system filter that ran in a subprocess used any of
6927 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6928 the wrong values were passed to the pipe command ($thisaddress had the
6929 value of $0, $0 had the value of $1, etc). This bug was introduced by
6930 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6932 38. Improved the line breaking for long SMTP error messages from ACLs.
6933 Previously, if there was no break point between 40 and 75 characters, Exim
6934 left the rest of the message alone. Two changes have been made: (a) I've
6935 reduced the minimum length to 35 characters; (b) if it can't find a break
6936 point between 35 and 75 characters, it looks ahead and uses the first one
6937 that it finds. This may give the occasional overlong line, but at least the
6938 remaining text gets split now.
6940 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6941 file descriptors might be low, and that setting 1000 would always raise it.
6942 It turns out that in some environments, the limit is already over 1000 and
6943 that lowering it causes trouble. So now Exim takes care not to decrease it.
6945 40. When delivering a message, the value of $return_path is set to $sender_
6946 address at the start of routing (routers may change the value). By an
6947 oversight, this default was not being set up when an address was tested by
6948 -bt or -bv, which affected the outcome if any router or filter referred to
6951 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6952 log only when "message" or "log_message" are set. However, if one of the
6953 conditions was an address verification, or a call to a nested ACL, the
6954 messages generated by the underlying test were being passed through. This
6955 no longer happens. The underlying message is available in $acl_verify_
6956 message for both "message" and "log_message" expansions, so it can be
6957 passed through if needed.
6959 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6960 new expansion $bh_ to give the encoded byte string without charset
6961 translation. Translation happens only if iconv() is available; HAVE_ICONV
6962 indicates this at build time. HEADERS_CHARSET gives the charset to
6963 translate to; headers_charset can change it in the configuration, and
6964 "headers charset" can change it in an individual filter file.
6966 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6967 that creates RFC 2047 encoded "words" labels them as that charset instead
6968 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6969 expansion operator; (ii) when Exim creates a From: line for a local
6970 message; (iii) when a header line is rewritten to include a "phrase" part.
6972 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6973 buggy, causing it to skip the first lines of messages whose message ID
6974 ended in 'D'. This would not have bitten before Exim release 4.14, because
6975 message IDs were unlikely to end in 'D' before then. The effect was to have
6976 incorrect size information for certain domains.
6978 45. #include "config.h" was missing at the start of the crypt16.c module. This
6979 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6982 46. If there was a timeout during a "random" callout check, Exim treated it as
6983 a failure of the random address, and carried on sending RSET and the real
6984 address. If the delay was just some slowness somewhere, the response to the
6985 original RCPT would be taken as a response to RSET and so on, causing
6986 mayhem of various kinds.
6988 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6989 when I implemented it. It didn't allow for the fact that some option values
6990 may legitimately be negative (e.g. size_addition), and it didn't even do
6991 the right test for positive values.
6993 48. Domain names in DNS records are case-independent. Exim always looks them up
6994 in lower case. Some resolvers return domain names in exactly the case they
6995 appear in the zone file, that is, they may contain uppercase letters. Not
6996 all resolvers do this - some return always lower case. Exim was treating a
6997 change of case by a resolver as a change of domain, similar to a widening
6998 of a domain abbreviation. This triggered its re-routing code and so it was
6999 trying to route what was effectively the same domain again. This normally
7000 caused routing to fail (because the router wouldn't handle the domain
7001 twice). Now Exim checks for this case specially, and just changes the
7002 casing of the domain that it ultimately uses when it transmits the message
7005 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
7008 50. If a filter generated a file delivery with a non-absolute name (possible if
7009 no home directory exists for the router), the forbid_file option was not
7012 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
7013 the existing equality matching.
7015 52. Exim was using ints instead of ino_t variables in some places where it was
7016 dealing with inode numbers.
7018 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
7019 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
7020 called TMPDIR, and if it finds it is different, it changes its value.
7022 54. The smtp_printf() function is now made available to local_scan() so
7023 additional output lines can be written before returning. There is also an
7024 smtp_fflush() function to enable the detection of a dropped connection.
7025 The variables smtp_input and smtp_batched_input are exported to
7028 55. Changed the default runtime configuration: the message "Unknown user"
7029 has been removed from the ACL, and instead placed on the localuser router,
7030 using the cannot_route_message feature. This means that any verification
7031 failures that generate their own messages won't get overridden. Similarly,
7032 the "Unrouteable address" message that was in the ACL for unverifiable
7033 relay addresses has also been removed.
7035 56. Added hosts_avoid_esmtp to the smtp transport.
7037 57. The exicyclog script was not checking for the esoteric option
7038 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7039 will work only if exicyclog is run under the appropriate euid.
7041 58. Following a discussion on the list, the rules by which Exim recognises line
7042 endings on incoming messages have been changed. The -dropcr and drop_cr
7043 options are now no-ops, retained only for backwards compatibility. The
7044 following line terminators are recognized: LF CRLF CR. However, special
7045 processing applies to CR:
7047 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7048 nor a local message in the state where . is a terminator.
7050 (ii) If a bare CR is encountered in a header line, an extra space is added
7051 after the line terminator so as not to end the header. The reasoning
7052 behind this is that bare CRs in header lines are most likely either
7053 to be mistakes, or people trying to play silly games.
7055 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7056 was being incorrectly given as 18 bytes larger than it should have been.
7057 This is a VOB (very old bug).
7059 60. This may never have affected anything current, but just in case it has:
7060 When the local host is found other than at the start of a list of hosts,
7061 the local host, those with the same MX, and any that follow, are discarded.
7062 When the list in question was part of a longer list of hosts, the following
7063 hosts (not currently being processed) were also being discarded. This no
7064 longer happens. I'm not sure if this situation could ever has previously
7067 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7070 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7071 header. This is something that is recommended in a new Internet Draft, and
7072 is something that is documented as being done by Sendmail. There are two
7073 possible values. For messages generated by the autoreply transport, Exim
7076 Auto-Submitted: auto-replied
7078 whereas for all other generated messages (e.g. bounces) it adds
7080 Auto-Submitted: auto-generated
7082 63. The "personal" condition in filters now includes a test for the
7083 Auto-Submitted: header. If it contains the string "auto-" the message it
7084 not considered personal.
7086 64. Added rcpt_include_affixes as a generic transport option.
7088 65. Added queue_only_override (default true).
7090 66. Added the syslog_duplication option.
7092 67. If what should have been the first header line of a message consisted of
7093 a space followed by a colon, Exim was mis-interpreting it as a header line.
7094 It isn't of course - it is syntactically invalid and should therefore be
7095 treated as the start of the message body. The misbehaviour could have
7096 caused a number of strange effects, including loss of data in subsequent
7097 header lines, and spool format errors.
7099 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7100 client host had authenticated. This control can now be exercised by an ACL
7101 for more flexibility.
7103 69. By default, callouts do not happen when testing with -bh. There is now a
7104 variant, -bhc, which does actually run the callout code, including
7105 consulting and updating the callout cache.
7107 70. Added support for saslauthd authentication, courtesy of Alexander
7110 71. If statvfs() failed on the spool or log directories while checking their
7111 size for availability, Exim confusingly gave the error "space shortage".
7112 Furthermore, in debugging mode it crashed with a floating point exception.
7113 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7114 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7115 serious problem, Exim now writes to the main and panic logs when this
7116 happens, with details of the failure. It then refuses to accept the
7117 incoming message, giving the message "spool directory problem" or "log
7118 directory problem" with a 421 code for SMTP messages.
7120 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7121 0, 1, and 2 exist, because some OS complain for execs without them (see
7122 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7123 descriptors. However, the code omitted to check that the open succeeded,
7124 causing mysterious errors if for some reason the permissions on /dev/null
7125 got screwed. Now Exim writes a message to the main and panic logs, and
7126 bombs out if it can't open /dev/null.
7128 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7129 interact so that it is all more flexible. It is supposed to remain
7130 backwards compatible. Also added extra_local_interfaces.
7132 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7133 to bomb out with an assertion failure - to the client this appears as a
7134 connection drop. This problem occurs in the part of the code that was taken
7135 from the Samba project. Fortunately, the assertion is in a very simple
7136 function, so I have fixed this by reproducing the function inline in the
7137 one place where it is called, and arranging for authentication to fail
7138 instead of killing the process with assert().
7140 75. The SPA client code was not working when the server requested OEM rather
7141 than Unicode encoding.
7143 76. Added code to make require_files with a specific uid setting more usable in
7144 the case where statting the file as root fails - usually a non-root-mounted
7145 NFS file system. When this happens and the failure is EACCES, Exim now
7146 forks a subprocess and does the per-uid checking as the relevant uid.
7148 77. Added process_log_path.
7150 78. If log_file_path was not explicitly set, a setting of check_log_space or
7151 check_log_inodes was ignored.
7153 79. If a space check for the spool or log partitions fails, the incident is now
7154 logged. Of course, in the latter case the data may get lost...
7156 80. Added the %p formatting code to string_format() so that it can be used to
7157 print addresses in debug_print(). Adjusted all the address printing in the
7158 debugging in store.c to use %p rather than %d.
7160 81. There was a concern that a line of code in smtp_in.c could overflow a
7161 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7162 initially expressed, the concern was not well-founded, because trailing
7163 spaces are removed early. However, if the trailing spaces were followed by
7164 a NULL, they did not get removed, so the overflow was possible. Two fixes
7167 (a) I re-wrote the offending code in a cleaner fashion.
7168 (b) If an incoming SMTP command contains a NULL character, it is rejected
7171 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7172 runs initgroups(), so that if the Exim user is in any additional groups,
7173 they will be used during message reception.
7179 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob