1 Change log file for Exim from version 4.21
2 ------------------------------------------
3 This document describes *changes* to previous versions, that might
4 affect Exim's operation, with an unchanged configuration file. For new
5 options, and new features, see the NewStuff file next to this ChangeLog.
11 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
12 When redis_servers is set to a list of > 1 element, and the Redis servers
13 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
14 case of MOVED into a DEFER case instead, thus moving the query onto the
15 next server in the list. For a cluster of N elements, all N servers must
16 be defined in redis_servers.
18 GF/02 Catch and remove uninitialized value warning in exiqsumm
19 Check for existence of @ARGV before looking at $ARGV[0]
21 JH/01 Replace the store_release() internal interface with store_newblock(),
22 which internalises the check required to safely use the old one, plus
23 the allocate and data copy operations duplicated in both (!) of the
26 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
27 modifier. This matches the restriction on the commandline.
29 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
30 Previously only the last row was returned.
32 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
33 we assumed that tags in the header were well-formed, and parsed the
34 element content after inspecting only the first char of the tag.
35 Assumptions at that stage could crash the receive process on malformed
38 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
39 While running the DKIM ACL we operate on the Permanent memory pool so that
40 variables created with "set" persist to the DATA ACL. Also (at any time)
41 DNS lookups that fail create cache records using the Permanent pool. But
42 expansions release any allocations made on the current pool - so a dnsdb
43 lookup expansion done in the DKIM ACL releases the memory used for the
44 DNS negative-cache, and bad things result. Solution is to switch to the
45 Main pool for expansions.
46 While we're in that code, add checks on the DNS cache during store_reset,
47 active in the testsuite.
48 Problem spotted, and debugging aided, by Wolfgang Breyha.
50 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
51 When none of the hosts presented to a transport match an already-open
52 connection, close it and proceed with the list. Previously we would
53 queue the message. Spotted by Lena with Yahoo, probably involving
56 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
57 Previously a spurious "250 OK id=" response was appended to the proper
60 JH/08 The "support for" informational output now, which built with Content
61 Scanning support, has a line for the malware scanner interfaces compiled
62 in. Interface can be individually included or not at build time.
64 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
65 by the template makefile "src/EDITME". The "STREAM" support for an older
66 ClamAV interface method is removed.
68 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
69 rows affected is given instead).
71 JH/11 The runtime Berkeley DB library version is now additionally output by
72 "exim -d -bV". Previously only the compile-time version was shown.
74 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
75 SMTP connection. Previously, when one had more receipients than the
76 first, an abortive onward connection was made. Move to full support for
77 multiple onward connections in sequence, handling cutthrough connection
78 for all multi-message initiating connections.
80 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
81 routers. Previously, a multi-recipient message would fail to match the
82 onward-connection opened for the first recipient, and cause its closure.
84 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
85 a timeout on read on a GnuTLS initiating connection, resulting in the
86 initiating connection being dropped. This mattered most when the callout
87 was marked defer_ok. Fix to keep the two timeout-detection methods
90 JH/15 Relax results from ACL control request to enable cutthrough, in
91 unsupported situations, from error to silently (except under debug)
92 ignoring. This covers use with PRDR, frozen messages, queue-only and
95 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
97 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
98 metadata, resulting in a crash in free().
100 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
101 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
102 Broken also in d185889f4, with init system revamp.
104 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
105 Previously we abruptly closed the connection after reading a malware-
106 found indication; now we go on to read the "scan ok" response line,
109 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
110 ACL. Previously, a crash would result.
112 JH/19 Speed up macro lookups during configuration file read, by skipping non-
113 macro text after a replacement (previously it was only once per line) and
114 by skipping builtin macros when searching for an uppercase lead character.
116 JH/20 DANE support moved from Experimental to mainline. The Makefile control
117 for the build is renamed.
119 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
120 was allocated for every new TLS startup, meaning one per message. Fix
121 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
123 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
124 reported the original. Fix to report (as far as possible) the ACL
125 result replacing the original.
127 JH/23 Fix memory leak during multi-message connections using STARTTLS under
128 OpenSSL. Certificate information is loaded for every new TLS startup,
129 and the resources needed to be freed.
131 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
133 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
136 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
137 DATA response info to the (existing) per-recipient response info for
138 the "C=" log element. It can have useful tracking info from the
139 destination system. Patch from Simon Arlott.
141 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
142 length value. Previously this would segfault.
144 HS/02 Support Avast multiline protoocol, this allows passing flags to
145 newer versions of the scanner.
147 JH/28 Ensure that variables possibly set during message acceptance are marked
148 dead before release of memory in the daemon loop. This stops complaints
149 about them when the debug_store option is enabled. Discovered specifically
150 for sender_rate_period, but applies to a whole set of variables.
151 Do the same for the queue-runner and queue-list loops, for variables set
152 from spool message files. Do the same for the SMTP per-message loop, for
153 certain variables indirectly set in ACL operations.
155 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
156 as a multi-recipient message from a mailinglist manager). The coding had
157 an arbitrary cutoff number of characters while checking for more input;
158 enforced by writing a NUL into the buffer. This corrupted long / fast
159 input. The problem was exposed more widely when more pipelineing of SMTP
160 responses was introduced, and one Exim system was feeding another.
161 The symptom is log complaints of SMTP syntax error (NUL chars) on the
162 receiving system, and refused recipients seen by the sending system
163 (propating to people being dropped from mailing lists).
164 Discovered and pinpointed by David Carter.
166 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
167 replaced by the ${authresults } expansion.
169 JH/31 Bug 2257: Fix pipe transport to not use a socket-only syscall.
171 HS/03 Set a handler for SIGTERM and call exit(3) if running as PID 1. This
172 allows proper process termination in container environments.
174 JH/32 Bug 2258: Fix spool_wireformat in combination with LMTP transport.
175 Previously the "final dot" had a newline after it; ensure it is CR,LF.
177 JH/33 SPF: remove support for the "spf" ACL condition outcome values "err_temp"
178 and "err_perm", deprecated since 4.83 when the RFC-defined words
179 "temperror" and "permerror" were introduced.
181 JH/34 Re-introduce enforcement of no cutthrough delivery on transports having
182 transport-filters or DKIM-signing. The restriction was lost in the
183 consolidation of verify-callout and delivery SMTP handling.
184 Extend the restriction to also cover ARC-signing.
186 JH/35 Cutthrough: for a final-dot response timeout (and nonunderstood responses)
187 in defer=pass mode supply a 450 to the initiator. Previously the message
190 PP/02 DANE: add dane_require_tls_ciphers SMTP Transport option; if unset,
191 tls_require_ciphers is used as before.
193 HS/03 Malware Avast: Better match the Avast multiline protocol. Add
194 "pass_unscanned". Only tmpfails from the scanner are written to
195 the paniclog, as they may require admin intervention (permission
196 denied, license issues). Other scanner errors (like decompression
197 bombs) do not cause a paniclog entry.
199 JH/36 Fix reinitialisation of DKIM logging variable between messages.
200 Previously it was possible to log spurious information in receive log
203 JH/37 Bug 2255: Revert the disable of the OpenSSL session caching. This
204 triggered odd behaviour from Outlook Express clients.
206 PP/03 Add util/renew-opendmarc-tlds.sh script for safe renewal of public
209 JH/38 DKIM: accept Ed25519 pubkeys in SubjectPublicKeyInfo-wrapped form,
210 since the IETF WG has not yet settled on that versus the original
211 "bare" representation.
213 JH/39 Fix syslog logging for syslog_timestamp=no and log_selector +millisec.
214 Previously the millisecond value corrupted the output.
215 Fix also for syslog_pid=no and log_selector +pid, for which the pid
216 corrupted the output.
222 JH/01 Rework error string handling in TLS interface so that the caller in
223 more cases is responsible for logging. This permits library-sourced
224 string to be attached to addresses during delivery, and collapses
225 pairs of long lines into single ones.
227 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
228 during configuration. Wildcards are allowed and expanded.
230 JH/02 Rework error string handling in DKIM to pass more info back to callers.
231 This permits better logging.
233 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
234 do not close it down and have the child transport start it up again on
235 the passed-on TCP connection. Instead, proxy the child (and any
236 subsequent ones) for TLS via a unix-domain socket channel. Logging is
237 affected: the continued delivery log lines do not have any DNSSEC, TLS
238 Certificate or OCSP information. TLS cipher information is still logged.
240 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
241 identical IP addresses on different listening ports. Will also affect
244 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
245 add noisy ifdef guards to special-case this sillyness.
246 Patch from Bernd Kuhls.
248 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
249 than 255 are no longer allowed.
251 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
252 Disable the session-cache too, which might reduce our load. Since we
253 currrectly use a new context for every connection, both as server and
254 client, there is no benefit for these.
255 GnuTLS appears to not support tickets server-side by default (we don't
256 call gnutls_session_ticket_enable_server()) but client side is enabled
257 by default on recent versions (3.1.3 +) unless the PFS priority string
260 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
261 <https://reproducible-builds.org/specs/source-date-epoch/>.
263 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
264 the check for any unsuccessful recipients did not notice the limit, and
265 erroneously found still-pending ones.
267 JH/08 Pipeline CHUNKING command and data together, on kernels that support
268 MSG_MORE. Only in-clear (not on TLS connections).
270 JH/09 Avoid using a temporary file during transport using dkim. Unless a
271 transport-filter is involved we can buffer the headers in memory for
272 creating the signature, and read the spool data file once for the
273 signature and again for transmission.
275 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
276 4.77 as the kernel support then wasn't solid, having issues in 64bit
277 mode. Now, it's been long enough. Add support for FreeBSD also.
279 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
280 case where the routing stage had gathered several addresses to send to
281 a host before calling the transport for the first, we previously failed
282 to close down TLS in the old transport process before passing the TCP
283 connection to the new process. The new one sent a STARTTLS command
284 which naturally failed, giving a failed delivery and bloating the retry
285 database. Investigation and fix prototype from Wolfgang Breyha.
287 JH/12 Fix check on SMTP command input synchronisation. Previously there were
288 false-negatives in the check that the sender had not preempted a response
289 or prompt from Exim (running as a server), due to that code's lack of
290 awareness of the SMTP input buffering.
292 PP/04 Add commandline_checks_require_admin option.
293 Exim drops privileges sanely, various checks such as -be aren't a
294 security problem, as long as you trust local users with access to their
295 own account. When invoked by services which pass untrusted data to
296 Exim, this might be an issue. Set this option in main configuration
297 AND make fixes to the calling application, such as using `--` to stop
300 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
301 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
302 responses to those, into a single TLS record each way (this usually means
303 a single packet). As a side issue, smtp_enforce_sync now works on TLS
306 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
307 affects you only if you're dancing at the edge of the param size limits.
308 If you are, and this message makes sense to you, then: raise the
309 configured limit or use OpenSSL 1.1. Nothing we can do for older
312 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
313 cmdline element to get the documented default one. Previously it was
316 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
317 are unsuable for non-message-specific reasons - eg. connection timeout,
318 banner-time rejection.
320 JH/16 Fix logging of delivery remote port, when specified by router, under
323 PP/06 Repair manualroute's ability to take options in any order, even if one
324 is the name of a transport.
327 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
329 JH/17 Change the list-building routines interface to use the expanding-string
330 triplet model, for better allocation and copying behaviour.
332 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
333 Previously it was constructed the first time a possibly-matching string
334 was met in the configuration file input during startup; now it is done
337 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
338 compatible one, to avoid the (poorly documented) possibility of a config
339 file in the working directory redirecting the DB files, possibly correpting
340 some existing file. CVE-2017-10140 assigned for BDB.
342 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
343 cache-hot. Previously, although the result was properly cached, the
344 initial verify call returned a defer.
346 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
347 the main verify for receipient in uncached-mode.
349 JH/22 Retire historical build files to an "unsupported" subdir. These are
350 defined as "ones for which we have no current evidence of testing".
352 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
353 if present. Previously it was ignored.
355 JH/24 Start using specified-initialisers in C structure init coding. This is
356 a C99 feature (it's 2017, so now considered safe).
358 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
359 if was a fixed-sized field and bitmask ops via macros; it is now more
362 PP/07 GitHub PR 56: Apply MariaDB build fix.
363 Patch provided by Jaroslav Škarvada.
365 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
366 during Coverity cleanups [4.87 JH/47]
367 Diagnosis and fix provided by Michael Fischer v. Mollard.
369 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
370 the right size to place the terminating semicolon on its own folded
371 line, the header hash was calculated to an incorrect value thanks to
372 the (relaxed) space the fold became.
374 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
375 and confused the parent.
377 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
378 which could crash as a result. This could lead to undeliverable messages.
380 JH/28 Logging: "next input sent too soon" now shows where input was truncated
383 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
384 matters on fast-turnover and PID-randomising systems, which were getting
385 out-of-order delivery.
387 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
388 a possibly-overlapping copy. The symptom was that "Remote host closed
389 connection in response to HELO" was logged instead of the actual 4xx
392 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
393 Previously only that bufferd was discarded, resulting in SYMTP command
396 JH/32 DKIM: when a message has multiple signatures matching an identity given
397 in dkim_verify_signers, run the dkim acl once for each. Previously only
398 one run was done. Bug 2189.
400 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
401 "panic the current process" to "deliberately defer". The panic log is
402 still written with the problem list name; the mail and reject logs now
403 get a temp-reject line for the message that was being handled, saying
404 something like "domains check lookup or other defer". The SMTP 451
405 message is still "Temporary local problem".
407 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
408 A crafted sequence of BDAT commands could result in in-use memory beeing
409 freed. CVE-2017-16943.
411 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
412 from SMTP input. Previously it was always done; now only done for DATA
413 and not BDAT commands. CVE-2017-16944.
415 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
416 to the message (such as an overlong header line). Previously this was
417 not done and we did not exit BDAT mode. Followon from the previous item
418 though a different problem.
424 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
425 than -2003 did; needs libidn2 in addition to libidn.
427 JH/02 The path option on a pipe transport is now expanded before use.
429 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
430 Patch provided by "Björn", documentation fix added too.
432 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
433 missing a wire-to-host endian conversion.
435 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
436 close after a BDAT command line could be taken as a following command,
437 giving a synch failure. Fix by only checking for synch immediately
438 before acknowledging the chunk.
440 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
441 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
442 macro. Patches provided by Josh Soref.
444 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
445 Previously we did not; the RFC seems ambiguous and VRFY is not listed
446 by IANA as a service extension. However, John Klensin suggests that we
449 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
450 the dkim code may be unix-mode line endings rather than smtp wire-format
451 CRLF, so prepend a CR to any bare LF.
453 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
454 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
456 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
457 the first were themselves being wrongly included in the feed into dkim
458 processing; with most chunk sizes in use this resulted in an incorrect
459 body hash calculated value.
461 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
462 DKIM signature block, for verification. Although advised against by
463 standards it is specifically not ruled illegal.
465 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
467 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
468 missing a body hash (the bh= tag).
470 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
471 It seems that HAProxy sends the Proxy Protocol information in clear and
472 only then does a TLS startup, so do the same.
474 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
475 TCP connections (such as for Spamd) unless the daemon successfully set
476 Fast Open mode on its listening sockets. This fixes breakage seen on
477 too-old kernels or those not configured for Fast Open, at the cost of
478 requiring both directions being enabled for TFO, and TFO never being used
479 by non-daemon-related Exim processes.
481 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
482 endings, at least on the first header line. Try to canonify any that get
483 past that check, despite the cost.
485 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
486 now limited to an arbitrary five deep, while parsing addresses with the
487 strip_excess_angle_brackets option enabled.
489 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
490 instead leave the unprompted TLS handshake in socket buffer for the
491 TLS library to consume.
493 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
495 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
497 JH/16 Drop variables when they go out of scope. Memory management drops a whole
498 region in one operation, for speed, and this leaves assigned pointers
499 dangling. Add checks run only under the testsuite which checks all
500 variables at a store-reset and panics on a dangling pointer; add code
501 explicitly nulling out all the variables discovered. Fixes one known
502 bug: a transport crash, where a dangling pointer for $sending_ip_address
503 originally assigned in a verify callout, is re-used.
505 PP/06 Drop '.' from @INC in various Perl scripts.
507 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
509 PP/08 Reduce a number of compilation warnings under clang; building with
510 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
511 should be warning-free.
513 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
515 HS/01 Fix portability problems introduced by PP/08 for platforms where
516 realloc(NULL) is not equivalent to malloc() [SunOS et al].
518 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
519 chunk. This allows us to accept broken chunked messages. We need a more
520 general solution here.
522 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
523 already-broken messages in the queue.
525 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
527 JH/19 Fix reference counting bug in routing-generated-address tracking.
533 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
534 supports it and a size is available (ie. the sending peer gave us one).
536 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
537 years of being deprecated). The replacements are the ACLs
538 acl_smtp_mime and acl_not_smtp_mime.
540 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
541 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
542 or even an in-clear connection were permitted. Now, if the host lookup
543 was dnssec and dane was requested then the host is only used if the
544 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
545 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
546 if one fails this test.
547 This means that a poorly-configured remote DNS will make it incommunicado;
548 but it protects against a DNS-interception attack on it.
550 JH/04 Bug 1810: make continued-use of an open smtp transport connection
551 non-noisy when a race steals the message being considered.
553 JH/05 If main configuration option tls_certificate is unset, generate a
554 self-signed certificate for inbound TLS connections.
556 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
557 in rewrites and routers.
559 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
560 and logged a warning sing 4.83; now they are a configuration file error.
562 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
563 (lacking @domain). Apply the same qualification processing as RCPT.
565 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
567 JH/10 Support ${sha256:} applied to a string (as well as the previous
570 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
571 a cutthrough deliver is pending, as we always want to make a connection.
572 This also avoids re-routing the message when later placing the cutthrough
573 connection after a verify cache hit.
574 Do not update it with the verify result either.
576 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
577 when routing results in more than one destination address.
579 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
580 signing (which inhibits the cutthrough capability). Previously only
581 the presence of an option was tested; now an expansion evaluating as
582 empty is permissible (obviously it should depend only on data available
583 when the cutthrough connection is made).
585 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
586 the relevant preceding SMTP command did not note the pipelining mode.
588 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
589 Previously they were not counted.
591 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
592 as one having no matching records. Previously we deferred the message
593 that needed the lookup.
595 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
596 distinguished as "(=".
598 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
599 for missing MX records. Previously it only worked for missing A records.
601 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
603 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
604 after the data-go-ahead and data-ack. Patch from Jason Betts.
606 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
607 even for a "none" policy. Patch from Tony Meyer.
609 JH/22 Fix continued use of a connection for further deliveries. If a port was
610 specified by a router, it must also match for the delivery to be
613 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
614 When one of the recipients of a message was unsuitable for the connection
615 (has no matching addresses), we lost track of needing to mark it
616 deferred. As a result mail would be lost.
618 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
620 JH/25 Decoding ACL controls is now done using a binary search; the source code
621 takes up less space and should be simpler to maintain. Merge the ACL
622 condition decode tables also, with similar effect.
624 JH/26 Fix problem with one_time used on a redirect router which returned the
625 parent address unchanged. A retry would see the parent address marked as
626 delivered, so not attempt the (identical) child. As a result mail would
629 JH/27 Fix a possible security hole, wherein a process operating with the Exim
630 UID can gain a root shell. Credit to http://www.halfdog.net/ for
631 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
634 JH/28 Enable {spool,log} filesystem space and inode checks as default.
635 Main config options check_{log,spool}_{inodes,space} are now
636 100 inodes, 10MB unless set otherwise in the configuration.
638 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
639 Previously it only applied to the main-section connection policy
642 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
644 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
645 by me. Added RFC7919 DH primes as an alternative.
647 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
648 are not in the system include path.
650 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
651 GnuTLS, when a session startup failed (eg because the client disconnected)
652 Exim did stdio operations after fclose. This was exposed by a recent
653 change which nulled out the file handle after the fclose.
655 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
656 signed directly by the cert-signing cert, rather than an intermediate
657 OCSP-signing cert. This is the model used by LetsEncrypt.
659 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
661 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
662 an incoming connection.
664 HS/02 Bug 1802: Do not half-close the connection after sending a request
667 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
668 fallback to "prime256v1".
670 JH/34 SECURITY: Use proper copy of DATA command in error message.
671 Could leak key material. Remotely exploitable. CVE-2016-9963.
677 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
678 and 3.4.4 - once the server is enabled to respond to an OCSP request
679 it does even when not requested, resulting in a stapling non-aware
680 client dropping the TLS connection.
682 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
683 support variable-length bit vectors. No functional change.
685 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
686 The I= interface field on outgoing lines is now after the H= remote
687 host field, same as incoming lines. There is a separate
688 outgoing_interface log selector which allows you to disable the
691 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
692 If not running log_selector +smtp_connection the mainlog would be held
693 open indefinitely after a "too many connections" event, including to a
694 deleted file after a log rotate. Leave the per net connection logging
695 leaving it open for efficiency as that will be quickly detected by the
696 check on the next write.
698 HS/01 Bug 1671: Fix post transport crash.
699 Processing the wait-<transport> messages could crash the delivery
700 process if the message IDs didn't exist for some reason. When
701 using 'split_spool_directory=yes' the construction of the spool
702 file name failed already, exposing the same netto behaviour.
704 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
705 mime_regex ACL conditions.
707 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
708 to DSN fail messages (bounces): remote IP, remote greeting, remote response
709 to HELO, local diagnostic string.
711 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
712 log line to debug. Even when configured with a tls authenticator many
713 client connections are expected to not authenticate in this way, so
714 an authenticate fail is not an error.
716 HS/02 Add the Exim version string to the process info. This way exiwhat
717 gives some more detail about the running daemon.
719 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
720 matter for fast-change records such as DNSBLs.
722 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
723 retry record, even if constant. There may be multiple transports with
724 different interface settings and the retry behaviour needs to be kept
727 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
729 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
731 JH/10 Bug 840: fix log_defer_output option of pipe transport
733 JH/11 Bug 830: use same host for all RCPTS of a message, even under
734 hosts_randomize. This matters a lot when combined with mua_wrapper.
736 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
737 ${quote_pgsql:<string>} operator.
739 JH/13 Bug 1708: avoid misaligned access in cached lookup.
741 JH/14 Change header file name for freeradius-client. Relevant if compiling
742 with Radius support; from the Gentoo tree and checked under Fedora.
744 JH/15 Bug 1712: Introduce $prdr_requested flag variable
746 JH/16 Bug 1714: Permit an empty string as expansion result for transport
747 option transport_filter, meaning no filtering.
749 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
751 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
752 defaults to "*" (all hosts). The variable is now available when not built
753 with TLS, default unset, mainly to enable keeping the testsuite sane.
754 If a server certificate is not supplied (via tls_certificate) an error is
755 logged, and clients will find TLS connections fail on startup. Presumably
756 they will retry in-clear.
757 Packagers of Exim are strongly encouraged to create a server certificate
758 at installation time.
760 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
761 with the $config_file variable.
763 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
764 in transport context, after the attempt, and per-recipient. The latter type
765 is per host attempted. The event data is the error message, and the errno
766 information encodes the lookup type (A vs. MX) used for the (first) host,
767 and the trailing two digits of the smtp 4xx response.
769 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
770 to write to mainlog (or rejectlog, paniclog) in the window between file
771 creation and permissions/ownership being changed. Particularly affects
772 installations where exicyclog is run as root, rather than exim user;
773 result is that the running daemon panics and dies.
775 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
777 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
778 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
779 "pri" and "weight". Note that the previous implicit priority given by the
780 list order is no longer honoured.
782 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
785 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
786 by defining SUPPORT_SOCKS.
788 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
789 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
790 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
791 variables are renamed to proxy_{local,external}_{address,port}.
793 JH/27 Move Internationalisation support from Experimental to mainline, enabled
794 for a build by defining SUPPORT_I18N
796 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
797 of the query string, and make ${quote_redis:} do that quoting.
799 JH/29 Move Events support from Experimental to mainline, enabled by default
800 and removable for a build by defining DISABLE_EVENT.
802 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
804 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
805 cached by the daemon.
807 JH/32 Move Redis support from Experimental to mainline, enabled for a build
808 by defining LOOKUP_REDIS. The libhiredis library is required.
810 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
811 keys are given for lookup.
813 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
814 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
815 only supported when built with TLS support. The PolarSSL SHA routines
816 are still used when the TLS library is too old for convenient support.
818 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
819 openssl_options), for security. OpenSSL forces this from version 1.1.0
820 server-side so match that on older versions.
822 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
823 allocation for $value could be released as the expansion processing
824 concluded, but leaving the global pointer active for it.
826 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
827 and to use the domains and local_parts ACL conditions.
829 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
830 incorrectly not doubled on cutthrough transmission, hence seen as a
831 body-termination at the receiving system - resulting in truncated mails.
832 Commonly the sender saw a TCP-level error, and retransmitted the message
833 via the normal store-and-forward channel. This could result in duplicates
834 received - but deduplicating mailstores were liable to retain only the
835 initial truncated version.
837 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
839 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
841 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
842 we're in there, support oversigning also; bug 1309.
844 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
846 HS/04 Add support for keep_environment and add_environment options.
848 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
849 either intentional arithmetic overflow during PRNG, or testing config-
852 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
853 delivery resulted in actual delivery. Cancel cutthrough before DATA
856 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
857 rejects a recipient: pass the reject to the originator.
859 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
860 Many were false-positives and ignorable, but it's worth fixing the
863 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
864 for the new environment-manipulation done at startup. Move the routines
865 from being local to tls.c to being global via the os.c file.
867 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
868 an extract embedded as result-arg for a map, the first arg for extract
869 is unavailable so we cannot tell if this is a numbered or keyed
870 extraction. Accept either.
876 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
879 JH/02 The smtp transport option "multi_domain" is now expanded.
881 JH/03 The smtp transport now requests PRDR by default, if the server offers
884 JH/04 Certificate name checking on server certificates, when exim is a client,
885 is now done by default. The transport option tls_verify_cert_hostnames
886 can be used to disable this per-host. The build option
887 EXPERIMENTAL_CERTNAMES is withdrawn.
889 JH/05 The value of the tls_verify_certificates smtp transport and main options
890 default to the word "system" to access the system default CA bundle.
891 For GnuTLS, only version 3.0.20 or later.
893 JH/06 Verification of the server certificate for a TLS connection is now tried
894 (but not required) by default. The verification status is now logged by
895 default, for both outbound TLS and client-certificate supplying inbound
898 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
901 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
902 Status Notification (bounce) messages are now MIME format per RFC 3464.
903 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
904 under the control of the dsn_advertise_hosts option, and routers may
905 have a dsn_lasthop option.
907 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
908 default, modifiable by a malware= option. The list separator for
909 the options can now be changed in the usual way. Bug 68.
911 JH/10 The smtp_receive_timeout main option is now expanded before use.
913 JH/11 The incoming_interface log option now also enables logging of the
914 local interface on delivery outgoing connections.
916 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
917 if the interface and destination host and port all match.
919 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
922 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
923 Patch from Andrew Lewis.
925 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
926 now supports optional time-restrictions, weighting, and priority
927 modifiers per server. Patch originally by <rommer@active.by>.
929 JH/16 The spamd_address main option now supports a mixed list of local
930 and remote servers. Remote servers can be IPv6 addresses, and
931 specify a port-range.
933 JH/17 Bug 68: The spamd_address main option now supports an optional
934 timeout value per server.
936 JH/18 Bug 1581: Router and transport options headers_add/remove can
937 now have the list separator specified.
939 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
942 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
945 JH/21 Support for the A6 type of dns record is withdrawn.
947 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
948 rather than the verbs used.
950 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
951 from 255 to 1024 chars.
953 JH/24 Verification callouts now attempt to use TLS by default.
955 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
956 are generic router options now. The defaults didn't change.
958 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
959 Original patch from Alexander Shikoff, worked over by JH.
961 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
962 files if ZCAT_COMMAND is not executable.
964 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
966 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
968 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
969 Normally benign, it bites when the pair was led to by a CNAME;
970 modern usage is to not canonicalize the domain to a CNAME target
971 (and we were inconsistent anyway for A-only vs AAAA+A).
973 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
975 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
976 when evaluating $sender_host_dnssec.
978 JH/31 Check the HELO verification lookup for DNSSEC, adding new
979 $sender_helo_dnssec variable.
981 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
983 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
985 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
987 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
988 documented as working, but never had. Support all but $spam_report.
990 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
991 added for tls authenticator.
993 HS/03 Add perl_taintmode main config option
999 TL/01 When running the test suite, the README says that variables such as
1000 no_msglog_check are global and can be placed anywhere in a specific
1001 test's script, however it was observed that placement needed to be near
1002 the beginning for it to behave that way. Changed the runtest perl
1003 script to read through the entire script once to detect and set these
1004 variables, reset to the beginning of the script, and then run through
1005 the script parsing/test process like normal.
1007 TL/02 The BSD's have an arc4random API. One of the functions to induce
1008 adding randomness was arc4random_stir(), but it has been removed in
1009 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
1010 function when detected.
1012 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
1013 cause callback expansion.
1015 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
1016 syntax errors in an expansion can be treated as a string instead of
1017 logging or causing an error, due to the internal use of bool_lax
1018 instead of bool when processing it.
1020 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
1021 server certificates when making smtp deliveries.
1023 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
1025 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
1027 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
1029 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
1030 Merged patch from Sebastian Wiedenroth.
1032 JH/05 Fix results-pipe from transport process. Several recipients, combined
1033 with certificate use, exposed issues where response data items split
1034 over buffer boundaries were not parsed properly. This eventually
1035 resulted in duplicates being sent. This issue only became common enough
1036 to notice due to the introduction of connection certificate information,
1037 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1039 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1040 size buffer was used, resulting in syntax errors when an expansion
1043 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1044 version 3.3.6 or later.
1046 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1047 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1048 both become "event_action", the variables become $event_name, $event_data
1049 and $event_defer_errno. There is a new variable $verify_mode, usable in
1050 routers, transports and related events. The tls:cert event is now also
1051 raised for inbound connections, if the main configuration event_action
1054 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1055 early OCSP support, but no stapling (appears to be less than 1.0.0).
1057 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1058 server certificate names available under the smtp transport option
1059 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1062 JH/10 Time-related extraction expansions from certificates now use the main
1063 option "timezone" setting for output formatting, and are consistent
1064 between OpenSSL and GnuTLS compilations. Bug 1541.
1066 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1067 encoded parameter in the incoming message. Bug 1558.
1069 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1070 include certificate info, eximon was claiming there were spoolfile
1073 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1075 JH/14 Log delivery-related information more consistently, using the sequence
1076 "H=<name> [<ip>]" wherever possible.
1078 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1079 are problematic for Debian distribution, omit them from the release
1082 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1084 JH/16 Fix string representation of time values on 64bit time_t architectures.
1087 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1088 output list separator was used.
1093 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1094 checkers that were complaining about end of non-void function with no
1097 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1098 This was a regression introduced in 4.83 by another bugfix.
1100 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1102 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1103 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1109 TF/01 Correctly close the server side of TLS when forking for delivery.
1111 When a message was received over SMTP with TLS, Exim failed to clear up
1112 the incoming connection properly after forking off the child process to
1113 deliver the message. In some situations the subsequent outgoing
1114 delivery connection happened to have the same fd number as the incoming
1115 connection previously had. Exim would try to use TLS and fail, logging
1116 a "Bad file descriptor" error.
1118 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1119 utilities have not been installed.
1121 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1122 temporary space as the ACL may create new global variables.
1124 TL/01 LDAP support uses per connection or global context settings, depending
1125 upon the detected version of the libraries at build time.
1127 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1128 to extract and use the src ip:port in logging and expansions as if it
1129 were a direct connection from the outside internet. PPv2 support was
1130 updated based on HAProxy spec change in May 2014.
1132 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1134 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1135 Properly escape header and check for NULL return.
1137 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1140 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1142 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1143 characters in header names, implemented as a verify condition.
1144 Contributed by Michael Fischer v. Mollard.
1146 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1147 results permerror and temperror. Previous values are deprecated but
1148 still accepted. In a future release, err_perm and err_temp will be
1149 completely removed, which will be a backward incompatibility if the
1150 ACL tests for either of these two old results. Patch contributed by
1151 user bes-internal on the mailing list.
1153 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1155 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1156 selectors, in both main and reject logs.
1158 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1161 JH/07 Add malware type "sock" for talking to simple daemon.
1163 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1165 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1166 routers/transports under cutthrough routing.
1168 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1169 numbers. Touch up "bool" conditional to keep the same definition.
1171 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1173 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1175 JH/12 Expand items in router/transport headers_add or headers_remove lists
1176 individually rather than the list as a whole. Bug 1452.
1178 Required for reasonable handling of multiple headers_ options when
1179 they may be empty; requires that headers_remove items with embedded
1180 colons must have them doubled (or the list-separator changed).
1182 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1183 view the policy declared in the DMARC record. Currently, $dmarc_status
1184 is a combined value of both the record presence and the result of the
1187 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1189 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1190 dnslookup router and the smtp transport (applying to the forward
1193 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1194 of ldap servers used for a specific lookup. Patch provided by Heiko
1197 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1198 New variable $lookup_dnssec_authenticated for observability.
1200 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1201 Patch submitted by Lars Timman.
1203 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1205 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1206 Requires trusted mode and valid format message id, aborts otherwise.
1207 Patch contributed by Heiko Schlichting.
1209 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1210 certextract with support for various fields. Bug 1358.
1212 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1213 is requested by default, modifiable by smtp transport option
1216 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1217 operate on certificate variables to give certificate fingerprints
1218 Also new ${sha256:cert_variable}.
1220 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1222 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1223 Christian Aistleitner.
1225 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1227 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1228 file. Patch from Wolfgang Breyha.
1230 JH/25 Expand the coverage of the delivery $host and $host_address to
1231 client authenticators run in verify callout. Bug 1476.
1233 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1234 align with daemon_smtp_ports. Bug 72.
1236 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1237 support and error reporting did not work properly.
1239 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1240 and is readable. Patch from Andrew Colin Kissa.
1242 TL/14 Enhance documentation of ${run expansion and how it parses the
1243 commandline after expansion, particularly in the case when an
1244 unquoted variable expansion results in an empty value.
1246 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1248 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1249 by renaming away from it. Addresses GH issue 15, reported by
1252 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1254 TL/15 SECURITY: prevent double expansion in math comparison functions
1255 (can expand unsanitized data). Not remotely exploitable.
1262 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1264 PP/02 Make -n do something, by making it not do something.
1265 When combined with -bP, the name of an option is not output.
1267 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1270 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1271 $sender_host_name and config options to manage this, and basic check
1274 PP/05 DSCP support for outbound connections and control modifier for inbound.
1276 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1277 (Only plugin which currently uses this is kerberos4, which nobody should
1278 be using, but we should make it available and other future plugins might
1279 conceivably use it, even though it would break NAT; stuff *should* be
1280 using channel bindings instead).
1282 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1283 name; added for Sendmail compatibility; requires admin caller.
1284 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1285 just ignore it); requires trusted caller.
1286 Also parse but ignore: -Ac -Am -X<logfile>
1289 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1291 TL/02 Add +smtp_confirmation as a default logging option.
1293 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1294 Patch by Magnus Holmgren from 2007-02-20.
1296 TL/04 Bugzilla 1281 - Spec typo.
1297 Bugzilla 1283 - Spec typo.
1298 Bugzilla 1290 - Spec grammar fixes.
1300 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1302 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1304 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1305 mailing list by by Dmitry Isaikin.
1307 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1309 JH/02 Support "G" suffix to numbers in ${if comparisons.
1311 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1313 NM/01 Bugzilla 1197 - Spec typo
1314 Bugzilla 1196 - Spec examples corrections
1316 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1318 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1319 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1322 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1323 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1325 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1326 "acl {{name}{arg}...}", and optional args on acl condition
1329 JH/05 Permit multiple router/transport headers_add/remove lines.
1331 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1333 JH/07 Avoid using a waiting database for a single-message-only transport.
1334 Performance patch from Paul Fisher. Bugzilla 1262.
1336 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1339 JH/09 Add $headers_added variable, with content from use of ACL modifier
1340 add_header (but not yet added to the message). Bugzilla 199.
1342 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1343 Pulled from Bugzilla 817 by Wolfgang Breyha.
1345 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1347 (nb: this is the same fix as in Exim 4.80.1)
1349 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1352 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1355 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1357 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1358 advertise SMTP AUTH mechanism to us, instead of a generic
1359 protocol violation error. Also, make Exim more robust to bad
1360 data from the Dovecot auth socket.
1362 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1364 When a queue runner is handling a message, Exim first routes the
1365 recipient addresses, during which it prunes them based on the retry
1366 hints database. After that it attempts to deliver the message to
1367 any remaining recipients. It then updates the hints database using
1370 So if a recipient address works intermittently, it can get repeatedly
1371 deferred at routing time. The retry hints record remains fresh so the
1372 address never reaches the final cutoff time.
1374 This is a fairly common occurrence when a user is bumping up against
1375 their storage quota. Exim had some logic in its local delivery code
1376 to deal with this. However it did not apply to per-recipient defers
1377 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1379 This change adds a proper retry rule check during routing so that the
1380 final cutoff time is checked against the message's age. We only do
1381 this check if there is an address retry record and there is not a
1382 domain retry record; this implies that previous attempts to handle
1383 the address had the retry_use_local_parts option turned on. We use
1384 this as an approximation for the destination being like a local
1385 delivery, as in LMTP.
1387 I suspect this new check makes the old local delivery cutoff check
1388 redundant, but I have not verified this so I left the code in place.
1390 TF/02 Correct gecos expansion when From: is a prefix of the username.
1392 Test 0254 submits a message to Exim with the header
1396 When I ran the test suite under the user fanf2, Exim expanded
1397 the header to contain my full name, whereas it should have added
1398 a Resent-Sender: header. It erroneously treats any prefix of the
1399 username as equal to the username.
1401 This change corrects that bug.
1403 GF/01 DCC debug and logging tidyup
1404 Error conditions log to paniclog rather than rejectlog.
1405 Debug lines prefixed by "DCC: " to remove any ambiguity.
1407 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1409 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1410 Bug spotted by Jeremy Harris; was flawed since initial commit.
1411 Would have resulted in OCSP responses post-SNI triggering an Exim
1412 NULL dereference and crash.
1414 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1416 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1417 Bug detection, analysis and fix by Samuel Thibault.
1418 Bugzilla 1331, Debian bug #698092.
1420 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1422 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1423 Server implementation by Todd Lyons, client by JH.
1424 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1425 config variable "prdr_enable" controls whether the server
1426 advertises the facility. If the client requests PRDR a new
1427 acl_data_smtp_prdr ACL is called once for each recipient, after
1428 the body content is received and before the acl_smtp_data ACL.
1429 The client is controlled by both of: a hosts_try_prdr option
1430 on the smtp transport, and the server advertisement.
1431 Default client logging of deliveries and rejections involving
1432 PRDR are flagged with the string "PRDR".
1434 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1435 fclose(). Diagnosis by Todd Lyons.
1437 PP/17 Update configure.default to handle IPv6 localhost better.
1438 Patch by Alain Williams (plus minor tweaks).
1441 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1442 This is now consistent with GnuTLS, and is now documented: the
1443 previous undocumented portable approach to treating the option as
1444 unset was to force an expansion failure. That still works, and
1445 an empty string is now equivalent.
1447 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1448 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1449 not performing validation itself.
1451 PP/20 Added force_command boolean option to pipe transport.
1452 Patch from Nick Koston, of cPanel Inc.
1454 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1457 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1459 PP/21 Fix eximon continuous updating with timestamped log-files.
1460 Broken in a format-string cleanup in 4.80, missed when I repaired the
1461 other false fix of the same issue.
1462 Report and fix from Heiko Schlichting.
1465 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1466 Report from Prashanth Katuri.
1468 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1469 It's SecureTransport, so affects any MacOS clients which use the
1470 system-integrated TLS libraries, including email clients.
1472 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1473 using a MIME ACL for non-SMTP local injection.
1474 Report and assistance in diagnosis by Warren Baker.
1476 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1478 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1480 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1481 last id that failed so it may be referenced in subsequent ACL's.
1483 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1486 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1487 ldap library initialization, allowing self-signed CA's to be
1488 used. Also properly sets require_cert option later in code by
1489 using NULL (global ldap config) instead of ldap handle (per
1490 session). Bug diagnosis and testing by alxgomz.
1492 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1493 the src/util/ subdirectory.
1495 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1496 renamed to Transport Post Delivery Action by Jeremy Harris, as
1499 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1500 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1501 redis_servers = needs to be configured which will be used by the redis
1502 lookup. Patch from Warren Baker, of The Packet Hub.
1504 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1506 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1507 hostname or reverse DNS when processing a host list. Used suggestions
1508 from multiple comments on this bug.
1510 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1512 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1513 Missed a few lines, added it to make the runtest require no keyboard
1516 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1517 contains upper case chars. Make router use caseful_local_part.
1519 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1520 support when GnuTLS has been built with p11-kit.
1526 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1528 This, or similar/improved, will also be change PP/11 of 4.82.
1534 PP/01 Handle short writes when writing local log-files.
1535 In practice, only affects FreeBSD (8 onwards).
1536 Bugzilla 1053, with thanks to Dmitry Isaikin.
1538 NM/01 Bugzilla 949 - Documentation tweak
1540 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1543 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1545 PP/02 Implemented gsasl authenticator.
1547 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1549 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1550 `pkg-config foo` for cflags/libs.
1552 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1553 with rest of GSASL and with heimdal_gssapi.
1555 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1556 `pkg-config foo` for cflags/libs for the TLS implementation.
1558 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1559 properties get this fed in as external SSF. A number of robustness
1560 and debugging improvements to the cyrus_sasl authenticator.
1562 PP/08 cyrus_sasl server now expands the server_realm option.
1564 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1565 Patch by Jeremy Harris.
1567 PP/10 Added dbmjz lookup type.
1569 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1571 PP/12 MAIL args handles TAB as well as SP, for better interop with
1572 non-compliant senders.
1573 Analysis and variant patch by Todd Lyons.
1575 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1576 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1577 Patch from Dirk Mueller <dmueller@suse.com>
1579 PP/13 tls_peerdn now print-escaped for spool files.
1580 Observed some $tls_peerdn in wild which contained \n, which resulted
1581 in spool file corruption.
1583 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1584 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1585 or write after TLS renegotiation, which otherwise led to messages
1588 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1589 as a tracking header (ie: a signed header comes before the signature).
1590 Patch from Wolfgang Breyha.
1592 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1593 comma-sep list; embedded commas doubled.
1595 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1597 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1599 Report and patch from Dmitry Banschikov.
1601 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1602 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1603 failures. We appear to now support TLS1.1+ with Exim.
1605 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1606 lets Exim select keys and certificates based upon TLS SNI from client.
1607 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1608 before an outbound SMTP session. New log_selector, +tls_sni.
1610 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1611 NULL dereference. Report and patch from Alun Jones.
1613 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1614 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1615 Not seeing resolver debug output on NetBSD, but suspect this is a
1616 resolver implementation change.
1618 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1619 Left warnings. Added "eximon gdb" invocation mode.
1621 PP/21 Defaulting "accept_8bitmime" to true, not false.
1623 PP/22 Added -bw for inetd wait mode support.
1625 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1626 locate the relevant includes and libraries. Made this the default.
1628 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1629 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1631 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1632 This may cause build issues on older platforms.
1634 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1635 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1636 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1637 Added SNI support via GnuTLS too.
1638 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1640 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1642 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1643 Applied second patch from Janne, implementing suggestion to default
1644 multiple-strings-in-record handling to match SPF spec.
1646 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1648 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1649 read-only, out of scope).
1650 Patch from Wolfgang Breyha, report from Stuart Northfield.
1652 PP/29 Fix three issues highlighted by clang analyser static analysis.
1653 Only crash-plausible issue would require the Cambridge-specific
1654 iplookup router and a misconfiguration.
1655 Report from Marcin Mirosław.
1657 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1659 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1660 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1661 As part of this, removing so much warning spew let me fix some minor
1662 real issues in debug logging.
1664 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1665 assignment on my part. Fixed.
1667 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1668 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1669 Janne Snabb (who went above and beyond: thank you).
1671 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1672 string otherwise requires a connection and a bunch more work and it's
1673 relatively easy to get wrong. Should also expose TLS library linkage
1676 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1677 64-bit ${eval} (JH/03).
1679 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1680 GNU libc to support some of the 64-bit stuff, should not lead to
1681 conflicts. Defined before os.h is pulled in, so if a given platform
1682 needs to override this, it can.
1684 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1685 protection layer was required, which is not implemented.
1686 Bugzilla 1254, patch from Wolfgang Breyha.
1688 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1689 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1690 tls_dhparam take prime identifiers. Also unbreak combination of
1691 OpenSSL+DH_params+TLSSNI.
1693 PP/39 Disable SSLv2 by default in OpenSSL support.
1699 PP/01 Solaris build fix for Oracle's LDAP libraries.
1700 Bugzilla 1109, patch from Stephen Usher.
1702 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1704 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1707 TF/02 Fix a couple more cases where we did not log the error message
1708 when unlink() failed. See also change 4.74-TF/03.
1710 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1711 lock up or crash if it happened to be inside a call to libc when it
1712 got a SIGUSR1 from exiwhat.
1714 The SIGUSR1 handler appends the current process status to the process
1715 log which is later printed by exiwhat. It used to use the general
1716 purpose logging code to do this, but several functions it calls are
1717 not safe for signals.
1719 The new output code in the SIGUSR1 handler is specific to the process
1720 log, and simple enough that it's easy to inspect for signal safety.
1721 Removing some special cases also simplifies the general logging code.
1722 Removing the spurious timestamps from the process log simplifies
1725 TF/04 Improved ratelimit ACL condition.
1727 The /noupdate option has been deprecated in favour of /readonly which
1728 has clearer semantics. The /leaky, /strict, and /readonly update modes
1729 are mutually exclusive. The update mode is no longer included in the
1730 database key; it just determines when the database is updated. (This
1731 means that when you upgrade Exim will forget old rate measurements.)
1733 Exim now checks that the per_* options are used with an update mode that
1734 makes sense for the current ACL. For example, when Exim is processing a
1735 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1736 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1737 must specify per_mail/readonly. If you omit the update mode it defaults to
1738 /leaky where that makes sense (as before) or /readonly where required.
1740 The /noupdate option is now undocumented but still supported for
1741 backwards compatibility. It is equivalent to /readonly except that in
1742 ACLs where /readonly is required you may specify /leaky/noupdate or
1743 /strict/noupdate which are treated the same as /readonly.
1745 A useful new feature is the /count= option. This is a generalization
1746 of the per_byte option, so that you can measure the throughput of other
1747 aggregate values. For example, the per_byte option is now equivalent
1748 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1750 The per_rcpt option has been generalized using the /count= mechanism
1751 (though it's more complicated than the per_byte equivalence). When it is
1752 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1753 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1754 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1755 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1756 Note that using per_rcpt with a non-readonly update mode in more than
1757 one ACL will cause the recipients to be double-counted. (The per_mail
1758 and per_byte options don't have this problem.)
1760 The handling of very low rates has changed slightly. If the computed rate
1761 is less than the event's count (usually one) then this event is the first
1762 after a long gap. In this case the rate is set to the same as this event's
1763 count, so that the first message of a spam run is counted properly.
1765 The major new feature is a mechanism for counting the rate of unique
1766 events. The new per_addr option counts the number of different
1767 recipients that someone has sent messages to in the last time period. It
1768 behaves like per_rcpt if all the recipient addresses are different, but
1769 duplicate recipient addresses do not increase the measured rate. Like
1770 the /count= option this is a general mechanism, so the per_addr option
1771 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1772 example, measure the rate that a client uses different sender addresses
1773 with the options per_mail/unique=$sender_address. There are further
1774 details in the main documentation.
1776 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1778 TF/06 Removed a few PCRE remnants.
1780 TF/07 Automatically extract Exim's version number from tags in the git
1781 repository when doing development or release builds.
1783 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1784 Bugzilla 879. Patch from Paul Fisher.
1786 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1787 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1790 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1792 PP/05 Variable $av_failed, true if the AV scanner deferred.
1793 Bugzilla 1078. Patch from John Horne.
1795 PP/06 Stop make process more reliably on build failure.
1796 Bugzilla 1087. Patch from Heiko Schlittermann.
1798 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1799 Bugzilla 1089. Patch from Heiko Schlittermann.
1801 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1802 Bugzilla 1131. Patch from Holger Weiß.
1804 PP/09 Handle IPv6 addresses with SPF.
1805 Bugzilla 860. Patch from Wolfgang Breyha.
1807 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1809 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1812 PP/11 match_* no longer expand right-hand-side by default.
1813 New compile-time build option, EXPAND_LISTMATCH_RHS.
1814 New expansion conditions, "inlist", "inlisti".
1816 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1818 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1820 PP/14 fix log_write() format string regression from TF/03.
1821 Bugzilla 1152. Patch from Dmitry Isaikin.
1827 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1829 PP/02 Harmonised TLS library version reporting; only show if debugging.
1830 Layout now matches that introduced for other libraries in 4.74 PP/03.
1832 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1834 PP/04 New "dns_use_edns0" global option.
1836 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1839 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1840 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1842 TK/01 Updated PolarSSL code to 0.14.2.
1843 Bugzilla 1097. Patch from Andreas Metzler.
1845 PP/07 Catch divide-by-zero in ${eval:...}.
1846 Fixes bugzilla 1102.
1848 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1851 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1852 format-string attack -- SECURITY: remote arbitrary code execution.
1854 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1855 time unintentionally subject to list matching rules, letting the header
1856 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1857 arbitrary string expansion). This allowed for information disclosure.
1859 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1860 INT_MIN/-1 -- value coerced to INT_MAX.
1866 NM/01 Workaround for PCRE version dependency in version reporting
1869 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1870 This fixes portability to compilers other than gcc, notably
1871 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1873 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1874 makefiles for portability to HP-UX and POSIX correctness.
1876 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1877 Also via indented variable definition in the Makefile.
1878 (Debugging by Oliver Heesakkers).
1880 PP/02 Restore caching of spamd results with expanded spamd_address.
1881 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1883 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1884 Improves build reliability. Fix from: Frank Elsner
1886 NM/02 Fix wide character breakage in the rfc2047 coding
1887 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1889 NM/03 Allow underscore in dnslist lookups
1890 Fixes bug 1026. Patch from Graeme Fowler
1892 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1893 Code patches from Adam Ciarcinski of NetBSD.
1895 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1898 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1899 is logged, to avoid truncation. Patch from John Horne.
1901 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1902 Patch from Jakob Hirsch.
1904 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1905 SQL string expansion failure details.
1906 Patch from Andrey Oktyabrski.
1908 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1909 Patch from Simon Arlott.
1911 PP/09 New lookups functionality failed to compile on old gcc which rejects
1912 extern declarations in function scope.
1913 Patch from Oliver Fleischmann
1915 PP/10 Use sig_atomic_t for flags set from signal handlers.
1916 Check getgroups() return and improve debugging.
1917 Fixed developed for diagnosis in bug 927 (which turned out to be
1920 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1921 Patch from Mark Zealey.
1923 PP/12 Bugzilla 1056: Improved spamd server selection.
1924 Patch from Mark Zealey.
1926 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1927 Based on patch from Heiko Schlittermann.
1929 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1930 Patch from Uwe Doering, sign-off by Michael Haardt.
1932 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1933 variable declaration deep within a block. Bug and patch from
1936 PP/15 lookups-Makefile IRIX compatibility coercion.
1938 PP/16 Make DISABLE_DKIM build knob functional.
1940 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1941 Patch by Simon Arlott
1943 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1944 variable argument macros. Our copy now differs from upstream.
1950 TF/01 Failure to get a lock on a hints database can have serious
1951 consequences so log it to the panic log.
1953 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1954 controlled using the smtp_confirmation log selector.
1956 TF/03 Include the error message when we fail to unlink a spool file.
1958 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1959 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1960 for maintaining out-of-tree patches for some time.
1962 PP/01 Bugzilla 139: Documentation and portability issues.
1963 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1964 Handle per-OS dynamic-module compilation flags.
1966 PP/02 Let /dev/null have normal permissions.
1967 The 4.73 fixes were a little too stringent and complained about the
1968 permissions on /dev/null. Exempt it from some checks.
1969 Reported by Andreas M. Kirchwitz.
1971 PP/03 Report version information for many libraries, including
1972 Exim version information for dynamically loaded libraries. Created
1973 version.h, now support a version extension string for distributors
1974 who patch heavily. Dynamic module ABI change.
1976 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1977 privilege escalation vulnerability whereby the Exim run-time user
1978 can cause root to append content of the attacker's choosing to
1981 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1984 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1985 If dropping privileges for untrusted macros, we disabled normal logging
1986 on the basis that it would fail; for the Exim run-time user, this is not
1987 the case, and it resulted in successful deliveries going unlogged.
1988 Fixed. Reported by Andreas Metzler.
1994 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1995 only prepend for the Resent-* case. Fixes regression introduced in
1996 Exim 4.70 by NM/22 for Bugzilla 607.
1998 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1999 increasing numbers of administrators be bitten by this.
2001 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
2003 PP/03 Bugzilla 994: added openssl_options main configuration option.
2005 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
2007 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
2009 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
2011 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
2012 without a peer certificate, leading to a segfault because of an
2013 assumption that peers always have certificates. Be a little more
2014 paranoid. Problem reported by Martin Tscholak.
2016 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
2017 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
2018 NB: ClamAV planning to remove STREAM in "middle of 2010".
2019 CL also introduces -bmalware, various -d+acl logging additions and
2020 more caution in buffer sizes.
2022 PP/09 Implemented reverse_ip expansion operator.
2024 PP/10 Bugzilla 937: provide a "debug" ACL control.
2026 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
2028 PP/12 Bugzilla 973: Implement --version.
2030 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2032 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2034 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2036 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2037 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2038 ignore trailing whitespace.
2040 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2042 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2045 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2046 Notification from Dr Andrew Aitchison.
2048 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2049 ExtendedDetectionInfo response format.
2050 Notification from John Horne.
2052 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2055 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2056 XSL and documented dependency on system catalogs, with examples of how
2059 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2062 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2063 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2064 configuration file which is writeable by the Exim user or group.
2066 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2067 of configuration files to cover files specified with the -C option if
2068 they are going to be used with root privileges, not just the default
2071 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2072 option (effectively making it always true).
2074 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2075 files to be used while preserving root privileges.
2077 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2078 that rogue child processes cannot use them.
2080 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2081 run-time user, instead of root.
2083 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2084 Exim run-time user without dropping privileges.
2086 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2087 result string, instead of calling string_vformat() twice with the same
2090 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2091 for other users. Others should always drop root privileges if they use
2092 -C on the command line, even for a whitelisted configure file.
2094 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2096 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2102 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2103 $data_path, and $header_path variables; fixed documentation bugs and
2106 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2107 exipick to access non-standard spools, including the "frozen" queue
2110 NM/01 Bugzilla 965: Support mysql stored procedures.
2111 Patch from Alain Williams
2113 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2115 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2116 Patch from Andreas Metzler
2118 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2119 Patch from Kirill Miazine
2121 NM/05 Bugzilla 671: Added umask to procmail example.
2123 JJ/03 installed exipick 20100323.0, fixing doc bug
2125 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2126 directory. Notification and patch from Dan Rosenberg.
2128 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2130 TK/02 Improve log output when DKIM signing operation fails.
2132 MH/01 Treat the transport option dkim_domain as a colon separated
2133 list, not as a single string, and sign the message with each element,
2134 omitting multiple occurences of the same signer.
2136 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2137 Bugzilla 985, 986. Patch by Simon Arlott
2139 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2140 Patch by Simon Arlott
2142 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2143 MBX locking. Notification from Dan Rosenberg.
2149 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2151 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2153 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2155 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2157 NM/04 Bugzilla 915: Flag broken perl installation during build.
2163 TK/01 Added patch by Johannes Berg that expands the main option
2164 "spamd_address" if it starts with a dollar sign.
2166 TK/02 Write list of recipients to X-Envelope-Sender header when building
2167 the mbox-format spool file for content scanning (suggested by Jakob
2170 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2171 (http://www.dcc-servers.net/) support via dccifd. Activated by
2172 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2174 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2175 by Mark Daniel Reidel <mr@df.eu>.
2177 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2178 When building exim an external PCRE library is now needed -
2179 PCRE is a system library on the majority of modern systems.
2180 See entry on PCRE_LIBS in EDITME file.
2182 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2183 conversation. Added nologin parameter to request.
2184 Patch contributed by Kirill Miazine.
2186 TF/01 Do not log submission mode rewrites if they do not change the address.
2188 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2190 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2191 log files in place. Contributed by Roberto Lima.
2193 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2195 TF/03 Bugzilla 615: When checking the local_parts router precondition
2196 after a local_part_suffix or local_part_prefix option, Exim now
2197 does not use the address's named list lookup cache, since this
2198 contains cached lookups for the whole local part.
2200 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2201 Robert Millan. Documentation is in experimental-spec.txt.
2203 TF/04 Bugzilla 668: Fix parallel build (make -j).
2205 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2207 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2208 Patch provided by Jan Srzednicki.
2210 TF/05 Leading white space used to be stripped from $spam_report which
2211 wrecked the formatting. Now it is preserved.
2213 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2214 that they are available at delivery time.
2216 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2218 TF/08 TLS error reporting now respects the incoming_interface and
2219 incoming_port log selectors.
2221 TF/09 Produce a more useful error message if an SMTP transport's hosts
2222 setting expands to an empty string.
2224 NM/06 Bugzilla 744: EXPN did not work under TLS.
2225 Patch provided by Phil Pennock.
2227 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2228 Patch provided by Richard Godbee.
2230 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2231 acl_smtp_notquit, added index entry.
2233 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2234 Patch provided by Eugene Bujak.
2236 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2237 accept(). Patch provided by Maxim Dounin.
2239 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2240 Patch provided by Phil Pennock.
2242 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2244 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2245 Patch provided by Brad "anomie" Jorsch.
2247 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2248 Patch provided by Dean Brooks.
2250 TK/05 Add native DKIM support (does not depend on external libraries).
2252 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2253 Patch provided by Graeme Fowler.
2255 NM/16 Bugzilla 851: Documentation example syntax fix.
2257 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2259 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2262 NM/19 Bugzilla 745: TLS version reporting.
2263 Patch provided by Phil Pennock.
2265 NM/20 Bugzilla 167: bool: condition support.
2266 Patch provided by Phil Pennock.
2268 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2269 clients. Patch provided by Phil Pennock.
2271 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2272 Patch provided by Brad "anomie" Jorsch.
2274 NM/23 Bugzilla 687: Fix misparses in eximstats.
2275 Patch provided by Heiko Schlittermann.
2277 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2278 Patch provided by Heiko Schlittermann.
2280 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2281 plus update to original patch.
2283 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2285 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2286 Patch provided by David Brownlee.
2288 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2290 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2292 NM/30 Bugzilla 888: TLS documentation bugfixes.
2294 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2296 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2297 Unlike the original bugzilla I have changed all shell scripts in src tree.
2299 NM/33 Bugzilla 898: Transport filter timeout fix.
2300 Patch by Todd Rinaldo.
2302 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2303 Patch by Serge Demonchaux.
2305 NM/35 Bugzilla 39: Base64 decode bug fixes.
2306 Patch by Jakob Hirsch.
2308 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2310 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2312 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2314 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2320 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2321 ALT-N's libdkim that I have put here:
2322 http://duncanthrax.net/exim-experimental/
2324 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2325 were called 'true' and it seems that C99 defines that as a reserved
2326 keyword to be used with 'bool' variable types. That means you could
2327 not include C99-style headers which use bools without triggering
2328 build errors in sieve.c.
2330 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2331 as mailq or other aliases. Changed the --help handling significantly
2332 to do whats expected. exim_usage() emits usage/help information.
2334 SC/01 Added the -bylocaldomain option to eximstats.
2336 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2338 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2340 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2346 PH/01 Another patch from the Sieve maintainer.
2348 PH/02 When an IPv6 address is converted to a string for single-key lookup
2349 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2350 dots are used instead of colons so that keys in lsearch files need not
2351 contain colons. This was done some time before quoting was made available
2352 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2353 (notated using the quote facility) so as to distinguish them from IPv4
2354 keys. This meant that lookups for IP addresses in host lists did not work
2355 for iplsearch lookups.
2357 This has been fixed by arranging for IPv6 addresses to be expressed with
2358 colons if the lookup type is iplsearch. This is not incompatible, because
2359 previously such lookups could never work.
2361 The situation is now rather anomalous, since one *can* have colons in
2362 ordinary lsearch keys. However, making the change in all cases is
2363 incompatible and would probably break a number of configurations.
2365 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2368 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2369 conversion specification without a maximum field width, thereby enabling
2370 a rogue spamd server to cause a buffer overflow. While nobody in their
2371 right mind would setup Exim to query an untrusted spamd server, an
2372 attacker that gains access to a server running spamd could potentially
2373 exploit this vulnerability to run arbitrary code as the Exim user.
2375 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2376 $primary_hostname instead of what libspf2 thinks the hosts name is.
2378 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2379 a directory entry by the name of the lookup key. Previously, if a
2380 symlink pointed to a non-existing file or a file in a directory that
2381 Exim lacked permissions to read, a lookup for a key matching that
2382 symlink would fail. Now it is enough that a matching directory entry
2383 exists, symlink or not. (Bugzilla 503.)
2385 PH/03 The body_linecount and body_zerocount variables are now exported in the
2388 PH/04 Added the $dnslist_matched variable.
2390 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2391 This means they are set thereafter only if the connection becomes
2394 PH/06 Added the client_condition to authenticators so that some can be skipped
2395 by clients under certain conditions.
2397 PH/07 The error message for a badly-placed control=no_multiline_responses left
2398 "_responses" off the end of the name.
2400 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2402 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2403 (without spaces) instead of just copying the configuration text.
2405 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2407 PH/11 Added $max_received_linelength.
2409 PH/12 Added +ignore_defer and +include_defer to host lists.
2411 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2412 way in which PCRE > 7.0 is built.
2414 PH/14 Implemented queue_only_load_latch.
2416 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2417 MAIL command. The effect was to mangle the value on 64-bit systems.
2419 PH/16 Another patch from the Sieve maintainer.
2421 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2423 PH/18 If a system quota error occurred while trying to create the file for
2424 a maildir delivery, the message "Mailbox is full" was not appended to the
2425 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2426 only to a quota excession during the actual writing of the file.
2428 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2429 characters?) which causes problems in log lines. The DN values are now
2430 passed through string_printing() before being added to log lines.
2432 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2433 and InterBase are left for another time.)
2435 PH/21 Added message_body_newlines option.
2437 PH/22 Guard against possible overflow in moan_check_errorcopy().
2439 PH/23 POSIX allows open() to be a macro; guard against that.
2441 PH/24 If the recipient of an error message contained an @ in the local part
2442 (suitably quoted, of course), incorrect values were put in $domain and
2443 $local_part during the evaluation of errors_copy.
2449 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2450 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2453 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2454 issue a MAIL command.
2456 PH/02 In an ACL statement such as
2458 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2460 if a client was not listed at all, or was listed with a value other than
2461 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2462 the condition was not true (as it should be), so access was not denied.
2463 The bug was that the ! inversion was incorrectly passed on to the second
2464 item. This has been fixed.
2466 PH/03 Added additional dnslists conditions == and =& which are different from
2467 = and & when the dns lookup returns more than one IP address.
2469 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2470 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2472 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2473 FSYNC, which compiles an option called disable_fsync that allows for
2474 bypassing fsync(). The documentation is heavily laced with warnings.
2476 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2478 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2479 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2480 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2481 including adding "make clean"; (3) Added -fPIC when compiling the test
2482 dynamically loaded module, to get rid of a warning.
2484 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2485 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2486 = 0s. The bug is otherwise harmless.
2488 PH/07 There was a bug in the dovecot authenticator such that the value of
2489 $auth1 could be overwritten, and so not correctly preserved, after a
2490 successful authentication. This usually meant that the value preserved by
2491 the server_setid option was incorrect.
2493 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2495 PH/09 Installed PCRE release 7.0.
2497 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2498 run for batched SMTP input. It is now run at the start of every message
2499 in the batch. While fixing this I discovered that the process information
2500 (output by running exiwhat) was not always getting set for -bs and -bS
2501 input. This is fixed, and it now also says "batched" for BSMTP.
2503 PH/11 Added control=no_pipelining.
2505 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2506 patch, slightly modified), and move the expansion of helo_data till after
2507 the connection is made in the smtp transport (so it can use these
2510 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2512 PH/14 Added log_selector = +pid.
2514 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2516 PH/16 Add ${if forany and ${if forall.
2518 PH/17 Added dsn_from option to vary the From: line in DSNs.
2520 PH/18 Flush SMTP output before performing a callout, unless control =
2521 no_callout_flush is set.
2523 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2524 was true (the default) a successful delivery failed to delete the retry
2525 item, thus causing premature timeout of the address. The bug is now
2528 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2530 PH/21 Long custom messages for fakedefer and fakereject are now split up
2531 into multiline responses in the same way that messages for "deny" and
2532 other ACL rejections are.
2534 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2535 with slight modification.
2537 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2538 draft, changing the syntax and factoring some duplicate code".
2540 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2541 for deliveries of the second and subsequent messages over the same SMTP
2544 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2545 ${reduce, with only minor "tidies".
2547 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2549 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2550 expansion side effects.
2552 PH/27 When a message times out after an over-quota error from an Exim-imposed
2553 quota, the bounce message says "mailbox is full". This message was not
2554 being given when it was a system quota that was exceeded. It now should
2557 MH/03 Made $recipients available in local_scan(). local_scan() already has
2558 better access to the recipient list through recipients_list[], but
2559 $recipients can be useful in postmaster-provided expansion strings.
2561 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2562 in the case of a MAIL command with additional options following the
2563 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2564 were accidentally chopped off.
2566 PH/29 SMTP synchronization checks are implemented when a command is read -
2567 there is a check that no more input is waiting when there shouldn't be
2568 any. However, for some commands, a delay in an ACL can mean that it is
2569 some time before the response is written. In this time, more input might
2570 arrive, invalidly. So now there are extra checks after an ACL has run for
2571 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2572 pipelining has not been advertised.
2574 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2576 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2577 returned in the SMTP response when smtp_return_error_details was false.
2578 This has been fixed.
2580 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2581 instead of the C I/O that was originally supplied, because problems were
2582 reported on Solaris.
2584 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2585 Exim which did not show up earlier: it was assuming that a call to
2586 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2587 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2588 was a macro that became an assignment, so it seemed to work. This has
2589 changed to a proper function call with a void return, hence the compile
2590 error. Exim's code has been fixed.
2592 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2595 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2597 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2599 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2600 because it turns out to be too big or there is a timeout) there is no
2601 "Completed" line in the log. When some messages of this type were
2602 selected by exigrep, they were listed as "not completed". Others were
2603 picked up by some special patterns. I have improved the selection
2604 criteria to be more general.
2606 PH/38 The host_find_failed option in the manualroute router can now be set
2607 to "ignore", to completely ignore a host whose IP address cannot be
2608 found. If all hosts are ignored, the behaviour is controlled by the new
2609 host_all_ignored option.
2611 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2612 homing or because of multiple MX records with /mx) generated more than
2613 one IP address, and the following item turned out to be the local host,
2614 all the secondary addresses of the first item were incorrectly removed
2615 from the list, along with the local host and any following hosts (which
2616 is what is supposed to happen).
2618 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2619 whoever called Exim into the -H file. In the case of the daemon it was
2620 behaving confusingly. When first started, it used values for whoever
2621 started the daemon, but after a SIGHUP it used the Exim user (because it
2622 calls itself on a restart). I have changed the code so that it now always
2625 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2626 message are rejected with the same error (e.g. no authentication or bad
2627 sender address), and a DATA command is nevertheless sent (as can happen
2628 with PIPELINING or a stupid MUA), the error message that was given to the
2629 RCPT commands is included in the rejection of the DATA command. This is
2630 intended to be helpful for MUAs that show only the final error to their
2633 PH/42 Another patch from the Sieve maintainer.
2635 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2636 Eximstats - Fixed some broken HTML links and added missing column headers
2638 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2639 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2641 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2643 PH/43 Yet another patch from the Sieve maintainer.
2645 PH/44 I found a way to check for a TCP/IP connection going away before sending
2646 the response to the final '.' that terminates a message, but only in the
2647 case where the client has not sent further data following the '.'
2648 (unfortunately, this is allowed). However, in many cases there won't be
2649 any further data because there won't be any more messages to send. A call
2650 to select() can be used: if it shows that the input is "ready", there is
2651 either input waiting, or the socket has been closed. An attempt to read
2652 the next input character can distinguish the two cases. Previously, Exim
2653 would have sent an OK response which the client would never have see.
2654 This could lead to message repetition. This fix should cure that, at
2655 least in a lot of common cases.
2657 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2658 advertised in response to EHLO.
2664 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2665 fixed by 4.65/MH/01 (is this a record?) are fixed:
2667 (i) An empty string was always treated as zero by the numeric comparison
2668 operators. This behaviour has been restored.
2670 (ii) It is documented that the numeric comparison operators always treat
2671 their arguments as decimal numbers. This was broken in that numbers
2672 starting with 0 were being interpreted as octal.
2674 While fixing these problems I realized that there was another issue that
2675 hadn't been noticed. Values of message_size_limit (both the global option
2676 and the transport option) were treated as octal if they started with 0.
2677 The documentation was vague. These values are now always treated as
2678 decimal, and I will make that clear in the documentation.
2684 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2685 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2688 MH/01 Don't check that the operands of numeric comparison operators are
2689 integers when their expansion is in "skipping" mode (fixes bug
2690 introduced by 4.64-PH/07).
2692 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2693 child addresses, Exim now panics and dies. Previously, because the count
2694 is held in a short int, deliveries were likely to be lost. As such a
2695 large number of recipients for a single message is ridiculous
2696 (performance will be very, very poor), I have chosen to impose a limit
2697 rather than extend the field.
2703 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2704 leftover -K file (the existence of which was triggered by #402).
2705 While we were at it, introduced process PID as part of the -K
2706 filename. This should rule out race conditions when creating
2709 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2710 processing considerably. Previous code took too long for large mails,
2711 triggering a timeout which in turn triggers #401.
2713 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2714 in the DK code in transports.c. sendfile() is not really portable,
2715 hence the _LINUX specificness.
2717 TF/01 In the add_headers option to the mail command in an Exim filter,
2718 there was a bug that Exim would claim a syntax error in any
2719 header after the first one which had an odd number of characters
2722 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2723 callout verification, Exim cached a "reject" for the entire domain. This
2724 is correct for most verifications, but it is not correct for a recipient
2725 verification with use_sender or use_postmaster set, because in that case
2726 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2727 case of MAIL FROM:<> rejection from other early rejections (e.g.
2728 rejection of HELO). When verifying a recipient using a non-null MAIL
2729 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2730 Whatever the result of the callout, the value of the domain cache is
2731 left unchanged (for any other kind of callout, getting as far as trying
2732 RCPT means that the domain itself is ok).
2734 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2737 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2738 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2739 session, the daemon ignores ECONNECT errors and logs others; it now
2740 ignores EPIPE as well.
2742 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2743 (quoted-printable decoding).
2745 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2746 later the small subsequent patch to fix an introduced bug.
2748 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2750 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2752 PH/08 An error is now given if message_size_limit is specified negative.
2754 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2755 to be given (somewhat) arbitrary names.
2757 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2760 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2761 miscellaneous code fixes
2763 PH/10 Added the log_reject_target ACL modifier to specify where to log
2766 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2767 hostname. This is wrong, because it relates to the incoming message (and
2768 probably the interface on which it is arriving) and not to the outgoing
2769 callout (which could be using a different interface). This has been
2770 changed to use the value of the helo_data option from the smtp transport
2771 instead - this is what is used when a message is actually being sent. If
2772 there is no remote transport (possible with a router that sets up host
2773 addresses), $smtp_active_hostname is used.
2775 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2776 tweaks were necessary in order to get it to work (see also 21 below):
2777 (a) The code assumed that strncpy() returns a negative number on buffer
2778 overflow, which isn't the case. Replaced with Exim's string_format()
2780 (b) There were several signed/unsigned issues. I just did the minimum
2781 hacking in of casts. There is scope for a larger refactoring.
2782 (c) The code used strcasecmp() which is not a standard C function.
2783 Replaced with Exim's strcmpic() function.
2784 (d) The code set only $1; it now sets $auth1 as well.
2785 (e) A simple test gave the error "authentication client didn't specify
2786 service in request". It would seem that Dovecot has changed its
2787 interface. Fortunately there's a specification; I followed it and
2788 changed what the client sends and it appears to be working now.
2790 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2793 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2794 address A is aliased to B and C, where B exists and C does not. Without
2795 -v the output is "A verified" because verification stops after a
2796 successful redirection if more than one address is generated. However,
2797 with -v the child addresses are also verified. Exim was outputting "A
2798 failed to verify" and then showing the successful verification for C,
2799 with its parentage. It now outputs "B failed to verify", showing B's
2800 parentage before showing the successful verification of C.
2802 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2803 look up a TXT record in a specific list after matching in a combined
2806 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2807 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2808 they consult the DNS. I had assumed they would set it the way they
2809 wanted; and indeed my experiments on Linux seem to show that in some
2810 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2811 To be on the safe side, however, I have now made the interface to
2812 host_find_byname() similar to host_find_bydns(), with an argument
2813 containing the DNS resolver options. The host_find_byname() function now
2814 sets these options at its start, just as host_find_bydns() does. The smtp
2815 transport options dns_qualify_single and dns_search_parents are passed to
2816 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2817 of host_find_byname() use the default settings of RES_DEFNAMES
2818 (qualify_single) but not RES_DNSRCH (search_parents).
2820 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2821 spool_read_header() do less string testing, by means of a preliminary
2822 switch on the second character of optional "-foo" lines. (This is
2823 overdue, caused by the large number of possibilities that now exist.
2824 Originally there were few.) While I was there, I also converted the
2825 str(n)cmp tests so they don't re-test the leading "-" and the first
2826 character, in the hope this might squeeze out yet more improvement.
2828 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2829 flag allowing group syntax was set by the header_syntax check but not
2830 turned off, possible causing trouble later; (2) The flag was not being
2831 set at all for the header_verify test, causing "group"-style headers to
2832 be rejected. I have now set it in this case, and also caused header_
2833 verify to ignore an empty address taken from a group. While doing this, I
2834 came across some other cases where the code for allowing group syntax
2835 while scanning a header line wasn't quite right (mostly, not resetting
2836 the flag correctly in the right place). These bugs could have caused
2837 trouble for malformed header lines. I hope it is now all correct.
2839 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2840 with the "reply" argument non-NULL. The code, however (which originally
2841 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2842 but it didn't always do it. This confused somebody who was copying the
2843 code for some other use. I have removed all the tests.
2845 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2846 feature that was used to support insecure browsers during the U.S. crypto
2847 embargo. It requires special client support, and Exim is probably the
2848 only MTA that supported it -- and would never use it because real RSA is
2849 always available. This code has been removed, because it had the bad
2850 effect of slowing Exim down by computing (never used) parameters for the
2851 RSA_EXPORT functionality.
2853 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2854 authenticator to fail if there's a tab character in the incoming data
2855 (there should never be unless someone is messing about, as it's supposed
2856 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2857 if the connection is using TLS or if the remote IP is the same as the
2858 local IP, and the "valid-client-cert option" if a client certificate has
2861 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2862 authenticators. This can be used for authorization after authentication
2863 succeeds. (In the case of plaintext, it servers for both authentication
2866 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2867 if any retry times were supplied.
2869 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2870 connection, where there is no client IP address to check. In this
2871 situation, the verify now always succeeds.
2873 PH/25 Applied John Jetmore's -Mset patch.
2875 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2877 PH/27 In a string expansion for a processed (not raw) header when multiple
2878 headers of the same name were present, leading whitespace was being
2879 removed from all of them, but trailing whitespace was being removed only
2880 from the last one. Now trailing whitespace is removed from each header
2881 before concatenation. Completely empty headers in a concatenation (as
2882 before) are ignored.
2884 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2885 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2887 PH/29 [Removed. This was a change that I later backed out, and forgot to
2888 correct the ChangeLog entry (that I had efficiently created) before
2889 committing the later change.]
2891 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2892 address errors (4xx response to RCPT) over the same connection as other
2893 messages routed to the same hosts. Such deliveries are always "forced",
2894 so retry times are not inspected. This resulted in far too many retries
2895 for the affected addresses. The effect occurred only when there were more
2896 hosts than the hosts_max_try setting in the smtp transport when it had
2897 the 4xx errors. Those hosts that it had tried were not added to the list
2898 of hosts for which the message was waiting, so if all were tried, there
2899 was no problem. Two fixes have been applied:
2901 (i) If there are any address or message errors in an SMTP delivery, none
2902 of the hosts (tried or untried) are now added to the list of hosts
2903 for which the message is waiting, so the message should not be a
2904 candidate for sending over the same connection that was used for a
2905 successful delivery of some other message. This seems entirely
2906 reasonable: after all the message is NOT "waiting for some host".
2907 This is so "obvious" that I'm not sure why it wasn't done
2908 previously. Hope I haven't missed anything, but it can't do any
2909 harm, as the worst effect is to miss an optimization.
2911 (ii) If, despite (i), such a delivery is accidentally attempted, the
2912 routing retry time is respected, so at least it doesn't keep
2913 hammering the server.
2915 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2916 in ${readsocket because some servers need this prod.
2918 PH/32 Added some extra debug output when updating a wait-xxx database.
2920 PH/33 The hint "could be header name not terminated by colon", which has been
2921 given for certain expansion errors for a long time, was not being given
2922 for the ${if def:h_colon_omitted{... case.
2924 PH/34 The spec says: "With one important exception, whenever a domain list is
2925 being scanned, $domain contains the subject domain." There was at least
2926 one case where this was not true.
2928 PH/35 The error "getsockname() failed: connection reset by peer" was being
2929 written to the panic log as well as the main log, but it isn't really
2930 panic-worthy as it just means the connection died rather early on. I have
2931 removed the panic log writing for the ECONNRESET error when getsockname()
2934 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2935 runs only) independently of the message's sender address. This meant
2936 that, if the 4xx error was in fact related to the sender, a different
2937 message to the same recipient with a different sender could confuse
2938 things. In particular, this can happen when sending to a greylisting
2939 server, but other circumstances could also provoke similar problems.
2940 I have changed the default so that the retry time for these errors is now
2941 based a combination of the sender and recipient addresses. This change
2942 can be overridden by setting address_retry_include_sender=false in the
2945 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2946 remote server are returned as part of bounce messages. This was not
2947 happening for LMTP over a pipe (the lmtp transport), but now it is the
2948 same for both kinds of LMTP.
2950 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2951 in header lines that were in fact CNAMEs. This is no longer the case.
2953 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2954 and queue runs started by the daemon processed all messages. This has
2955 been fixed so that -R and -S can now usefully be given with -q<time>.
2957 PH/40 Import PCRE release 6.7 (fixes some bugs).
2959 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2961 PH/42 Give an error if -q is specified more than once.
2963 PH/43 Renamed the variables $interface_address and $interface_port as
2964 $received_ip_address and $received_port, to make it clear that these
2965 values apply to message reception, and not to the outgoing interface when
2966 a message is delivered. (The old names remain recognized, of course.)
2968 PH/44 There was no timeout on the connect() call when using a Unix domain
2969 socket in the ${readsocket expansion. There now is.
2971 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2972 be meaningful with "accept".
2974 SC/01 Eximstats V1.43
2975 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2977 SC/02 Eximstats V1.44
2978 Use a glob alias rather than an array ref in the generated
2979 parser. This improves both readability and performance.
2981 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2982 Collect SpamAssassin and rejection statistics.
2983 Don't display local sender or destination tables unless
2984 there is data to show.
2985 Added average volumes into the top table text output.
2987 SC/04 Eximstats V1.46
2988 Collect data on the number of addresses (recipients)
2989 as well as the number of messages.
2991 SC/05 Eximstats V1.47
2992 Added 'Message too big' to the list of mail rejection
2993 reasons (thanks to Marco Gaiarin).
2995 SC/06 Eximstats V1.48
2996 Mainlog lines which have GMT offsets and are too short to
2997 have a flag are now skipped.
2999 SC/07 Eximstats V1.49 (Alain Williams)
3000 Added the -emptyok flag.
3002 SC/08 Eximstats V1.50
3003 Fixes for obtaining the IP address from reject messages.
3005 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
3006 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
3007 whitespace changes from 4.64-PH/27
3009 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
3012 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
3013 are found, allow negative numbers in numeric criteria)
3015 JJ/06 exipick.20061117.2, added new $message_body_missing variable
3017 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
3018 to match changes made in 4.64-PH/43
3020 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
3022 PH/47 Put in an explicit test for a DNS lookup of an address record where the
3023 "domain" is actually an IP address, and force a failure. This locks out
3024 those revolvers/nameservers that support "A-for-A" lookups, in
3025 contravention of the specifications.
3027 PH/48 When a host name was looked up from an IP address, and the subsequent
3028 forward lookup of the name timed out, the host name was left in
3029 $sender_host_name, contrary to the specification.
3031 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3032 restricted to single-key lookups, Exim was not diagnosing an error if
3033 * or *@ was used with a query-style lookup.
3035 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3037 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3038 long ago, but noone interested enough thought of it. Let's just say that
3039 the "1.1" means that there are some new functions that weren't there at
3040 some point in the past.
3042 PH/51 Error processing for expansion failure of helo_data from an smtp
3043 transport during callout processing was broken.
3045 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3046 tested/used via the -bh/-bhc/-bs options.
3048 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3049 bug, fixed in subsequent PCRE releases).
3051 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3052 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3054 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3060 SC/01 Use a glob alias rather than an array ref in eximstats generated
3061 parser. This improves both readability and performance.
3063 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3064 Don't display local sender or destination tables in eximstats unless
3065 there is data to show.
3066 Added average volumes into the eximstats top table text output.
3068 SC/03 Collect data on the number of addresses (recipients) as well
3069 as the number of messages in eximstats.
3071 TF/01 Correct an error in the documentation for the redirect router. Exim
3072 does (usually) call initgroups() when daemonizing.
3074 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3075 with consistent privilege compared to when running as a daemon.
3077 TF/03 Note in the spec that $authenticated_id is not set for local
3078 submissions from trusted users.
3080 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3081 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3083 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3084 by adding some example configuration directives to the default
3085 configuration file. A little bit of work is required to uncomment the
3086 directives and define how usernames and passwords are checked, but
3087 there is now a framework to start from.
3089 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3090 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3091 without this. I don't know how relevant this is to other LDAP libraries.
3093 PH/02 Add the verb name to the "unknown ACL verb" error.
3095 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3097 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3099 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3100 directory not expanded when it should be if an expanded home directory
3101 was set for the address (which is overridden by the transport).
3103 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3106 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3107 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3108 because it is too late at that time, and has no effect.
3110 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3111 security issue with \' (bugzilla #107). I could not use the
3112 PQescapeStringConn() function, because it needs a PGconn value as one of
3115 PH/08 When testing addresses using -bt, indicate those final addresses that
3116 are duplicates that would not cause an additional delivery. At least one
3117 person was confused, thinking that -bt output corresponded to deliveries.
3118 (Suppressing duplicates isn't a good idea as you lose the information
3119 about possibly different redirections that led to the duplicates.)
3121 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3122 systems where poll() doesn't work, in particular OS X.
3124 PH/10 Added more information to debugging output for retry time not reached.
3126 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3127 operations in malware.c.
3129 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3132 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3133 syslog_duplication set false, log lines that would normally be written
3134 both the the main log and to the reject log were not written to syslog at
3137 PH/14 In the default configuration, change the use of "message" in ACL warn
3138 statements to "add_header".
3140 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3141 not followed by a command (e.g. "seen endif").
3143 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3144 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3147 PH/17 Added extra conditions to the default value of delay_warning_condition
3151 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3152 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3153 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3156 The Auto-Submitted: and various List- headers are standardised, whereas I
3157 don't think Precedence: ever was.
3159 PH/18 Refactored debugging code in route_finduser() to show more information,
3160 in particular, the error code if getpwnam() issues one.
3162 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3163 This is apparently needed in addition to the PH/07 change above to avoid
3164 any possible encoding problems.
3166 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3167 but not after initializing Perl.
3169 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3170 output them only if debugging. By default they are written stderr,
3171 apparently, which is not desirable.
3173 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3176 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3179 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3181 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3182 authenticated or an ident call has been made. Suppress the default
3183 values for $authenticated_id and $authenticated_sender (but permit -oMai
3184 and -oMas) when testing with -bh.
3186 PH/24 Re-jigged the order of the tests in the default configuration so that the
3187 tests for valid domains and recipients precede the DNS black list and CSA
3188 tests, on the grounds that those ones are more expensive.
3190 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3191 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3192 This bug exists in every version of Exim that I still have, right back to
3195 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3196 However, an attempt to turn on case-sensitivity in a regex key by
3197 including (?-i) didn't work because the subject string was already
3198 lowercased, and the effects were non-intuitive. It turns out that a
3199 one-line patch can be used to allow (?-i) to work as expected.
3205 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3206 other effects) broke the use of negated acl sub-conditions.
3208 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3211 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3212 "Deny" causes Exim to reject the incoming connection with a 554 error.
3213 Unfortunately, if there is a major crisis, such as a disk failure,
3214 tcp-wrappers gives "deny", whereas what one would like would be some
3215 kind of temporary error. A kludge has been added to help with this.
3216 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3217 554 error is used if errno is still zero or contains ENOENT (which occurs
3218 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3221 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3223 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3224 errors. Otherwise a message that provokes a temporary error (when other
3225 messages do not) can cause a whole host to time out.
3227 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3228 addresses were routed directly to files or pipes from a redirect router.
3229 File deliveries just didn't batch; pipe deliveries might have suffered
3232 PH/06 A failure to get a lock for a hints database would erroneously always say
3233 "Failed to get write lock", even when it was really a read lock.
3235 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3236 of 0600. This has been changed to use the value of the lockfile_mode
3237 option (which defaults to 0600).
3239 PH/08 Applied small patch from the Sieve maintainer.
3241 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3242 folder from quota calculations, a direct delivery into this folder messed
3243 up the contents of the maildirsize file. This was because the regex was
3244 used only to exclude .Trash (or whatever) when the size of the mailbox
3245 was calculated. There was no check that a delivery was happening into an
3246 excluded directory. This bug has been fixed by ignoring all quota
3247 processing for deliveries into excluded directories.
3249 PH/10 Added the maildirfolder_create_regex option to appendfile.
3255 PH/01 The code for finding all the local interface addresses on a FreeBSD
3256 system running IPv6 was broken. This may well have applied to all BSD
3257 systems, as well as to others that have similar system calls. The broken
3258 code found IPv4 interfaces correctly, but gave incorrect values for the
3259 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3260 that it would not match correctly against @[] and not recognize the IPv6
3263 PH/02 The ipliteral router was not recognizing addresses of the form user@
3264 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3266 PH/03 Added disable_ipv6.
3268 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3269 decoded form, because it is most often used to construct To: headers
3270 lines in autoreplies, and the decoded form may well be syntactically
3271 invalid. However, $reply_address has leading white space removed, and all
3272 newlines turned into spaces so that the autoreply transport does not
3275 PH/05 If group was specified without a user on a router, and no group or user
3276 was specified on a transport, the group from the router was ignored.
3278 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3279 for visible compile-time settings that can be used to change these
3280 numbers, for those that want even more. Backwards compatibility with old
3281 spool files has been maintained. However, going back to a previous Exim
3282 release will lost any variables that are in spool files.
3284 PH/07 Two small changes when running in the test harness: increase delay when
3285 passing a TCP/IP connection to a new process, in case the original
3286 process has to generate a bounce, and remove special handling of
3287 127.0.0.2 (sic), which is no longer necessary.
3289 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3290 be the same on different OS.
3292 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3295 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3296 whether --show-vars was specified or not
3298 JJ/02 exipick: Added support for new ACL variable spool format introduced
3301 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3302 syntactically invalid From: or Reply-to: line, and a filter used this to
3303 generate an autoreply, and therefore failed to obtain an address for the
3304 autoreply, Exim could try to deliver to a non-existent relative file
3305 name, causing unrelated and misleading errors. What now happens is that
3306 it logs this as a hard delivery error, but does not attempt to create a
3309 PH/11 The exinext utility has a -C option for testing purposes, but although
3310 the given file was scanned by exinext itself; it wasn't being passed on
3311 when Exim was called.
3313 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3314 an end-of-file indication when reading a command response.
3316 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3317 compiled. In many other places in Exim, IPv6 addresses are always
3318 recognized, so I have changed this. It also means that IPv4 domain
3319 literals of the form [IPV4:n.n.n.n] are now always recognized.
3321 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3322 used if the router is not running as root, for example, when verifying at
3323 ACL time, or when using -bh. The debugging output from this situation was
3324 non-existent - all you got was a failure to exec. I have made two
3327 (a) Failures to set uid/gid, the current directory, or a process leader
3328 in a subprocess such as that created by queryprogram now generate
3329 suitable debugging output when -d is set.
3331 (b) The queryprogram router detects when it is not running as root,
3332 outputs suitable debugging information if -d is set, and then runs
3333 the subprocess without attempting to change uid/gid.
3335 PH/15 Minor change to Makefile for building test_host (undocumented testing
3338 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3339 additional section of a DNS packet that returns MX or SRV records.
3340 Instead, it always explicitly searches for A/AAAA records. This avoids
3341 major problems that occur when a DNS server includes only records of one
3342 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3343 fixed another bug: if SRV records were looked up and the corresponding
3344 address records were *not* found in the additional section, the port
3345 values from the SRV records were lost.
3347 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3348 using the correct key (the original address) when searching the retry
3349 rules in order to find which one to use for generating the retry hint.
3351 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3352 adding the default one. Similarly, if it contains a Reply-To: header, the
3353 errors_reply_to option, if set, is not used.
3355 PH/19 When calculating a retry time, Exim used to measure the "time since
3356 failure" by looking at the "first failed" field in the retry record. Now
3357 it does not use this if it is later than than the arrival time of the
3358 message. Instead it uses the arrival time. This makes for better
3359 behaviour in cases where some deliveries succeed, thus re-setting the
3360 "first failed" field. An example is a quota failure for a huge message
3361 when small messages continue to be delivered. Without this change, the
3362 "time since failure" will always be short, possible causing more frequent
3363 delivery attempts for the huge message than are intended.
3364 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3366 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3367 $1, $2, $3) because the numerical variables can be reset during some
3368 expansion items (e.g. "match"), thereby losing the authentication data.
3370 PH/21 Make -bV show the size of off_t variables so that the test suite can
3371 decide whether to run tests for quotas > 2G.
3373 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3374 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3375 filecount value is greater than 2G or if a quota value is greater than 2G
3376 on a system where the size of off_t is not greater than 4, a panic error
3379 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3380 never match. The debug and -bh output now contains an explicit error
3381 message indicating a malformed IPv4 address or mask.
3383 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3384 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3385 PH/23 above applies.
3387 PH/25 Do not write to syslog when running in the test harness. The only
3388 occasion when this arises is a failure to open the main or panic logs
3389 (for which there is an explicit test).
3391 PH/26 Added the /no_tell option to "control=freeze".
3393 PH/27 If a host name lookup failed very early in a connection, for example, if
3394 the IP address matched host_lookup and the reverse lookup yielded a name
3395 that did not have a forward lookup, an error message of the form "no IP
3396 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3397 could be logged. Now it outputs the IP address instead of "NULL".
3399 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3400 allows the sender and the authenticated sender to be set when
3401 submitting a message from within Exim. Since child_open_exim() is
3402 documented for local_scan(), the new function should be too.
3404 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3405 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3406 results in an empty string is now treated as unset.
3408 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3410 PH/31 Added sender_verify_fail logging option.
3412 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3413 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3414 tidied the source and removed it altogether.
3416 PH/33 When a queue run was abandoned because the load average was too high, a
3417 log line was always written; now it is written only if the queue_run log
3418 selector is set. In addition, the log line for abandonment now contains
3419 information about the queue run such as the pid. This is always present
3420 in "start" and "stop" lines but was omitted from the "abandon" line.
3422 PH/34 Omit spaces between a header name and the colon in the error message that
3423 is given when verify = headers_syntax fails (if there are lots of them,
3424 the message gets confusing).
3426 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3427 names, as there are now some PTR records that contain slashes. This check
3428 is only to protect against broken name servers that fall over on strange
3429 characters, so the fact that it applies to all lookups doesn't matter.
3431 PH/36 Now that the new test suite is complete, we can remove some of the
3432 special code in Exim that was needed for the old test suite. For example,
3433 sorting DNS records because real resolvers return them in an arbitrary
3434 order. The new test suite's fake resolver always returns records in the
3437 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3438 bounces) except when queue_only is set, to avoid logging races between
3439 the different processes.
3441 PH/38 Panic-die if .include specifies a non-absolute path.
3443 PH/39 A tweak to the "H" retry rule from its user.
3445 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3446 a label. They prevented compilation on older perls.
3448 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3449 a warning to be raised on newish perls.
3451 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3452 on queue. Changes to match documented behaviour of showing count of
3453 messages matching specified criteria.
3455 PH/40 Changed the default ident timeout from 30s to 5s.
3457 PH/41 Added support for the use of login_cap features, on those BSD systems
3458 that have them, for controlling the resources used by pipe deliveries.
3460 PH/42 The content-scanning code uses fopen() to create files in which to put
3461 message data. Previously it was not paying any attention to the mode of
3462 the files. Exim runs with umask(0) because the rest of the code creates
3463 files with open(), and sets the required mode explicitly. Thus, these
3464 files were ending up world-writeable. This was not a big issue, because,
3465 being within the spool directory, they were not world-accessible. I have
3466 created a function called modefopen, which takes an additional mode
3467 argument. It sets umask(777), creates the file, chmods it to the required
3468 mode, then resets the umask. All the relevant calls to fopen() in the
3469 content scanning code have been changed to use this function.
3471 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3472 to 24 hours. This avoids potential overflow problems when processing G
3473 and H retry rules. I suspect nobody ever tinkers with this value.
3475 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3477 PH/45 When the plaintext authenticator is running as a client, the server's
3478 challenges are checked to ensure they are valid base64 strings. By
3479 default, the authentication attempt is cancelled if an invalid string is
3480 received. Setting client_ignore_invalid_base64 true ignores these errors.
3481 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3482 they are received. Thus, the responses can be made to depend on the
3483 challenges. If an invalid string is ignored, an empty string is placed in
3486 PH/46 Messages that are created by the autoreply transport now contains a
3487 References: header, in accordance with RFCs 2822 and 3834.
3489 PH/47 Added authenticated_sender_force to the smtp transport.
3491 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3493 PH/49 Installed latest patch from the Sieve maintainer.
3495 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3496 was also set, the appendfile transport was unnecessarily scanning a
3497 directory of message files (e.g. for maildir delivery) to find the count
3498 of files (along with the size), even though it did not need this
3499 information. It now does the scan only if it needs to find either the
3500 size of the count of files.
3502 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3504 PH/52 Two bugs concerned with error handling when the smtp transport is
3507 (i) Exim was not creating retry information for temporary errors given
3508 for individual recipients after the DATA command when the smtp transport
3509 was used in LMTP mode. This meant that they could be retried too
3510 frequently, and not timed out correctly.
3512 (ii) Exim was setting the flag that allows error details to be returned
3513 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3514 recipients that were returned after the DATA command.
3516 PH/53 This is related to PH/52, but is more general: for any failing address,
3517 when detailed error information was permitted to be returned to the
3518 sender, but the error was temporary, then after the final timeout, only
3519 "retry timeout exceeded" was returned. Now it returns the full error as
3520 well as "retry timeout exceeded".
3522 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3523 do this, and (what is worse) MTAs that accept it.
3525 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3526 will now be deprecated.
3528 PH/56 New os.c-cygwin from the Cygwin maintainer.
3530 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3531 formats (previously only available in exim formats via -bpr, -bpru,
3532 and -bpra. Now also available in native and exiqgrep formats)
3534 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3535 with very large, slow to parse queues
3537 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3539 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3541 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3542 responses to authentication challenges, though it was showing the
3543 challenges; (ii) I've removed the CR characters from the debug output for
3546 PH/58 Allow for the insertion of a newline as well as a space when a string
3547 is turned into more than one encoded-word during RFC 2047 encoding. The
3548 Sieve code now uses this.
3550 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3551 data_4xx, lost_connection, tls_required.
3553 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3554 message was being sent as an SMTP response.
3556 PH/61 Add -l and -k options to exicyclog.
3558 PH/62 When verifying, if an address was redirected to one new address, so that
3559 verification continued, and the new address failed or deferred after
3560 having set something in $address_data, the value of $address_data was not
3561 passed back to the ACL. This was different to the case when no
3562 redirection occurred. The value is now passed back in both cases.
3564 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3565 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3566 use login_cap.h, so on its own it isn't the distinguishing feature. The
3567 new name refers directly to the setclassresources() function.
3569 PH/65 Added configuration files for NetBSD3.
3571 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3573 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3574 is preferred over IPv4.
3576 PH/68 The bounce_return_message and bounce_return_body options were not being
3577 honoured for bounces generated during the reception of non-SMTP messages.
3578 In particular, this applied to messages rejected by the ACL. This bug has
3579 been fixed. However, if bounce_return_message is true and bounce_return_
3580 body is false, the headers that are returned for a non-SMTP message
3581 include only those that have been read before the error was detected.
3582 (In the case of an ACL rejection, they have all been read.)
3584 PH/69 The HTML version of the specification is now built in a directory called
3585 spec_html instead of spec.html, because the latter looks like a path with
3586 a MIME-type, and this confuses some software.
3588 PH/70 Catch two compiler warnings in sieve.c.
3590 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3591 function verify_get_ident() calls ip_connect() to connect a socket, but
3592 if the "connect()" function timed out, ip_connect() used to close the
3593 socket. However, verify_get_ident() also closes the socket later, and in
3594 between Exim writes to the log, which may get opened at this point. When
3595 the socket was closed in ip_connect(), the log could get the same file
3596 descriptor number as the socket. This naturally causes chaos. The fix is
3597 not to close the socket in ip_connect(); the socket should be closed by
3598 the function that creates it. There was only one place in the code where
3599 this was missing, in the iplookup router, which I don't think anybody now
3600 uses, but I've fixed it anyway.
3602 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3603 well as to direct DNS lookups. Otherwise the handling of names in host
3604 lists is inconsistent and therefore confusing.
3610 PH/01 Two changes to the default runtime configuration:
3612 (1) Move the checks for relay_from_hosts and authenticated clients from
3613 after to before the (commented out) DNS black list checks.
3615 (2) Add control=submission to the relay_from_hosts and authenticated
3616 clients checks, on the grounds that messages accepted by these
3617 statements are most likely to be submissions.
3619 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3621 (1) Generate an error if the third argument for the ${prvs expansion is
3624 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3627 (3) Reset the variables that are obtained from the first argument of
3628 ${prvscheck and used in the second argument before leaving the code,
3629 because their memory is reclaimed, so using them afterwards may do
3632 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3633 one (it's much easier than Tom thought :-).
3635 (5) Because of (4), we can now allow for the use of $prvscheck_result
3636 inside the third argument.
3638 PH/03 For some reason, the default setting of PATH when running a command from
3639 a pipe transport was just "/usr/bin". I have changed it to
3642 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3643 anything to be listed in the output from -bV.
3645 PH/05 When a filter generated an autoreply, the entire To: header line was
3646 quoted in the delivery log line, like this:
3648 => >A.N.Other <ano@some.domain> <original@ddress> ...
3650 This has been changed so that it extracts the operative address. There
3651 may be more than one such address. If so, they are comma-separated, like
3654 => >ano@some.domain,ona@other.domain <original@ddress> ...
3656 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3657 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3658 not being looked up in the rDNS to get a host name, Exim was showing the
3659 IP address twice in Received: lines, even though the IP addresses were
3660 identical. For example:
3662 Received: from [1.2.3.4] (helo=[1.2.3.4])
3664 However, if the real host name was known, it was omitting the HELO data
3665 if it matched the actual IP address. This has been tidied up so that it
3666 doesn't show the same IP address twice.
3668 PH/07 When both +timestamp and +memory debugging was on, the value given by
3669 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3670 called by the debug printing, thereby overwriting the timestamp buffer.
3671 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3673 PH/08 When the original message was included in an autoreply transport, it
3674 always said "this is a copy of the message, including all the headers",
3675 even if body_only or headers_only was set. It now gives an appropriate
3678 PH/09 Applied a patch from the Sieve maintainer which:
3680 o fixes some comments
3681 o adds the (disabled) notify extension core
3682 o adds some debug output for the result of if/elsif tests
3683 o points to the current vacation draft in the documentation
3684 and documents the missing references header update
3688 o fixes a bug in processing the envelope test (when testing
3689 multiple envelope elements, the last element determined the
3692 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3693 Electronic Mail") by including:
3695 Auto-submitted: auto-generated
3697 in the messages that it generates (bounce messages and others, such as
3698 warnings). In the case of bounce messages for non-SMTP messages, there was
3699 also a typo: it was using "Auto_submitted" (underscore instead of
3700 hyphen). Since every message generated by Exim is necessarily in response
3701 to another message, thes have all been changed to:
3703 Auto-Submitted: auto-replied
3705 in accordance with these statements in the RFC:
3707 The auto-replied keyword:
3709 - SHOULD be used on messages sent in direct response to another
3710 message by an automatic process,
3712 - MUST NOT be used on manually-generated messages,
3714 - MAY be used on Delivery Status Notifications (DSNs) and Message
3715 Disposition Notifications (MDNs),
3717 - MUST NOT be used on messages generated by automatic or periodic
3718 processes, except for messages which are automatic responses to
3721 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3722 to the default Received: header definition.
3724 PH/12 Added log selector acl_warn_skipped (default on).
3726 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3727 variables because (a) they are in the wrong storage pool and (b) even if
3728 they were copied, it wouldn't work properly because of the caching.
3730 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3731 checking when decoding. Apparently there are clients that generate
3732 overlong encoded strings. Why am I not surprised?
3734 PH/15 If the first argument of "${if match_address" was not empty, but did not
3735 contain an "@" character, Exim crashed. Now it writes a panic log message
3736 and treats the condition as false.
3738 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3740 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3741 extension "envelope-auth". The code is finished and in agreement with
3742 other implementations, but there is no documentation so far and in fact,
3743 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3744 not changing the active code.
3746 Print executed "if" and "elsif" statements when debugging is used. This
3747 helps a great deal to understand what a filter does.
3749 Document more things not specified clearly in RFC3028. I had all this
3750 sorted out, when out of a sudden new issues came to my mind. Oops."
3752 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3755 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3756 canonical form (as documented). However, after a host name lookup from
3757 the IP address, check_host() was doing a simple string comparison with
3758 addresses acquired from the DNS when checking that the found name did
3759 have the original IP as one of its addresses. Since any found IPv6
3760 addresses are likely to be in abbreviated form, the comparison could
3761 fail. Luckily, there already exists a function for doing the comparison
3762 by converting both addresses to binary, so now that is used instead of
3763 the text comparison.
3765 PH/20 There was another similar case to PH/19, when a complete host name was
3766 given in a host list; looking up its IP address could give an abbreviated
3767 form, whereas the current host's name might or might not be abbreviated.
3768 The same fix has been applied.
3774 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3775 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3778 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3779 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3781 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3783 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3784 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3785 Also, for "Auto-Submitted", treat anything other than "no" as
3786 non-personal, in accordance with RFC 3834. (Previously it treated
3787 anything starting "auto-" as non-personal.)
3789 TF/01 The control=submission/name=... option had a problem with syntax
3790 errors if the name included a slash character. The /name= option
3791 now slurps the rest of the string, so it can include any characters
3792 but it must come last in the list of options (after /sender_retain
3795 PH/05 Some modifications to the interface to the fake nameserver for the new
3803 TK/01 Added the "success_on_redirect" address verification option. See
3804 NewStuff for rationale and an example.
3806 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3808 PH/02 Patch to exigrep to allow it to work on syslog lines.
3810 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3811 fread() to skip over the body file's header line, because in Cygwin the
3812 header line is locked and is inaccessible.
3814 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3815 co-exist for some time) to make it clear that it is the Exim ID that is
3816 referenced, not the Message-ID: header line.
3818 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3819 string_format() function, because snprintf() does not exist on all
3822 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3823 ${stat: expansion item.
3825 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3826 protocol synchronization error", to keep the pedants happy.
3828 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3829 well as for IRIX systems, when gcc is being used. See the host.c source
3832 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3834 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3837 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3838 transport and to the smtp transport in LMTP mode.
3840 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3842 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3843 run a filter in a subprocess. This could lead to confusion in subsequent
3844 lookups in the parent process. There should also be a search_tidyup() at
3845 the end of the subprocess.
3847 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3848 only if the host matched helo_try_verify_hosts, which caused the
3849 verification to occur when the EHLO/HELO command was issued. The ACL just
3850 tested the remembered result. Now, if a previous verification attempt has
3851 not happened, "verify = helo" does it there and then.
3853 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3855 TK/03 Fix log output including CR from clamd.
3857 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3858 exist provoked a memory error which could cause a segfault.
3860 PH/15 Installed PCRE 6.2
3862 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3864 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3865 of the problem. Specifically, suggested +O2 rather than +O1 for the
3868 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3870 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3871 record, the port was not correctly logged when the outgoing_port log
3872 selector was set (it logged the transort's default port).
3874 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3875 fallback_hosts, and "hosts" in the smtp transport.
3877 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3878 host errors such as "Connection refused".
3880 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3881 authentication with radiusclient 0.4.9:
3883 - Error returned from rc_read_config was caught wrongly
3884 - Username/password not passed on to radius server due to wrong length.
3886 The presumption is that some radiusclient API changes for 4.51/PH/17
3887 were not taken care of correctly. The code is still untested by me (my
3888 Linux distribution still has 0.3.2 of radiusclient), but it was
3889 contributed by a Radius user.
3891 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3892 expanding the "port" option of the smtp transport.
3894 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3895 while reading a MIME header. Thanks to Tom Hughes for a patch.
3897 PH/24 Include config.h inside local_scan.h so that configuration settings are
3900 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3901 that in an ACL for RCPT (for example), you can examine exactly what was
3904 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3905 commands, but it was not correctly comparing the address with the actual
3906 client host address. Thus, it would show the EHLO address in Received:
3907 header lines when this was not necessary.
3909 PH/27 Added the % operator to ${eval:}.
3911 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3912 it should be ignoring failures (because with -C, for example, it has lost
3913 privilege). It wasn't ignoring creation failures other than "already
3916 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3919 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3920 because an input line was too long, either on its own, or by virtue of
3921 too many continuations, the temporary file was not being removed, and the
3922 return code was incorrect.
3924 PH/31 Missing "BOOL" in function definition in filtertest.c.
3926 PH/32 Applied Sieve patches from the maintainer.
3928 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3930 PH/33 Added "verify = not_blind".
3932 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3933 Local/Makefile (with some defaults set). These are used in built scripts
3934 such as exicyclog, but they have never been used in the exim_install
3935 script (though there are many overriding facilities there). I have
3936 arranged that the exim_install script now takes note of these two
3939 PH/35 Installed configuration files for Dragonfly.
3941 PH/36 When a locally submitted message by a trusted user did not contain a
3942 From: header, and the sender address was obtained from -f or from an SMTP
3943 MAIL command, and the trusted user did not use -F to supply a sender
3944 name, $originator_name was incorrectly used when constructing a From:
3945 header. Furthermore, $originator_name was used for submission mode
3946 messages from external hosts without From: headers in a similar way,
3947 which is clearly wrong.
3949 PH/37 Added control=suppress_local_fixups.
3951 PH/38 When log_selector = +received_sender was set, and the addition of the
3952 sender made the log line's construction buffer exactly full, or one byte
3953 less than full, an overflow happened when the terminating "\n" was
3956 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3957 when the result of a list match is failure because a DNS lookup failed.
3959 PH/40 RM_COMMAND is now used in the building process.
3961 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3962 the "build-* directories that it finds.
3964 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3965 address in a domain literal was a prefix of an interface address.
3967 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3968 when verifying a sender address, unless rewrite_headers is false.
3970 PH/44 Wrote a long comment about why errors_to addresses are verified as
3971 recipients, not senders.
3973 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3974 the ratelimit ACL was added.
3976 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3978 PH/46 Added extra information about PostgreSQL errors to the error string.
3980 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3981 avoiding the need to install special zones in a real server. This is
3982 backwards compatible; if it can't find the fake resolver, it drops back.
3983 Thus, both old and new test suites can be run.
3985 TF/02 Added util/ratelimit.pl
3987 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3988 clock is set back in time.
3990 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3991 Candler <B.Candler@pobox.com>.
3993 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3994 OK for addresses that are the result of redirections.
3996 PH/48 A number of further additions for the benefit of the new test suite,
3997 including a fake gethostbyname() that interfaces to the fake DNS resolver
4000 TF/06 The fix for widen_domains has also been applied to qualify_single and
4001 search_parents which are the other dnslookup options that can cause
4004 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
4007 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
4009 TF/07 Exim produced the error message "an SRV record indicated no SMTP
4010 service" if it encountered an MX record with an empty target hostname.
4011 The message is now "an MX or SRV record indicated no SMTP service".
4013 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
4014 if the DNS of the sending site is misconfigured. This is quite a
4015 common situation. This change restores the behaviour of treating a
4016 helo verification defer as a failure.
4018 PH/51 If self=fail was set on a router, the bounce message did not include the
4019 actual error message.
4025 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
4027 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
4028 command itself ended in error, the underlying message about the transport
4029 filter timeout was being overwritten with the pipe command error. Now the
4030 underlying error message should be appended to the second error message.
4032 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4034 PH/02 Reduce the amount of output that "make" produces by default. Full output
4035 can still be requested.
4037 PH/03 The warning log line about a condition test deferring for a "warn" verb
4038 was being output only once per connection, rather than after each
4039 occurrence (because it was using the same function as for successful
4040 "warn" verbs). This seems wrong, so I have changed it.
4042 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4043 it should not have, which might have caused a crash in the right
4044 circumstances, but probably never did.
4046 PH/04 Installed a modified version of Tony Finch's patch to make submission
4047 mode fix the return path as well as the Sender: header line, and to
4048 add a /name= option so that you can make the user's friendly name appear
4051 TF/03 Added the control = fakedefer ACL modifier.
4053 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4054 Mark Lowes for thorough testing.
4056 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4058 TK/03 Merged latest SRS patch from Miles Wilton.
4060 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4061 with the definition in sysexits.h (which is #included earlier).
4062 Fortunately, Exim does not actually use EX_OK. The code used to try to
4063 preserve the sysexits.h value, by assuming that macro definitions were
4064 scanned for macro replacements. I have been disabused of this notion,
4065 so now the code just undefines EX_OK before #including unistd.h.
4067 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4068 in the smtp transport. When a block could not be written in a single
4069 write() function, the timeout was being re-applied to each part-write.
4070 This seems wrong - if the receiver was accepting one byte at a time it
4071 would take for ever. The timeout is now adjusted when this happens. It
4072 doesn't have to be particularly precise.
4074 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4075 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4077 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4078 without a domain if the check to <postmaster@domain> fails.
4080 SC/01 Eximstats: added -xls and the ability to specify output files
4081 (patch written by Frank Heydlauf).
4083 SC/02 Eximstats: use FileHandles for outputting results.
4085 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4087 SC/04 Eximstats: fixed display of large numbers with -nvr option
4089 SC/05 Eximstats: fixed merging of reports with empty tables.
4091 SC/06 Eximstats: added the -include_original_destination flag
4093 SC/07 Eximstats: removed tabs and trailing whitespace.
4095 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4097 TK/06 MBOX spool code: Add real "From " MBOX separator line
4098 so the .eml file is really in mbox format (even though
4099 most programs do not really care). Patch from Alex Miller.
4101 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4102 The latter is generated from $received_to and is only set if the
4103 message has one envelope recipient. SA can use these headers,
4104 obviously out-of-the-box. Patch from Alex Miller.
4106 PH/08 The ${def test on a variable was returning false if the variable's
4107 value was "0", contrary to what the specification has always said!
4108 The result should be true unless the variable is empty.
4110 PH/09 The syntax error of a character other than { following "${if
4111 def:variable_name" (after optional whitespace) was not being diagnosed.
4112 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4113 accidental colon was present, for example, could give incorrect results.
4115 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4116 result is used (not including appendfile, where other changes are about
4119 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4120 This involved changing a lot of size variables from int to off_t. It
4121 should work with maildirs and everything.
4123 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4124 spamd dying while we are connected to it.
4126 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4129 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4130 function for BDB 4.3.
4132 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4134 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4135 the sources for PCRE have been re-arranged and more files are now
4138 PH/15 The code I had for printing potentially long long variables in PH/11
4139 above was not the best (it lost precision). The length of off_t variables
4140 is now inspected at build time, and an appropriate printing format (%ld
4141 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4142 to be "long long int" or "long int". This is needed for the internal
4143 formatting function string_vformat().
4145 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4146 the configuration file to be ":syslog", then the script "guesses" where
4147 the logs files are, rather than using the compiled in default. In our
4148 case the guess is not the same as the compiled default, so the script
4149 suddenly stopped working when I started to use syslog. The patch checks
4150 to see if log_file_path is "". If so, it attempts to read it from exim
4151 with no configuration file to get the compiled in version, before it
4152 falls back to the previous guessing code."
4154 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4155 implementing BATV in an Exim configuration. See NewStuff for the gory
4158 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4159 Makefile that are specific to HP-UX.
4161 PH/18 If the "use_postmaster" option was set for a recipient callout together
4162 with the "random" option, the postmaster address was used as the MAIL
4163 FROM address for the random test, but not for the subsequent recipient
4164 test. It is now used for both.
4166 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4167 patch removes a few documentation additions to RFC 3028, because the
4168 latest draft now contains them. It adds the new en;ascii-case comparator
4169 and a new error check for 8bit text in MIME parts. Comparator and
4170 require names are now matched exactly. I enabled the subaddress
4171 extension, but it is not well tested yet (read: it works for me)."
4173 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4174 rework some of the code of TK/09 above to avoid the hardwired use of
4175 "%lld" and "long long". Replaced the call to snprintf() with a call to
4178 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4179 records point to non-existent hosts", "retry timeout exceeded", and
4180 "retry time not reached for any host after a long failure period".
4182 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4183 experimental DomainKeys support:
4185 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4186 (2) On an error such as an illegally used "control", the wrong name for
4187 the control was given.
4189 These problems did NOT occur unless DomainKeys support was compiled.
4191 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4193 PH/24 Added ${if match_ip condition.
4195 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4196 databases so that it will be absolutely obvious if a crash occurs in the
4197 DB library. This is a regular occurrence (often caused by mis-matched
4200 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4201 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4202 picked up on a user's system that detects such things. There doesn't seem
4203 to be a gcc warning option for this - only an attribute that has to be
4204 put on the function's prototype. It seems that in Fedora Core 4 they have
4205 set this on a number of new functions. No doubt there will be more in due
4208 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4209 specify a transport. However, if an address that was verified by such a
4210 router was the subject of a callout, Exim crashed because it tried to
4211 read the rcpt_include_affixes from the non-existent transport. Now it
4212 just assumes that the setting of that option is false. This bug was
4213 introduced by 4.51/PH/31.
4215 PH/28 Changed -d+all to exclude +memory, because that information is very
4216 rarely of interest, but it makes the output a lot bigger. People tend to
4217 do -d+all out of habit.
4219 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4220 code in os-type was giving problems when libc.so lives in lib64, like on
4223 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4224 aren't the modern standard, and it seems that some systems' include files
4225 don't always have them. Exim was already checking for some of the newer
4226 ones like T_AAAA, and defining it itself. I've added checks for all the
4227 record types that Exim uses.
4229 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4230 not automatically generating a new one, as it is supposed to. This
4231 prevented TLS from working. If the file did exist, but contained invalid
4232 data, a new version was generated, as expected. It was only the case of a
4233 non-existent file that was broken.
4235 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4236 with a change in libdomainkeys > 0.64.
4238 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4239 from DNS. If the selector record carries the flag, it now has
4240 precedence over the domain-wide flag.
4242 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4244 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4245 the use of a port name that isn't defined in /etc/services) occurred, the
4246 message was deferred as in a normal delivery, and thus remained on the
4247 spool, instead of being failed because of the mua_wrapper setting. This
4248 is now fixed, and I tidied up some of the mua_wrapper messages at the
4251 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4252 the messages in a hash of arrays rather than using individual hashes.
4253 This is a bit cleaner and results in dramatic memory savings, albeit
4254 at a slight CPU cost.
4256 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4257 as requested by Marc Sherman.
4259 SC/10 Eximstats: added histograms for user specified patterns as requested
4262 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4264 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4265 fopen() in the content-scanning modules that did not already have it.
4271 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4272 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4274 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4276 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4278 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4279 to test Sieve filters that use "vacation".
4281 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4282 that changes the way the GnuTLS parameters are stored in the cache file.
4283 The new format can be generated externally. For backward compatibility,
4284 if the data in the cache doesn't make sense, Exim assumes it has read an
4285 old-format file, and it generates new data and writes a new file. This
4286 means that you can't go back to an older release without removing the
4289 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4290 work if there are any delivery delays because "one_time" forces the
4291 parent to be marked "delivered", so its unseen clone is never tried
4292 again. For this reason, Exim now forbids the simultaneous setting of
4295 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4296 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4297 entry if you want to know the details. The fix, however, neglected to
4298 consider the case when local delivery batching is involved. The test for
4299 "previously delivered" was not happening when checking to see if an
4300 address could be batched with a previous (undelivered) one; under
4301 certain circumstances this could lead to multiple deliveries to the same
4304 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4305 in its include files, and this causes problems building Exim.
4307 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4308 header_syntax) but Exim was just ignoring anything given after a slash.
4309 In particular, this caused confusion with an attempt to use "verify =
4310 reverse_host_lookup/defer_ok". An error is now given when options are
4311 supplied for verify items that do not have them. (Maybe reverse_host_
4312 lookup should have a defer_ok option, but that's a different point.)
4314 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4315 defined by RFC 821) to 2048, because there were problems with some AUTH
4316 commands, and RFC 1869 says the size should be increased for extended
4317 SMTP commands that take arguments.
4319 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4322 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4323 "unknown" error; now it says that the functionality isn't in the binary.
4325 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4326 an address' error message when a string expansion fails (syntax or
4327 whatever). Otherwise the password may appear in the log. Following change
4328 PH/42 below, there is no longer a chance of it appearing in a bounce
4331 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4333 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4334 of its addresses was ever tried. (Bugzilla bug #2.)
4336 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4337 the result incorrectly in the debug output. (It correctly added a newline
4338 to what was transported.)
4340 TF/01 Added $received_time.
4342 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4343 commented out examples of how to interface to a virus scanner and to
4344 SpamAssassin. Also added commented examples of av_scanner and
4345 spamd_address settings.
4347 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4348 and controls are allowed in which ACLs. There were a couple of minor
4349 errors. Some of the entries in the conditions table (which is a table of
4350 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4351 negation of where the condition IS allowed.
4353 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4355 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4356 header file does not have a version number, so I've had to invent a new
4357 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4358 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4359 radiusclient), but it was contributed by a Radius user.
4361 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4362 files or directories, for OpenSSL.
4364 PH/19 When an Exim process that is running as root has to create an Exim log
4365 file, it does so in a subprocess that runs as exim:exim so as to get the
4366 ownership right at creation (otherwise, other Exim processes might see
4367 the file with the wrong ownership). There was no test for failure of this
4368 fork() call, which would lead to the process getting stuck as it waited
4369 for a non-existent subprocess. Forks do occasionally fail when resources
4370 run out. I reviewed all the other calls to fork(); they all seem to check
4373 PH/20 When checking for unexpected SMTP input at connect time (before writing
4374 the banner), Exim was not dealing correctly with a non-positive return
4375 from the read() function. If the client had disconnected by this time,
4376 the result was a log entry for a synchronization error with an empty
4377 string after "input=" when read() returned zero. If read() returned -1
4378 (an event I could not check), uninitialized data bytes were printed.
4379 There were reports of junk text (parts of files, etc) appearing after
4382 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4384 PH/22 Added support for macro redefinition, and (re)definition in between
4385 driver and ACL definitions.
4387 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4388 forgetting to use the resulting value; it was using the unexpanded value.
4390 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4391 hadn't been configured. The fix is from Juergen Kreileder, who
4392 understands it better than I do:
4394 "Here's what I see happening with three configured cyrus_sasl
4395 authenticators configured (plain, login, cram-md5):
4397 On startup auth_cyrus_sasl_init() gets called for each of these.
4398 This means three calls to sasl_listmech() without a specified mech_list.
4399 => SASL tests which mechs of all available mechs actually work
4400 => three warnings about OTP not working
4401 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4403 With the patch, sasl_listmech() also gets called three times. But now
4404 SASL's mech_list option is set to the server_mech specified in the the
4405 authenticator. Or in other words, the answer from sasl_listmech()
4406 gets limited to just the mech you're testing for (which is different
4408 => the return list contains just 'plain' or 'login', 'cram-md5' or
4409 nothing depending on the value of ob->server_mech.
4411 I've just tested the patch: Authentication still works fine,
4412 unavailable mechs specified in the exim configuration are still
4413 caught, and the auth.log warnings about OTP are gone."
4415 PH/25 When debugging is enabled, the contents of the command line are added
4416 to the debugging output, even when log_selector=+arguments is not
4419 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4420 answer is "GNU", and only if the return is "GNU/something" is the answer
4423 PH/27 $acl_verify_message is now set immediately after the failure of a
4424 verification in an ACL, and so is available in subsequent modifiers. In
4425 particular, the message can be preserved by coding like this:
4427 warn !verify = sender
4428 set acl_m0 = $acl_verify_message
4430 Previously, $acl_verify_message was set only while expanding "message"
4431 and "log_message" when a very denied access.
4433 PH/28 Modified OS/os.c-Linux with
4435 -#ifndef OS_LOAD_AVERAGE
4436 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4438 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4439 nomenclature these days.)
4441 PH/29 Installed patch from the Sieve maintainer that adds the options
4442 sieve_useraddress and sieve_subaddress to the redirect router.
4444 PH/30 In these circumstances:
4445 . Two addresses routed to the same list of hosts;
4446 . First host does not offer TLS;
4447 . First host accepts first address;
4448 . First host gives temporary error to second address;
4449 . Second host offers TLS and a TLS session is established;
4450 . Second host accepts second address.
4451 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4452 and peerdn, if requested) that were in fact used only for the second
4455 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4456 attention to any local part prefix or suffix that was matched by the
4457 router that accepted the address. It now behaves in the same way as it
4458 does for delivery: the affixes are removed from the local part unless
4459 rcpt_include_affixes is set on the transport.
4461 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4462 timeout during the DATA phase of an incoming message.
4464 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4465 applied a patch sanctioned by the Sieve maintainer.
4467 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4468 the uid or gid is negative. A case of a negative gid caused this to be
4469 noticed. The fix allows for either to be negative.
4471 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4472 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4475 PH/36 The change PH/12 above was broken. Fixed it.
4477 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4478 the grounds that routing the same address twice would always produce the
4479 same answer. This might have been true once, but it is certainly no
4480 longer true now. Routing a child address may depend on the previous
4481 routing that produced that child. Some complicated redirection strategies
4482 went wrong when messages had multiple recipients, and made Exim's
4483 behaviour dependent on the order in which the addresses were given.
4485 I have moved the duplicate checking until after the routing is complete.
4486 Exim scans the addresses that are assigned to local and remote
4487 transports, and removes any duplicates. This means that more work will be
4488 done, as duplicates will always all be routed, but duplicates are
4489 presumably rare, so I don't expect this is of any significance.
4491 For deliveries to pipes, files, and autoreplies, the duplicate checking
4492 still happens during the routing process, since they are not going to be
4495 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4496 It corrects a timeout issue with spamd. This is Ian's comment: "The
4497 background is that sometimes spamd either never reads data from a
4498 connection it has accepted, or it never writes response data. The exiscan
4499 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4500 blindly assumes that writes won't block so it may never time out."
4502 PH/39 Allow G after quota size as well as K and M.
4504 PH/40 The value set for $authenticated_id in an authenticator may not contain
4505 binary zeroes or newlines because the value is written to log lines and
4506 to spool files. There was no check on this. Now the value is run through
4507 the string_printing() function so that such characters are converted to
4508 printable escape sequences.
4510 PH/41 $message_linecount is a new variable that contains the total number of
4511 lines in the message. Compare $body_linecount, which is the count for the
4514 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4515 bounce and delay warning messages, except in certain special cases, which
4518 (a) An SMTP error message from a remote host;
4519 (b) A message specified in a :fail: redirection;
4520 (c) A message specified in a "fail" command in a system filter;
4521 (d) A message specified in a FAIL return from the queryprogram router;
4522 (e) A message specified by the cannot_route_message router option.
4524 In these cases only, Exim does include the error details in bounce and
4525 warning messages. There are also a few cases where bland messages such
4526 as "unrouteable address" or "local delivery error" are given.
4528 PH/43 $value is now also set for the "else" part of a ${run expansion.
4530 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4531 being worked on, but at least Exim now implements the latest version to
4534 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4535 process to complete was treated as a delivery failure, a timeout while
4536 writing the message to the pipe was logged, but erroneously treated as a
4537 successful delivery. Such timeouts include transport filter timeouts. For
4538 consistency with the overall process timeout, these timeouts are now
4539 treated as errors, giving rise to delivery failures by default. However,
4540 there is now a new Boolean option for the pipe transport called
4541 timeout_defer, which, if set TRUE, converts the failures into defers for
4542 both kinds of timeout. A transport filter timeout is now identified in
4545 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4546 systems where "make" and "gmake" are different, calling "gmake" at top
4547 level broke things. I've arranged for the value of $(MAKE) to be passed
4548 from the Makefile to this script so that it can call the same version of
4552 A note about Exim versions 4.44 and 4.50
4553 ----------------------------------------
4555 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4556 changes of various kinds. As a consequence, a big documentation update was
4557 needed. This delayed the release for rather longer than seemed good, especially
4558 in the light of a couple of (minor) security issues. Therefore, the changes
4559 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4560 release. So 4.44 and 4.50 are in effect two different branches that both start
4563 I have left the 4.50 change log unchanged; it contains all the changes since
4564 4.43. The change log for 4.44 is below; many of its items are identical to
4565 those for 4.50. This seems to be the most sensible way to preserve the
4566 historical information.
4572 1. Minor wording change to the doc/README.SIEVE file.
4574 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4575 computation of the current number of files was incorrect.
4577 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4578 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4581 4. Give more explanation in the error message when the command for a transport
4582 filter fails to execute.
4584 5. There are several places where Exim runs a non-Exim command in a
4585 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4586 was being done only for the command run by the queryprogram router. It is
4587 now done for all such subprocesses. The other cases are: ${run, transport
4588 filters, and the commands run by the lmtp and pipe transports.
4590 6. Added CONFIGURE_GROUP build-time option.
4592 7. Some older OS have a limit of 256 on the maximum number of file
4593 descriptors. Exim was using setrlimit() to set 1000 as a large value
4594 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4595 systems. I've change it so that if it can't get 1000, it tries for 256.
4597 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4598 was an oversight, and furthermore, ever since the addition of extra
4599 controls (e.g. 4.43/32), the checks on when to allow different forms of
4600 "control" were broken. There should now be diagnostics for all cases when a
4601 control that does not make sense is encountered.
4603 9. Added the /retain_sender option to "control=submission".
4605 10. $recipients is now available in the predata ACL (oversight).
4607 11. Tidy the search cache before the fork to do a delivery from a message
4608 received from the command line. Otherwise the child will trigger a lookup
4609 failure and thereby defer the delivery if it tries to use (for example) a
4610 cached ldap connection that the parent has called unbind on.
4612 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4613 of $address_data from the recipient verification was clobbered by the
4614 sender verification.
4616 13. The value of address_data from a sender verification is now available in
4617 $sender_address_data in subsequent conditions in the ACL statement.
4619 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4621 15. Added a new option "connect=<time>" to callout options, to set a different
4624 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4625 was its contents. (It was OK if the option was not defined at all.)
4627 17. A "Completed" log line is now written for messages that are removed from
4628 the spool by the -Mrm option.
4630 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4631 information about exactly what failed.
4633 19. Added -dd to debug only the daemon process.
4635 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4636 handles timeouts, both on the server side and network timeouts. Renamed the
4637 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4639 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4640 It is now set to "smtps".
4642 22. $host_address is now set to the target address during the checking of
4643 ignore_target_hosts.
4645 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4646 being passed; this would have caused $sender_host_name to have been used if
4647 matching the list had actually called for a host name (not very likely,
4648 since this list is usually IP addresses). A host name is now passed as
4651 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4652 code that specifies a non-restarting handler (typically sigaction() in
4653 modern systems) in an attempt to fix a rare and obscure crash bug.
4655 25. Narrowed the window for a race in the daemon that could cause it to ignore
4656 SIGCHLD signals. This is not a major problem, because they are used only to
4657 wake it up if nothing else does.
4659 26. A malformed maildirsize file could cause Exim to calculate negative values
4660 for the mailbox size or file count. Odd effects could occur as a result.
4661 The maildirsize information is now recalculated if the size or filecount
4664 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4665 support for a long time. Removed HAVE_SYS_VFS_H.
4667 28. Installed the latest version of exipick from John Jetmore.
4669 29. In an address list, if the pattern was not a regular expression, an empty
4670 subject address (from a bounce message) matched only if the pattern was an
4671 empty string. Non-empty patterns were not even tested. This was the wrong
4672 because it is perfectly reasonable to use an empty address as part of a
4673 database query. An empty address is now tested by patterns that are
4674 lookups. However, all the other forms of pattern expect the subject to
4675 contain a local part and a domain, and therefore, for them, an empty
4676 address still always fails if the pattern is not itself empty.
4678 30. Exim went into a mad DNS loop when attempting to do a callout where the
4679 host was specified on an smtp transport, and looking it up yielded more
4680 than one IP address.
4682 31. Re-factored the code for checking spool and log partition space into a
4683 function that finds that data and another that does the check. The former
4684 is then used to implement four new variables: $spool_space, $log_space,
4685 $spool_inodes, and $log_inodes.
4687 32. The RFC2047 encoding function was originally intended for short strings
4688 such as real names; it was not keeping to the 75-character limit for
4689 encoded words that the RFC imposes. It now respects the limit, and
4690 generates multiple encoded words if necessary. To be on the safe side, I
4691 have increased the buffer size for the ${rfc2047: expansion operator from
4694 33. It is now permitted to omit both strings after an "if" condition; if the
4695 condition is true, the result is "true". As before, when the second string
4696 is omitted, a false condition yields an empty string. This makes it less
4697 cumbersome to write custom ACL and router conditions.
4699 34. Failure to deliver a bounce message always caused it to be frozen, even if
4700 there was an errors_to setting on the router. The errors_to setting is now
4703 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4704 canonical form (fully expanded) before being placed in
4705 $sender_host_address.
4707 36. The table in the code that translates DNS record types into text (T_A to
4708 "A" for instance) was missing entries for NS and CNAME. It is just possible
4709 that this could have caused confusion if both these types were looked up
4710 for the same domain, because the text type is used as part of Exim's
4711 per-process caching. But the chance of anyone hitting this buglet seems
4714 37. The dnsdb lookup has been extended in a number of ways.
4716 (1) There is a new type, "zns", which walks up the domain tree until it
4717 finds some nameserver records. It should be used with care.
4719 (2) There is a new type, "mxh", which is like "mx" except that it returns
4720 just the host names, not the priorities.
4722 (3) It is now possible to give a list of domains (or IP addresses) to be
4723 looked up. The behaviour when one of the lookups defers can be
4724 controlled by a keyword.
4726 (4) It is now possible to specify the separator character for use when
4727 multiple records are returned.
4729 38. The dnslists ACL condition has been extended: it is now possible to supply
4730 a list of IP addresses and/or domains to be looked up in a particular DNS
4733 39. Added log_selector=+queue_time_overall.
4735 40. When running the queue in the test harness, wait just a tad after forking a
4736 delivery process, to get repeatability of debugging output.
4738 41. Include certificate and key file names in error message when GnuTLS fails
4739 to set them up, because the GnuTLS error message doesn't include the name
4740 of the failing file when there is a problem reading it.
4742 42. Allow both -bf and -bF in the same test run.
4744 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4746 44. The "Exiscan patch" is now merged into the mainline Exim source.
4748 45. Sometimes the final signoff response after QUIT could fail to get
4749 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4750 < 0 before doing a fflush(). This bug looks as though it goes back to the
4751 introduction of TLS in release 3.20, but "sometimes" must have been rare
4752 because the tests only now provoked it.
4754 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4755 (this can affect the format of dates).
4757 47. exim_tidydb, when checking for the continued existence of a message for
4758 which it has found a message-specific retry record, was not finding
4759 messages that were in split spool directories. Consequently, it was
4760 deleting retry records that should have stayed in existence.
4762 48. Steve fixed some bugs in eximstats.
4764 49. The SPA authentication driver was not abandoning authentication and moving
4765 on to the next authenticator when an expansion was forced to fail,
4766 contradicting the general specification for all authenticators. Instead it
4767 was generating a temporary error. It now behaves as specified.
4769 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4770 (the order specifies the preference for clients). The order is now AES256,
4771 AES128, 3DES, ARCFOUR128.
4773 51. Small patch to Sieve code - explicitly set From: when generating an
4776 52. Exim crashed if a remote delivery caused a very long error message to be
4777 recorded - for instance if somebody sent an entire SpamAssassin report back
4778 as a large number of 550 error lines. This bug was coincidentally fixed by
4779 increasing the size of one of Exim's internal buffers (big_buffer) that
4780 happened as part of the Exiscan merge. However, to be on the safe side, I
4781 have made the code more robust (and fixed the comments that describe what
4784 53. Now that there can be additional text after "Completed" in log lines (if
4785 the queue_time_overall log selector is set), a one-byte patch to exigrep
4786 was needed to allow it to recognize "Completed" as not the last thing in
4789 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4790 patch that reportedly fixes this has been added. I am not expert enough to
4791 create a test for it. This is what the patch creator wrote:
4793 "I found a little strange behaviour of ldap code when working with
4794 Windows 2003 AD Domain, where users was placed in more than one
4795 Organization Units. When I tried to give exim partial DN, the exit code
4796 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4797 But simultaneously result of request was absolutely normal ldap result,
4798 so I produce this patch..."
4800 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4801 REFERENCE, so I have modified the code to exclude the patch when that macro
4804 55. Some experimental protocols are using DNS PTR records for new purposes. The
4805 keys for these records are domain names, not reversed IP addresses. The
4806 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4807 leaves it alone. Component reversal etc. now happens only for IP addresses.
4810 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4812 57. Double the size of the debug message buffer (to 2048) so that more of very
4813 long debug lines gets shown.
4815 58. The exicyclog utility now does better if the number of log files to keep
4816 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4818 59. Two changes related to the smtp_active_hostname option:
4820 (1) $smtp_active_hostname is now available as a variable.
4821 (2) The default for smtp_banner uses $smtp_active_hostname instead
4822 of $primary_hostname.
4824 60. The host_aton() function is supposed to be passed a string that is known
4825 to be a valid IP address. However, in the case of IPv6 addresses, it was
4826 not checking this. This is a hostage to fortune. Exim now panics and dies
4827 if the condition is not met. A case was found where this could be provoked
4828 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4829 components; fortuitously, this particular loophole had already been fixed
4830 by change 4.50/55 above.
4832 If there are any other similar loopholes, the new check in host_aton()
4833 itself should stop them being exploited. The report I received stated that
4834 data on the command line could provoke the exploit when Exim was running as
4835 exim, but did not say which command line option was involved. All I could
4836 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4837 running as the user.
4840 61. There was a buffer overflow vulnerability in the SPA authentication code
4841 (which came originally from the Samba project). I have added a test to the
4842 spa_base64_to_bits() function which I hope fixes it.
4845 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4846 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4848 63. The daemon start-up calls getloadavg() while still root for those OS that
4849 need the first call to be done as root, but it missed one case: when
4850 deliver_queue_load_max is set with deliver_drop_privilege. This is
4851 necessary for the benefit of the queue runner, because there is no re-exec
4852 when deliver_drop_privilege is set.
4854 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4855 This has been fixed.
4857 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4858 was in use, was not putting the data itself into the right store pool;
4859 consequently, it could be overwritten for a subsequent message in the same
4860 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4863 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4865 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4866 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4867 the function were treating the return as a boolean value, which happened to
4868 work because 0=false and not-0=true, but is not correct code.
4870 68. The host_aton() function was not handling scoped IPv6 addresses (those
4871 with, for example, "%eth0" on the end) correctly.
4873 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4874 negated items (that is, ~something) in unsigned ints. Some compilers
4875 apparently mutter when there is no cast.
4877 70. If an address verification called from an ACL failed, and did not produce a
4878 user-specific message (i.e. there was only a "system" message), nothing was
4879 put in $acl_verify_message. In this situation, it now puts the system
4882 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4883 session; change 4.31/43 added the unwanted input to the log line - except
4884 that it did not do this in the start of session case. It now does.
4886 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4887 This is wrong and can cause the other end to generate a synchronization
4888 error if it is another Exim or anything else that does the synchronization
4889 check. A QUIT command is no longer sent after a timeout.
4891 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4892 during host lookups.
4894 74. The defer_ok option of callout verification was not working if it was used
4895 when verifying addresses in header lines, that is, for this case:
4897 verify = header_sender/callout=defer_ok
4899 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4900 those file descriptors could be used for SMTP connections. If anything
4901 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4902 could be sent to the SMTP client, causing chaos. The daemon now opens
4903 stdin, stdout, and stderr to /dev/null when it puts itself into the
4906 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4907 log by default. The user can override this with suitable Perl magic.
4909 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4910 the log message when discard triggers, was not working for the DATA ACL or
4911 for the non-SMTP ACL.
4913 78. Error message wording change in sieve.c.
4915 79. If smtp_accept_max_per_host was set, the number of connections could be
4916 restricted to fewer than expected, because the daemon was trying to set up
4917 a new connection before checking whether the processes handling previous
4918 connections had finished. The check for completed processes is now done
4919 earlier. On busy systems, this bug wouldn't be noticed because something
4920 else would have woken the daemon, and it would have reaped the completed
4923 80. If a message was submitted locally by a user whose login name contained one
4924 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4925 It caused a spool format error. I have fixed the spool reading code. A
4926 related problem was that the "from" clause in the Received: line became
4927 illegal because of the space(s). It is now covered by ${quote_local_part.
4929 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4932 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4934 83. Patch from Sieve maintainer to fix unterminated string problem in
4935 "vacation" handling.
4937 84. Some minor changes to the Linux configuration files to help with other
4938 OS variants using glibc.
4940 85. One more patch for Sieve to update vacation handling to latest spec.
4943 ----------------------------------------------------
4944 See the note above about the 4.44 and 4.50 releases.
4945 ----------------------------------------------------
4951 1. Change 4.43/35 introduced a bug that caused file counts to be
4952 incorrectly computed when quota_filecount was set in an appendfile
4955 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4956 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4959 3. Give more explanation in the error message when the command for a transport
4960 filter fails to execute.
4962 4. There are several places where Exim runs a non-Exim command in a
4963 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4964 was being done only for the command run by the queryprogram router. It is
4965 now done for all such subprocesses. The other cases are: ${run, transport
4966 filters, and the commands run by the lmtp and pipe transports.
4968 5. Some older OS have a limit of 256 on the maximum number of file
4969 descriptors. Exim was using setrlimit() to set 1000 as a large value
4970 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4971 systems. I've change it so that if it can't get 1000, it tries for 256.
4973 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4974 was an oversight, and furthermore, ever since the addition of extra
4975 controls (e.g. 4.43/32), the checks on when to allow different forms of
4976 "control" were broken. There should now be diagnostics for all cases when a
4977 control that does not make sense is encountered.
4979 7. $recipients is now available in the predata ACL (oversight).
4981 8. Tidy the search cache before the fork to do a delivery from a message
4982 received from the command line. Otherwise the child will trigger a lookup
4983 failure and thereby defer the delivery if it tries to use (for example) a
4984 cached ldap connection that the parent has called unbind on.
4986 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4987 of $address_data from the recipient verification was clobbered by the
4988 sender verification.
4990 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4991 was its contents. (It was OK if the option was not defined at all.)
4993 11. A "Completed" log line is now written for messages that are removed from
4994 the spool by the -Mrm option.
4996 12. $host_address is now set to the target address during the checking of
4997 ignore_target_hosts.
4999 13. When checking ignore_target_hosts for an ipliteral router, no host name was
5000 being passed; this would have caused $sender_host_name to have been used if
5001 matching the list had actually called for a host name (not very likely,
5002 since this list is usually IP addresses). A host name is now passed as
5005 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5006 code that specifies a non-restarting handler (typically sigaction() in
5007 modern systems) in an attempt to fix a rare and obscure crash bug.
5009 15. Narrowed the window for a race in the daemon that could cause it to ignore
5010 SIGCHLD signals. This is not a major problem, because they are used only to
5011 wake it up if nothing else does.
5013 16. A malformed maildirsize file could cause Exim to calculate negative values
5014 for the mailbox size or file count. Odd effects could occur as a result.
5015 The maildirsize information is now recalculated if the size or filecount
5018 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5019 support for a long time. Removed HAVE_SYS_VFS_H.
5021 18. Updated exipick to current release from John Jetmore.
5023 19. Allow an empty sender to be matched against a lookup in an address list.
5024 Previously the only cases considered were a regular expression, or an
5027 20. Exim went into a mad DNS lookup loop when doing a callout where the
5028 host was specified on the transport, if the DNS lookup yielded more than
5031 21. The RFC2047 encoding function was originally intended for short strings
5032 such as real names; it was not keeping to the 75-character limit for
5033 encoded words that the RFC imposes. It now respects the limit, and
5034 generates multiple encoded words if necessary. To be on the safe side, I
5035 have increased the buffer size for the ${rfc2047: expansion operator from
5038 22. Failure to deliver a bounce message always caused it to be frozen, even if
5039 there was an errors_to setting on the router. The errors_to setting is now
5042 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5043 canonical form (fully expanded) before being placed in
5044 $sender_host_address.
5046 24. Updated eximstats to version 1.33
5048 25. Include certificate and key file names in error message when GnuTLS fails
5049 to set them up, because the GnuTLS error message doesn't include the name
5050 of the failing file when there is a problem reading it.
5052 26. Expand error message when OpenSSL has problems setting up cert/key files.
5055 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5056 (this can affect the format of dates).
5058 28. exim_tidydb, when checking for the continued existence of a message for
5059 which it has found a message-specific retry record, was not finding
5060 messages that were in split spool directories. Consequently, it was
5061 deleting retry records that should have stayed in existence.
5063 29. eximstats updated to version 1.35
5064 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5065 1.35 - bugfix such that pie charts by volume are generated correctly
5067 30. The SPA authentication driver was not abandoning authentication and moving
5068 on to the next authenticator when an expansion was forced to fail,
5069 contradicting the general specification for all authenticators. Instead it
5070 was generating a temporary error. It now behaves as specified.
5072 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5073 (the order specifies the preference for clients). The order is now AES256,
5074 AES128, 3DES, ARCFOUR128.
5076 31. Small patch to Sieve code - explicitly set From: when generating an
5079 32. Exim crashed if a remote delivery caused a very long error message to be
5080 recorded - for instance if somebody sent an entire SpamAssassin report back
5081 as a large number of 550 error lines. This bug was coincidentally fixed by
5082 increasing the size of one of Exim's internal buffers (big_buffer) that
5083 happened as part of the Exiscan merge. However, to be on the safe side, I
5084 have made the code more robust (and fixed the comments that describe what
5087 33. Some experimental protocols are using DNS PTR records for new purposes. The
5088 keys for these records are domain names, not reversed IP addresses. The
5089 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5090 leaves it alone. Component reversal etc. now happens only for IP addresses.
5093 34. The host_aton() function is supposed to be passed a string that is known
5094 to be a valid IP address. However, in the case of IPv6 addresses, it was
5095 not checking this. This is a hostage to fortune. Exim now panics and dies
5096 if the condition is not met. A case was found where this could be provoked
5097 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5098 components; fortuitously, this particular loophole had already been fixed
5099 by change 4.50/55 or 4.44/33 above.
5101 If there are any other similar loopholes, the new check in host_aton()
5102 itself should stop them being exploited. The report I received stated that
5103 data on the command line could provoke the exploit when Exim was running as
5104 exim, but did not say which command line option was involved. All I could
5105 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5106 running as the user.
5109 35. There was a buffer overflow vulnerability in the SPA authentication code
5110 (which came originally from the Samba project). I have added a test to the
5111 spa_base64_to_bits() function which I hope fixes it.
5114 36. The daemon start-up calls getloadavg() while still root for those OS that
5115 need the first call to be done as root, but it missed one case: when
5116 deliver_queue_load_max is set with deliver_drop_privilege. This is
5117 necessary for the benefit of the queue runner, because there is no re-exec
5118 when deliver_drop_privilege is set.
5120 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5121 was in use, was not putting the data itself into the right store pool;
5122 consequently, it could be overwritten for a subsequent message in the same
5123 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5126 38. Sometimes the final signoff response after QUIT could fail to get
5127 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5128 < 0 before doing a fflush(). This bug looks as though it goes back to the
5129 introduction of TLS in release 3.20, but "sometimes" must have been rare
5130 because the tests only now provoked it.
5136 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5137 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5138 Now it returns an integer. A number of calls were still expecting a T/F
5139 return. Fortuitously, in all cases, the tests worked in OK situations,
5140 which is the norm. However, things would have gone wrong on any write
5141 failures on the smtp file descriptor. This function is used when sending
5142 messages over SMTP and also when doing verify callouts.
5144 2. When Exim is called to do synchronous delivery of a locally submitted
5145 message (the -odf or -odi options), it no longer closes stderr before doing
5148 3. Implemented the mua_wrapper option.
5150 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5152 5. Implemented the functions header_remove(), header_testname(),
5153 header_add_at_position(), and receive_remove_recipient(), and exported them
5156 6. If an ACL "warn" statement specified the addition of headers, Exim already
5157 inserted X-ACL-Warn: at the start if there was no header name. However, it
5158 was not making this test for the second and subsequent header lines if
5159 there were newlines in the string. This meant that an invalid header could
5160 be inserted if Exim was badly configured.
5162 7. Allow an ACL "warn" statement to add header lines at the start or after all
5163 the Received: headers, as well as at the end.
5165 8. Added the rcpt_4xx retry error code.
5167 9. Added postmaster_mailfrom=xxx to callout verification option.
5169 10. Added mailfrom=xxxx to the callout verification option, for verify=
5172 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5173 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5175 12. Inserted some casts to stop certain compilers warning when using pointer
5176 differences as field lengths or precisions in printf-type calls (mostly
5177 affecting debugging statements).
5179 13. Added optional readline() support for -be (dynamically loaded).
5181 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5182 same clock tick as a message's arrival, so that its received time was the
5183 same as the "first fail" time on the retry record, and that message
5184 remained on the queue past the ultimate address timeout, every queue runner
5185 would try a delivery (because it was past the ultimate address timeout) but
5186 after another failure, the ultimate address timeout, which should have then
5187 bounced the address, did not kick in. This was a "< instead of <=" error;
5188 in most cases the first failure would have been in the next clock tick
5189 after the received time, and all would be well.
5191 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5192 being recognized when the domain list was tested by the match_domain
5193 condition in an expansion string.
5195 16. Added the ${str2b64: operator.
5197 17. Exim was always calling setrlimit() to set a large limit for the number of
5198 processes, without checking whether the existing limit was already
5199 adequate. (It did check for the limit on file descriptors.) Furthermore,
5200 errors from getrlimit() and setrlimit() were being ignored. Now they are
5201 logged to the main and panic logs, but Exim does carry on, to try to do its
5202 job under whatever limits there are.
5204 18. Imported PCRE 5.0.
5206 19. Trivial typo in log message " temporarily refused connection" (the leading
5209 20. If the log selector return_path_on_delivery was set and an address was
5210 redirected to /dev/null, the delivery process crashed because it assumed
5211 that a return path would always be set for a "successful" delivery. In this
5212 case, the whole delivery is bypassed as an optimization, and therefore no
5215 21. Internal re-arrangement: the function for sending a challenge and reading
5216 a response while authentication was assuming a zero-terminated challenge
5217 string. It's now changed to take a pointer and a length, to allow for
5218 binary data in such strings.
5220 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5222 23. Exim was not respecting finduser_retries when seeking the login of the
5223 uid under which it was called; it was always trying 10 times. (The default
5224 setting of finduser_retries is zero.) Also, it was sleeping after the final
5225 failure, which is pointless.
5227 24. Implemented tls_on_connect_ports.
5229 25. Implemented acl_smtp_predata.
5231 26. If the domain in control=submission is set empty, Exim assumes that the
5232 authenticated id is a complete email address when it generates From: or
5233 Sender: header lines.
5235 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5236 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5237 chown and chgrp in /bin and hostname in /usr/bin.
5239 28. Exim was keeping the "process log" file open after each use, just as it
5240 does for the main log. This opens the possibility of it remaining open for
5241 long periods when the USR1 signal hits a daemon. Occasional processlog
5242 errors were reported, that could have been caused by this. Anyway, it seems
5243 much more sensible not to leave this file open at all, so that is what now
5246 29. The long-running daemon process does not normally write to the log once it
5247 has entered its main loop, and it closes the log before doing so. This is
5248 so that log files can straightforwardly be renamed and moved. However,
5249 there are a couple of unusual error situations where the daemon does write
5250 log entries, and I had neglected to close the log afterwards.
5252 30. The text of an SMTP error response that was received during a remote
5253 delivery was being truncated at 512 bytes. This is too short for some of
5254 the long messages that one sometimes sees. I've increased the limit to
5257 31. It is now possible to make retry rules that apply only when a message has a
5258 specific sender, in particular, an empty sender.
5260 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5261 it possible to be selective about when SMTP synchronization is enforced.
5263 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5265 32. Implemented hosts_connection_nolog.
5267 33. Added an ACL for QUIT.
5269 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5272 35. Added mailbox_size and mailbox_filecount to appendfile.
5274 36. Added control = no_multiline_responses to ACLs.
5276 37. There was a bug in the logic of the code that waits for the clock to tick
5277 in the case where the clock went backwards by a substantial amount such
5278 that the microsecond fraction of "now" was more than the microsecond
5279 fraction of "then" (but the whole seconds number was less).
5281 38. Added support for the libradius Radius client library this is found on
5282 FreeBSD (previously only the radiusclient library was supported).
5288 1. When certain lookups returned multiple values in the form name=value, the
5289 quoting of the values was not always being done properly. Specifically:
5290 (a) If the value started with a double quote, but contained no whitespace,
5292 (b) If the value contained whitespace other than a space character (i.e.
5293 tabs or newlines or carriage returns) it was not quoted.
5294 This fix has been applied to the mysql and pgsql lookups by writing a
5295 separate quoting function and calling it from the lookup code. The fix
5296 should probably also be applied to nisplus, ibase and oracle lookups, but
5297 since I cannot test any of those, I have not disturbed their existing code.
5299 2. A hit in the callout cache for a specific address caused a log line with no
5300 reason for rejecting RCPT. Now it says "Previous (cached) callout
5301 verification failure".
5303 3. There was an off-by-one bug in the queryprogram router. An over-long
5304 return line was truncated at 256 instead of 255 characters, thereby
5305 overflowing its buffer with the terminating zero. As well as fixing this, I
5306 have increased the buffer size to 1024 (and made a note to document this).
5308 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5309 when Exim is waiting for an SMTP response from a remote server, Exim
5310 restarts its select() call on the socket, thereby resetting its timeout.
5311 This is not a problem when such interrupts are rare. Somebody set up a cron
5312 job to run exiwhat every 2 minutes, which is less than the normal select()
5313 timeout (5 or 10 minutes). This meant that the select() timeout never
5314 kicked in because it was always reset. I have fixed this by comparing the
5315 time when an interrupt arrives with the time at the start of the first call
5316 to select(). If more time than the timeout has elapsed, the interrupt is
5317 treated as a timeout.
5319 5. Some internal re-factoring in preparation for the addition of Sieve
5320 extensions (by MH). In particular, the "personal" test is moved to a
5321 separate function, and given an option for scanning Cc: and Bcc: (which is
5322 not set for Exim filters).
5324 6. When Exim created an email address using the login of the caller as the
5325 local part (e.g. when creating a From: or Sender: header line), it was not
5326 quoting the local part when it contained special characters such as @.
5328 7. Installed new OpenBSD configuration files.
5330 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5331 try to make them clearer.
5333 9. Callout options, other than the timeout value, were being ignored when
5334 verifying sender addresses in header lines. For example, when using
5336 verify = header_sender/callout=no_cache
5338 the cache was (incorrectly) being used.
5340 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5341 only the Cygwin environment.
5343 11. When return_path_on_delivery was set as a log selector, if different remote
5344 addresses in the same message used different return paths and parallel
5345 remote delivery occurred, the wrong values would sometimes be logged.
5346 (Whenever a remote delivery process finished, the return path value from
5347 the most recently started remote delivery process was logged.)
5349 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5350 lines when AUTH and/or TLS are in use. This is the "received protocol"
5351 field. Exim used to use "asmtp" for authenticated SMTP, without any
5352 indication (in the protocol name) for TLS use. Now it follows the RFC and
5353 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5354 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5355 names appear in log lines as well as in Received: header lines.
5357 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5358 extensions, and comparison tests, and to fix some bugs.
5360 14. Changes to the "personal" filter test:
5362 (1) The test was buggy in that it was just doing the equivalent of
5363 "contains" tests on header lines. For example, if a user's address was
5364 anne@some.where, the "personal" test would incorrectly be true for
5366 To: susanne@some.where
5368 This test is now done by extracting each address from the header in turn,
5369 and checking the entire address. Other tests that are part of "personal"
5370 are now done using regular expressions (for example, to check local parts
5371 of addresses in From: header lines).
5373 (2) The list of non-personal local parts in From: addresses has been
5374 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5375 taken from the Sieve specification recommendations.
5377 (3) If the message contains any header line starting with "List-" it is
5378 treated as non-personal.
5380 (4) The test for "circular" in the Subject: header line has been removed
5381 because it now seems ill-conceived.
5383 15. Minor typos in src/EDITME comments corrected.
5385 16. Installed latest exipick from John Jetmore.
5387 17. If headers_add on a router specified a text string that was too long for
5388 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5389 of string_sprintf() is now avoided.
5391 18. $message_body_size was not set (it was always zero) when running the DATA
5392 ACL and the local_scan() function.
5394 19. For the "mail" command in an Exim filter, no default was being set for
5395 the once_repeat time, causing a random time value to be used if "once" was
5396 specified. (If the value happened to be <= 0, no repeat happened.) The
5397 default is now 0s, meaning "never repeat". The "vacation" command was OK
5398 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5399 (I found it when inspecting the code).
5401 20. There is now an overall timeout for performing a callout verification. It
5402 defaults to 4 times the callout timeout, which applies to individual SMTP
5403 commands during the callout. The overall timeout applies when there is more
5404 than one host that can be tried. The timeout is checked before trying the
5405 next host. This prevents very long delays if there are a large number of
5406 hosts and all are timing out (e.g. when the network connections are timing
5407 out). The value of the overall timeout can be changed by specifying an
5408 additional sub-option for "callout", called "maxwait". For example:
5410 verify = sender/callout=5s,maxwait=20s
5412 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5413 to the end before writing, but this should make it even safer).
5415 22. Exim was forgetting that it had advertised PIPELINING for the second and
5416 subsequent messages on an SMTP connection. It was also not resetting its
5417 memory on STARTTLS and an internal HELO.
5419 23. When Exim logs an SMTP synchronization error within a session, it now
5420 records whether PIPELINING has been advertised or not.
5422 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5423 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5424 rather than long int.
5426 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5428 26. Added the never_mail option to autoreply.
5434 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5435 crash if the getsockname() call failed; this can happen if a connection is
5436 closed very soon after it is established. The problem was simply in the
5437 order in which certain operations were done, causing Exim to try to write
5438 to the SMTP stream before it had set up the file descriptor. The bug has
5439 been fixed by making things happen in the correct order.
5445 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5446 before the connection was closed, thus losing the rejection response.
5448 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5449 some early Solaris releases, but causes trouble in current releases where
5450 socklen_t is defined.
5452 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5455 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5458 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5459 that releasing the top of it at the end releases what was used for sub-
5460 expansions (unless the block got too big). However, discard this block if
5461 the first thing is a variable or header, so that we can use its block when
5462 it is dynamic (useful for very large $message_headers, for example).
5464 6. Lookups now cache *every* query, not just the most recent. A new, separate
5465 store pool is used for this. It can be recovered when all lookup caches are
5466 flushed. Lookups now release memory at the end of their result strings.
5467 This has involved some general refactoring of the lookup sources.
5469 7. Some code has been added to the store_xxx() functions to reduce the amount
5470 of flapping under certain conditions.
5472 8. log_incoming_interface used to affect only the <= reception log lines. Now
5473 it causes the local interface and port to be added to several more SMTP log
5474 lines, for example "SMTP connection from", and rejection lines.
5476 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5478 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5480 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5481 could be overwritten at the end of the current message (or the start of a
5482 new message if it was set in a HELO ACL). The value is now preserved for
5483 the duration of the SMTP connection.
5485 12. If a transport had a headers_rewrite setting, and a matching header line
5486 contained an unqualified address, that address was qualified, even if it
5487 did not match any rewriting rules. The underlying bug was that the values
5488 of the flags that permit the existence of unqualified sender and recipient
5489 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5490 non-local messages, and by -bnq for local messages) were not being
5491 preserved with the message after it was received.
5493 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5494 "next input=" as part of the text comprising the host identity instead of
5495 the correct text. The code was using the same buffer for two different
5496 strings. However, depending on which order the printing function evaluated
5497 its arguments, the bug did not always show up. Under Linux, for example, my
5498 test suite worked just fine.
5500 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5501 doesn't work with some older versions of Perl. It has been changed to "my",
5502 which in any case is probably the better facility to use.
5504 15. A really picky compiler found some instances of statements for creating
5505 error messages that either had too many or two few arguments for the format
5508 16. The size of the buffer for calls to the DNS resolver has been increased
5509 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5510 for addresses that have a lot of PTR records. This alleviates a problem; it
5511 does not fully solve it.
5513 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5514 buffer now truncates the list and logs the incident, which is the same
5515 action as happens when Exim is looking up a host name and its aliases.
5516 Previously in this situation something unpredictable would happen;
5517 sometimes it was "internal error: store_reset failed".
5519 18. If a server dropped the connection unexpectedly when an Exim client was
5520 using GnuTLS and trying to read a response, the client delivery process
5521 crashed while trying to generate an error log message.
5523 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5524 string, for example:
5526 warn message = H1: something\nH2: something
5528 the text was added as a single header line from Exim's point of view
5529 though it ended up OK in the delivered message. However, searching for the
5530 second and subsequent header lines using $h_h2: did not work. This has been
5531 fixed. Similarly, if a system filter added multiple headers in this way,
5532 the routers could not see them.
5534 20. Expanded the error message when iplsearch is called with an invalid key to
5535 suggest using net-iplsearch in a host list.
5537 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5538 ACLs are no longer actually imposed (and a message to that effect is
5541 22. If a "gecos" field in a passwd entry contained escaped characters, in
5542 particular, if it contained a \" sequence, Exim got it wrong when building
5543 a From: or a Sender: header from that name. A second bug also caused
5544 incorrect handling when an unquoted " was present following a character
5545 that needed quoting.
5547 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5548 was not being matched caselessly.
5550 24. Arranged for all hyphens in the exim.8 source to be escaped with
5553 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5554 properly. Recipient callout cache records were still being keyed to include
5555 the sender, even when use_sender was set false. This led to far more
5556 callouts that were necessary. The sender is no longer included in the key
5557 when use_sender is false.
5559 26. Added "control = submission" modifier to ACLs.
5561 27. Added the ${base62d: operator to decode base 62 numbers.
5563 28. dnsdb lookups can now access SRV records.
5565 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5566 the configuration file.
5568 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5569 (-v) mode. This makes the output for a verbose queue run more intelligible.
5571 31. Added a use_postmaster feature to recipient callouts.
5573 32. Added the $body_zerocount variable, containing the number of binary zero
5574 bytes in the message body.
5576 33. The time of last modification of the "new" subdirectory is now used as the
5577 "mailbox time last read" when there is a quota error for a maildir
5580 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5582 35. Added +ignore_unknown as a special item in host lists.
5584 36. Code for decoding IPv6 addresses in host lists is now included, even if
5585 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5586 address was recognized as an IP address, but was then not correctly decoded
5587 into binary, causing unexpected and incorrect effects when compared with
5594 1. Very minor rewording of debugging text in manualroute to say "list of
5595 hosts" instead of "hostlist".
5597 2. If verify=header_syntax was set, and a header line with an unqualified
5598 address (no domain) and a large number of spaces between the end of the
5599 name and the colon was received, the reception process suffered a buffer
5600 overflow, and (when I tested it) crashed. This was caused by some obsolete
5601 code that should have been removed. The fix is to remove it!
5603 3. When running in the test harness, delay a bit after writing a bounce
5604 message to get a bit more predictability in the log output.
5606 4. Added a call to search_tidyup() just before forking a reception process. In
5607 theory, someone could use a lookup in the expansion of smtp_accept_max_
5608 per_host which, without the tidyup, could leave open a database connection.
5610 5. Added the variables $recipient_data and $sender_data which get set from a
5611 lookup success in an ACL "recipients" or "senders" condition, or a router
5612 "senders" option, similar to $domain_data and $local_part_data.
5614 6. Moved the writing of debug_print from before to after the "senders" test
5617 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5618 problems for message scanning, either using a data ACL, or using
5619 local_scan() because the Received: header was not generated till after they
5620 were called (in order to set the time as the time of reception completion).
5621 I have revised the way this works. The header is now generated after the
5622 body is received, but before the ACL or local_scan() are called. After they
5623 are run, the timestamp in the header is updated.
5629 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5630 before starting a queue runner without re-exec. This happened only when
5631 deliver_drop_privilege was set or when the Exim user was set to root. The
5632 effect of the bug was that timeouts during subsequent deliveries caused
5633 crashes instead of being properly handled. The handler is now left at its
5634 default (and expected) setting.
5636 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5637 message, again when deliver_drop_privilege is set or Exim is run as root.
5638 The bug described in (1) was not present in this case, but the tidying up
5639 of the other signals was missing. I have made the two cases consistent.
5641 3. The ignore_target_hosts setting on a manualroute router was being ignored
5642 for hosts that were looked up using the /MX notation.
5644 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5647 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5648 operated on the sender address. After changing the $sender_address to <>
5649 for the sender address verify, Exim was re-instated it as the original
5650 (before rewriting) address, but remembering that it had rewritten it, so it
5651 wasn't rewriting it again. This bug also had the effect of breaking the
5652 sender address verification caching when the sender address was rewritten.
5654 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5655 This has been changed so that if the ip literal address matches
5656 ignore_target_hosts, the router declines.
5658 7. Added expansion conditions match_domain, match_address, and match_local_
5659 part (NOT match_host).
5661 8. The placeholder for the Received: header didn't have a length field set.
5663 9. Added code to Exim itself and to exim_lock to test for a specific race
5664 condition that could lead to file corruption when using MBX delivery. The
5665 issue is with the lockfile that is created in /tmp. If this file is removed
5666 after a process has opened it but before that process has acquired a lock,
5667 there is the potential for a second process to recreate the file and also
5668 acquire a lock. This could lead to two Exim processes writing to the file
5669 at the same time. The added code performs the same test as UW imapd; it
5670 checks after acquiring the lock that its file descriptor still refers to
5671 the same named file.
5673 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5674 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5677 11. Added the smtp_active_hostname option. If used, this will typically be made
5678 to depend on the incoming interface address. Because $interface_address is
5679 not set up until the daemon has forked a reception process, error responses
5680 that can happen earlier (such as "too many connections") no longer contain
5683 12. If an expansion in a condition on a "warn" statement fails because a lookup
5684 defers, the "warn" statement is abandoned, and the next ACL statement is
5685 processed. Previously this caused the whole ACL to be aborted.
5687 13. Added the iplsearch lookup type.
5689 14. Added ident_timeout as a log selector.
5691 15. Added tls_certificate_verified as a log selector.
5693 16. Added a global option tls_require_ciphers (compare the smtp transport
5694 option of the same name). This controls incoming TLS connections.
5696 17. I finally figured out how to make tls_require_ciphers do a similar thing
5697 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5698 before starting the TLS session.
5700 18. Tabs are now shown as \t in -bP output.
5702 19. If the log selector return_path_on_delivery was set, Exim crashed when
5703 bouncing a message because it had too many Received: header lines.
5705 20. If two routers both had headers_remove settings, and the first one included
5706 a superfluous trailing colon, the final name in the first list and the
5707 first name in the second list were incorrectly joined into one item (with a
5708 colon in the middle).
5714 1. Added -C and -D options to the exinext utility, mainly to make it easier
5715 to include in the automated testing, but these could be helpful when
5716 multiple configurations are in use.
5718 2. The exinext utility was not formatting the output nicely when there was
5719 an alternate port involved in the retry record key, nor when there was a
5720 message id as well (for retries that were specific to a specific message
5721 and a specific host). It was also confused by IPv6 addresses, because of
5722 the additional colons they contain. I have fixed the IPv4 problem, and
5723 patched it up to do a reasonable job for IPv6.
5725 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5726 delivery, the log line now contains "pipelined" if PIPELINING was used.
5728 4. An SMTP transport process used to panic and die if the bind() call to set
5729 an explicit outgoing interface failed. This has been changed; it is now
5730 treated in the same way as a connect() failure.
5732 5. A reference to $sender_host_name in the part of a conditional expansion
5733 that was being skipped was still causing a DNS lookup. This no longer
5736 6. The def: expansion condition was not recognizing references to header lines
5737 that used bh_ and bheader_.
5739 7. Added the _cache feature to named lists.
5741 8. The code for checking quota_filecount in the appendfile transport was
5742 allowing one more file than it should have been.
5744 9. For compatibility with Sendmail, the command line option
5752 and sets the incoming protocol and host name (for trusted callers). The
5753 host name and its colon can be omitted when only the protocol is to be set.
5754 Note the Exim already has two private options, -pd and -ps, that refer to
5755 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5756 "s", but I don't think that's a major issue.
5758 10. A number of refactoring changes to the code, none of which should affect
5761 (a) The number of logging options was getting close to filling up the
5762 32-bit word that was used as a bit map. I have split them into two classes:
5763 those that are passed in the argument to log_write(), and those that are
5764 only ever tested independently outside of that function. These are now in
5765 separate 32-bit words, so there is plenty of room for expansion again.
5766 There is no change in the user interface or the logging behaviour.
5768 (b) When building, for example, log lines, the code previously used a
5769 macro that called string_cat() twice, in order to add two strings. This is
5770 not really sufficiently general. Furthermore, there was one instance where
5771 it was actually wrong because one of the argument was used twice, and in
5772 one call a function was used. (As it happened, calling the function twice
5773 did not affect the overall behaviour.) The macro has been replaced by a
5774 function that can join an arbitrary number of extra strings onto a growing
5777 (c) The code for expansion conditions now uses a table and a binary chop
5778 instead of a serial search (which was left over from when there were very
5779 few conditions). Also, it now recognizes conditions like "pam" even when
5780 the relevant support is not compiled in: a suitably worded error message is
5781 given if an attempt is made to use such a condition.
5783 11. Added ${time_interval:xxxxx}.
5785 12. A bug was causing one of the ddress fields not to be passed back correctly
5786 from remote delivery subprocesses. The field in question was not being
5787 subsequently used, so this caused to problems in practice.
5789 13. Added new log selectors queue_time and deliver_time.
5791 14. Might have fixed a bug in maildirsizefile handling that threw up
5792 "unexpected character" debug warnings, and recalculated the data
5793 unnecessarily. In any case, I expanded the warning message to give more
5796 15. Added the message "Restricted characters in address" to the statements in
5797 the default ACL that block characters like @ and % in local parts.
5799 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5800 Three changes have been made:
5802 (a) There was a serious bug; a negative response to MAIL caused the whole
5803 recipient domain to be cached as invalid, thereby blocking all messages
5804 to all local parts at the same domain, from all senders. This bug has
5805 been fixed. The domain is no longer cached after a negative response to
5806 MAIL if the sender used is not empty.
5808 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5811 (c) A new callout option, "use_sender" has been added for people who want
5812 the modified behaviour.
5818 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5821 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5822 indeed breaks things for older releases.
5824 3. Added additional logging to the case where there is a problem reading data
5825 from a filter that is running in a subprocess using a pipe, in order to
5826 try to track down a specific problem.
5828 4. Testing facility fudge: when running in the test harness and attempting
5829 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5830 getting "No route to host". Convert this to a timeout.
5832 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5835 6. Some OS don't have socklen_t but use size_t instead. This affects the
5836 fifth argument of getsockopt() amongst other things. This is now
5837 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5838 can be set for individual OS. I have set it for SunOS5, OSF1, and
5839 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5840 some earlier ones do not.
5842 7. Change 4.30/15 was not doing the test caselessly.
5844 8. The standard form for an IPv6 address literal was being rejected by address
5845 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5846 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5847 this, as well as the form without the "IPv6" on the front (but only when
5848 address literals are enabled, of course).
5850 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5852 10. Exim crashed if a message with an empty sender address specified by -f
5853 encountered a router with an errors_to setting. This could be provoked only
5854 by a command such as
5858 where an empty string was supplied; "<>" did not hit this bug.
5860 11. Installed PCRE release 4.5.
5862 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5863 remained set. It is now erased.
5865 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5866 times from message ids (which are base 36 rather than the normal 62).
5868 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5869 were being counted as actual protocol errors, and logged if the log
5870 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5871 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5872 and DATA following a set of rejected RCPTs do not count as protocol errors.
5873 In other words, Exim assumes they were pipelined, though this may not
5874 actually be the case. Of course, in all cases the client gets an
5875 appropriate error code.
5877 15. If a lookup fails in an ACL condition, a message about the failure may
5878 be available; it is used if testing the ACL cannot continue, because most
5879 such messages specify what the cause of the deferral is. However, some
5880 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5881 that caused an old message to be retained and used if a later statement
5882 caused a defer, replacing the real cause of the deferral.
5884 16. If an IP address had so many PTR records that the DNS lookup buffer
5885 was not large enough to hold them, Exim could crash while trying to process
5886 the truncated data. It now detects and logs this case.
5888 17. Further to 4.21/58, another change has been made: if (and only if) the
5889 first line of a message (the first header line) ends with CRLF, a bare LF
5890 in a subsequent header line has a space inserted after it, so as not to
5891 terminate the header.
5893 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5894 unnecessarily, used atoi() instead of strtol(), and didn't check the
5895 termination when getting file sizes from file names by regex.
5897 19. Completely re-implemented the support for maildirsize files, in the light
5898 of a number of problems with the previous contributed implementation
5899 (4.30/29). In particular:
5901 . If the quota is zero, the maildirsize file is maintained, but no quota is
5904 . If the maildir directory does not exist, it is created before any attempt
5905 to write a maildirsize file.
5907 . The quota value in the file is just a cache; if the quota is changed in
5908 the transport, the new value overrides.
5910 . A regular expression is available for excluding directories from the
5913 20. The autoreply transport checks the characters in options that define the
5914 message's headers; it allows continued headers, but it was checking with
5915 isspace() after an embedded newline instead of explicitly looking for a
5918 21. If all the "regular" hosts to which an address was routed had passed their
5919 expiry times, and had not reached their retry times, the address was
5920 bounced, even if fallback hosts were defined. Now Exim should go on to try
5923 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5924 equivalent code in the SMTP transport. Some hosts send humungous responses
5925 to HELO/EHLO, more than 1024 it seems.
5927 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5928 gives compiler warnings in some environments. I've now done it "properly",
5931 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5932 (because of problems with the built-in one) was declared to return uschar *
5933 instead of char *, causing compiler failure.
5935 25. Fixed a file descriptor leak when processing alias/forward files.
5937 26. Fixed a minor format string issue in dbfn.c.
5939 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5941 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5942 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5945 29. When a sender address is verified, it is cached, to save repeating the test
5946 when there is more than one recipient in a message. However, when the
5947 verification involves a callout, it is possible for different callout
5948 options to be set for different recipients. It is too complicated to keep
5949 track of this in the cache, so now Exim always runs a verification when a
5950 callout is required, relying on the callout cache for the optimization.
5951 The overhead is duplication of the address routing, but this should not be
5954 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5955 to be verified with callout=postmaster, and the main callout worked but the
5956 postmaster check failed, the verification correctly failed. However, if a
5957 subsequent RCPT command asked for sender verification *without* the
5958 postmaster check, incorrect caching caused this verification also to fail,
5961 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5962 it was not caching the DNS options (qualify_single, search_parents) that
5963 were used when the lookup failed. A subsequent lookup with different
5964 options therefore always gave the same answer, though there were cases
5965 where it should not have. (Example: a "domains = !$mx_any" option on a
5966 dnslookup router: the "domains" option is always processed without any
5967 widening, but the router might have qualify_single set.) Now Exim uses the
5968 cached value only when the same options are set.
5970 32. Added John Jetmore's "exipick" utility to the distribution.
5972 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5973 than a timeout (e.g. a certificate is required, and is not provided), an
5974 Exim server now closes the connection immediately. Previously it waited for
5975 the client to close - but if the client is SSL, it seems that they each
5976 wait for each other, leading to a delay before one of them times out.
5978 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5979 maintained 0.8.x compatibility because I don't think many are using it, and
5980 it is clearly obsolete.
5982 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5985 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5986 client certificate was expired. A simple patch fixes this, though I don't
5987 understand the full logic of why the verify callback is called multiple
5990 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5991 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5992 which causes problems with some clients (such as the Certicom SSL Plus
5993 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5994 disables the coutermeasure allowing Eudora to connect."
5996 38. Exim was not checking that a write() to a log file succeeded. This could
5997 lead to Bad Things if a log got too big, in particular if it hit a file
5998 size limit. Exim now panics and dies if it cannot write to a log file, just
5999 as it does if it cannot open a log file.
6001 39. Modified OS/Makefile-Linux so that it now contains
6003 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
6005 The two -D definitions ensure that Exim is compiled with large file
6006 support, which makes it possible to handle log files that are bigger than
6009 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
6010 instance) a domain was checked against a named list that involved a lookup,
6011 causing $domain_data to be set, then another domain was checked against the
6012 same list, then the first domain was re-checked, the value of $domain_data
6013 after the final check could be wrong. In particular, if the second check
6014 failed, it could be set empty. This bug probably also applied to
6017 41. The strip_trailing_dot option was not being applied to the address given
6018 with the -f command-line option.
6020 42. The code for reading a message's header from the spool was incrementing
6021 $received_count, but never initializing it. This meant that the value was
6022 incorrect (doubled) while delivering a message in the same process in which
6023 it was received. In the most common configuration of Exim, this never
6024 happens - a fresh exec is done - but it can happen when
6025 deliver_drop_privilege is set.
6027 43. When Exim logs an SMTP synchronization error - client data sent too soon -
6028 it now includes up to 150 characters of the unexpected data in the log
6031 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6032 and building data strings. The size of both of these buffers was 10 000
6033 bytes - far larger than anybody would *ever* want, thought I. Needless to
6034 say, somebody hit the limit. I have increased the maximum line length to
6035 20 000 and the maximum data length of concatenated lines to 100 000. I have
6036 also fixed two bugs, because there was no checking on these buffers. Tsk,
6037 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6038 buffer is too small.
6040 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6041 lsearch lookups. Now it does.
6043 46. When parsing a route_list item in a manualroute router, a fixed-length
6044 buffer was used for the list of hosts. I made this 1024 bytes long,
6045 thinking that nobody would ever have a list of hosts that long. Wrong.
6046 Somebody had a whole pile of complicated expansion conditions, and the
6047 string was silently truncated, leading to an expansion error. It turns out
6048 that it is easier to change to an unlimited length (owing to other changes
6049 that have happened since this code was originally written) than to build
6050 structure for giving a limitation error. The length of the item that
6051 expands into the list of hosts is now unlimited.
6053 47. The lsearch lookup could not handle data where the length of text line was
6054 more than 4095 characters. Such lines were truncated, leading to shortened
6055 data being returned. It should now handle lines of any length.
6057 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6058 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6061 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6062 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6063 no longer happens in comment lines. A list of replacements is now placed
6064 at the head of all of the source files, except those whose only change is
6065 to replace PERL_COMMAND in the very first #! line.
6067 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6068 messages on the queue, with a bottom-up merge sort, using code contributed
6069 by Michael Haardt. This should make operations like -bp somewhat faster on
6070 large queues. It won't affect queue runners, except when queue_run_in_order
6073 51. Installed eximstats 1.31 in the distribution.
6075 52. Added support for SRV lookups to the dnslookup router.
6077 53. If an ACL referred to $message_body or $message_body_end, the value was not
6078 reset for any messages that followed in the same SMTP session.
6080 54. The store-handling optimization for building very long strings was not
6081 differentiating between the different store pools. I don't think this
6082 actually made any difference in practice, but I've tidied it.
6084 55. While running the routers to verify a sender address, $sender_address
6085 was still set to the sender address. This is wrong, because when routing to
6086 send a bounce to the sender, it would be empty. Therefore, I have changed
6087 it so that, while verifying a sender address, $sender_address is set to <>.
6088 (There is no change to what happens when verifying a recipient address.)
6090 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6091 target A or AAAA records (if not already returned) without resetting the
6092 qualify_single or search_parents options of the DNS resolver. These are
6093 inappropriate in this case because the targets of MX and SRV records must
6094 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6095 target that, when qualified, matched something in the local domain. These
6096 two options are now turned off when doing these lookups.
6098 57. It seems that at least some releases of Reiserfs (which does not have the
6099 concept of a fixed number of inodes) returns zero and not -1 for the
6100 number of available inodes. This interacted badly with check_spool_inodes,
6101 which assumed that -1 was the "no such thing" setting. What I have done is
6102 to check that the total number of inodes is greater than zero before doing
6103 the test of how many are available.
6105 58. When a "warn" ACL statement has a log_message modifier, the message is
6106 remembered, and not repeated. This is to avoid a lot of repetition when a
6107 message has many recipients that cause the same warning to be written.
6108 However, Exim was preserving the list of already written lines for an
6109 entire SMTP session, which doesn't seem right. The memory is now reset if a
6110 new message is started.
6112 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6113 debugging output unless log_rewrite was also set.
6115 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6116 of (int)(handle) when we know that handle contains (void *)(-1).
6118 61. The Exim daemon panic-logs an error return when it closes the incoming
6119 connection. However "connection reset by peer" seems to be common, and
6120 isn't really an error worthy of noting specially, so that particular error
6123 62. When Exim is trying to find all the local interfaces, it used to panic and
6124 die if the ioctl to get the interface flags failed. However, it seems that
6125 on at least one OS (Solaris 9) it is possible to have an interface that is
6126 included in the list of interfaces, but for which you get a failure error
6127 for this call. This happens when the interface is not "plumbed" into a
6128 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6129 failure of the "get flags" call assumes that the interface is down.
6131 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6132 makes life easier for people who are doing arithmetic on fields extracted
6133 from dates, where you often get leading zeros that should not be
6134 interpreted as octal.
6136 64. Added qualify_domain to the redirect router, to override the global
6139 65. If a pathologically long header line contained very many addresses (the
6140 report of this problem mentioned 10 000) and each of them was rewritten,
6141 Exim could use up a very large amount of memory. (It kept on making new
6142 copies of the header line as it rewrote, and never released the old ones.)
6143 At the expense of a bit more processing, the header rewriting function has
6144 been changed so that it no longer eats memory in this way.
6146 66. The generation of the Received: header has been moved from the time that a
6147 message starts to be received, to the time that it finishes. The timestamp
6148 in the Received: header should now be very close to that of the <= log
6149 line. There are two side-effects of this change:
6151 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6152 logged header lines no longer include the local Received: line, because
6153 it has not yet been created. The same applies to a copy of the message
6154 that is returned to a non-SMTP sender when a message is rejected.
6156 (b) When a filter file is tested using -bf, no additional Received: header
6157 is added to the test message. After some thought, I decided that this
6160 This change does not affect the value of $received_for. It is still set
6161 after address rewriting, but before local_scan() is called.
6163 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6165 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6166 gave an unhelpful panic error message, and a defer error. I have managed to
6167 change this behaviour so that it now rejects any supplied certificate,
6168 which seems right, as the list of acceptable certificates is empty.
6170 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6171 gave an unhelpful defer error. I have not managed to make this reject any
6172 supplied certificates, but the error message it gives is "no certificate
6173 supplied", which is not helpful.
6175 70. exigrep's output now also includes lines that are not associated with any
6176 message, but which match the given pattern. Implemented by a patch from
6177 Martin Sluka, which also tidied up the Perl a bit.
6179 71. Recipient callout verification, like sender verification, was using <> in
6180 the MAIL FROM command. This isn't really the right thing, since the actual
6181 sender may affect whether the remote host accepts the recipient or not. I
6182 have changed it to use the actual sender in the callout; this means that
6183 the cache record is now keyed on a recipient/sender pair, not just the
6184 recipient address. There doesn't seem to be a real danger of callout loops,
6185 since a callout by the remote host to check the sender would use <>.
6186 [SEE ABOVE: changed after hitting problems.]
6188 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6189 temporary errors. However, in the case of such a code being given after
6190 the end of a data transmission (i.e. after ".") Exim was failing to write
6191 a retry record for the message. (Yes, there was some broken host that was
6192 actually sending 8xx at this point.)
6194 73. An unknown lookup type in a host list could cause Exim to panic-die when
6195 the list was checked. (An example that provoked this was putting <; in the
6196 middle of a list instead of at the start.) If this happened during a DATA
6197 ACL check, a -D file could be left lying around. This kind of configuration
6198 error no longer causes Exim to die; instead it causes a defer error. The
6199 incident is still logged to the main and panic logs.
6201 74. Buglet left over from Exim 3 conversion. The message "too many messages
6202 in one connection" was written to the rejectlog but not the mainlog, except
6203 when address rewriting (yes!) was being logged.
6205 75. Added write_rejectlog option.
6207 76. When a system filter was run not as root (that is, when system_filter_user
6208 was set), the values of the $n variables were not being returned to the
6209 main process; thus, they were not subsequently available in the $sn
6212 77. Added +return_path_on_delivery log selector.
6214 78. A connection timeout was being treated differently from recipients deferred
6215 when testing hosts_max_try with a message that was older than the host's
6216 retry timeout. (The host should not be counted, thus allowing all hosts to
6217 be tried at least once before bouncing.) This may have been the cause of an
6218 occasionally reported bug whereby a message would remain on the queue
6219 longer than the retry timeout, but would be bounced if a delivery was
6220 forced. I say "may" because I never totally pinned down the problem;
6221 setting up timeout/retry tests is difficult. See also the next item.
6223 79. The ultimate address timeout was not being applied to errors that involved
6224 a combination of host plus message (for example, a timeout on a MAIL
6225 command). When an address resolved to a number of possible hosts, and they
6226 were not all tried for each delivery (e.g. because of hosts_max_try), a
6227 message could remain on the queue longer than the retry timeout.
6229 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6232 81. Fixed an obscure SMTP outgoing bug which required at least the following
6233 conditions: (a) there was another message waiting for the same server;
6234 (b) the server returned 5xx to all RCPT commands in the first message so
6235 that the message was not completed; (c) the server dropped the connection
6236 or gave a negative response to the RSET that Exim sends to abort the
6237 transaction. The observed case was a dropped connection after DATA that had
6238 been sent in pipelining mode. That is, the server had advertised PIPELINING
6239 but was not implementing it correctly. The effect of the bug was incorrect
6240 behaviour, such as trying another host, and this could lead to a crash.
6246 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6247 and daemon.c were passed as pointers to ints; they should have been
6248 pointers to socklen_t variables (which are typically unsigned ints).
6250 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6253 3. Fixed a really odd bug that affected only the testing scheme; patching a
6254 certain fixed string in the binary changed the value of another string that
6255 happened to be identical to the end of the original first string.
6257 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6258 name", it returns that address as the IP address. On some operating
6259 systems (e.g. Solaris), it also passes back the IP address string as the
6260 "host name". However, on others (e.g. Linux), it passes back an empty
6261 string. Exim wasn't checking for this, and was changing the host name to an
6262 empty string, assuming it had been canonicalized.
6264 5. Although rare, it is permitted to have more than one PTR record for a given
6265 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6266 all the names associated with an address, because they do in Solaris.
6267 However, it seems that they do not in Linux for data that comes from the
6268 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6269 I found this out when I moved to a new Linux workstation and tried to run
6270 the Exim test suite.
6272 To get round this problem I have changed the code so that it now does its
6273 own call to the DNS to look up PTR records when searching for a host name.
6274 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6275 addresses that are only in /etc/hosts are still found.
6277 This behaviour is, however, controlled by an option called host_lookup_
6278 order, which defaults to "bydns:byaddr". If people want to use the other
6279 order, or indeed, just use one or the other means of lookup, they can
6280 specify it in this variable.
6282 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6283 some operating systems, this comes back from gethostbyaddr() as an empty
6284 string, and this is what Exim used to test for. However, it seems that in
6285 other systems, "." is yielded. Exim now tests for this case too.
6287 7. The values of check_spool_space and check_log_space are now held internally
6288 as a number of kilobytes instead of an absolute number of bytes. If a
6289 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6290 kilobyte. This means that much larger values can be stored.
6292 8. Exim monitor: an attempt to get the action menu when not actually pointing
6293 at a message produces an empty menu entitled "No message selected". This
6294 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6295 no entries in it ("Shell widget menu has zero width and/or height"). So I
6296 have added a single, blank menu entry in this case.
6298 9. Added ${quote_local_part.
6300 10. MIME decoding is now applied to the contents of Subject: header lines when
6303 11. Now that a reference to $sender_host_address automatically causes a reverse
6304 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6305 host lookup before query-style lookups in lists that might use this
6306 variable. This has therefore been abolished, and the "net-" prefix is no
6307 longer necessary for query-style lookups.
6309 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6310 have been a typo for LFLAGS, so it has been changed.
6312 13. The install script calls Exim with "-C /dev/null" in order to find the
6313 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6314 to be output. However, since Exim outputs its version number before the
6315 error, it didn't break the script. It just looked ugly. I fixed this by
6316 always allowing "-C /dev/null" if the caller is root.
6318 14. Ignore overlarge ACL variable number when reading spool file - insurance
6319 against a later release with more variables having written the file.
6321 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6322 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6323 this, as well as the form without the "IPv6" on the front.
6325 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6326 OS/Makefile-Darwin file.
6328 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6329 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6331 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6332 variable. However, some OS use ints for inodes. I've added cast to long int
6333 to get rid of the compiler warning.
6335 19. I had forgotten to lock out "/../" in configuration file names when
6336 ALT_CONFIG_PREFIX was set.
6338 20. Routers used for verification do not need to specify transports. However,
6339 if such a router generated a host list, and callout was configured, Exim
6340 crashed, because it could not find a port number from the (non-existent)
6341 transport. It now assumes port 25 in this circumstance.
6343 21. Added the -t option to exigrep.
6345 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6346 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6347 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6348 them is set, LOOKUP_LSEARCH is forced.
6350 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6352 24. Added sender and host information to the "rejected by local_scan()" log
6353 line; previously there was no indication of these.
6355 25. Added .include_if_exists.
6357 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6358 Linux. It turns out that not all file systems support this. Apparently some
6359 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6360 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6361 not supported on the file descriptor, is now ignored when Exim is trying to
6362 sync a directory. This applies only to Linux.
6364 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6366 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6367 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6370 29. Added support for maildirsize files from supplied patch (modified a bit).
6372 30. The use of :fail: followed by an empty string could lead Exim to respond to
6373 sender verification failures with (e.g.):
6375 550 Verification failed for <xxx>
6376 550 Sender verify failed
6378 where the first response line was missing the '-' that indicates it is not
6379 the final line of the response.
6381 31. The loop for finding the name of the user that called Exim had a hardwired
6382 limit of 10; it now uses the value of finduser_retries, which is used for
6383 all other user lookups.
6385 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6388 33. Exim was neglecting to zero errno before one call of strtol() when
6389 expanding a string and expecting an integer value. On some systems this
6390 resulted in spurious "integer overflow" errors. Also, it was casting the
6391 result into an int without checking.
6393 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6394 did not work. The code looks as if it has *never* worked, though it appears
6395 to have been documented since at least release 1.62. I have made it work.
6397 35. The "timeout_DNS" error in retry rules, also documented since at least
6398 1.62, also never worked. As it isn't clear exactly what this means, and
6399 clearly it isn't a major issue, I have abolished the feature by treating it
6400 as "timeout", and writing a warning to the main and panic logs.
6402 36. The display of retry rules for -brt wasn't always showing the error code
6405 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6406 timeout_connect_A, timeout_connect_MX.
6408 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6409 to the empty sender.
6411 39. The daemon was not analysing the content of -oX till after it had closed
6412 stderr and disconnected from the controlling terminal. This meant that any
6413 syntax errors were only noted on the panic log, and the return code from
6414 the command was 0. By re-arranging the code a little, I've made the
6415 decoding happen first, so such errors now appear on stderr, and the return
6416 code is 1. However, the actual setting up of the sockets still happens in
6417 the disconnected process, so errors there are still only recorded on the
6420 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6421 connections (as happens on some IP stacks) was logged at start up time as
6422 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6423 it from "IPv6 and IPv4", which means that two separate sockets are being
6426 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6427 says whether AF_INET or AF_INET6 was passed as an argument.
6429 42. Exiwhat output was messed up when time zones were included in log
6432 43. Exiwhat now gives more information about the daemon's listening ports,
6433 and whether -tls-on-connect was used.
6435 44. The "port" option of the smtp transport is now expanded.
6437 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6438 silently ignored. Now an error message is written to the main and panic
6441 46. There's a new ACL modifier called "logwrite" which writes to a log file
6442 as soon as it is encountered.
6444 47. Added $local_user_uid and $local_user_gid at routing time.
6446 48. Exim crashed when trying to verify a sender address that was being
6449 49. Exim was recognizing only a space character after ".include". It now also
6450 recognizes a tab character.
6452 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6453 extracting the relevant information from the specification. The man page no
6454 longer contains scrambled data for the -d option, and I've added a section
6455 at the front about calling Exim under different names.
6457 51. Added "extra_headers" argument to the "mail" command in filter files.
6459 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6462 53. Installed eximstats 1.29.
6464 54. Added transport_filter_timeout as a generic transport option.
6466 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6467 Cc: header lines. This was required by RFC 822, but it not required by RFC
6470 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6471 incoming messages that did not have them. Now it does so only if the
6472 message originates locally, that is, if there is no associated remote host
6473 address. When Resent- header lines are present, this applies to the Resent-
6474 lines rather than the non-Resent- lines.
6476 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6477 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6479 58. Messages for configuration errors now include the name of the main
6480 configuration file - useful now that there may be more than one file in a
6481 list (.included file names were always shown).
6483 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6484 for those rare installations that do not start the daemon as root or run it
6485 setuid root. I've cut out the call to initgroups() if the daemon is not
6488 60. The Exim user and group can now be bound into the binary as text strings
6489 that are looked up at the start of Exim's processing.
6491 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6493 62. Added $mailstore_basename variable.
6495 63. Installed patch to sieve.c from Michael Haardt.
6497 64. When Exim failed to open the panic log after failing to open the main log,
6498 the original message it was trying to log was written to stderr and debug
6499 output, but if they were not available (the usual case in production), it
6500 was lost. Now it is written to syslog before the two lines that record the
6501 failures to open the logs.
6503 65. Users' Exim filters run in subprocesses under the user's uid. It is
6504 possible for a "deliver" command or an alias in a "personal" command to
6505 provoke an address rewrite. If logging of address rewriting is configured,
6506 this fails because the process is not running as root or exim. There may be
6507 a better way of dealing with this, but for the moment (because 4.30 needs
6508 to be released), I have disabled address rewrite logging when running a
6509 filter in a non-root, non-exim process.
6515 1. The buildconfig auxiliary program wasn't quoting the value set for
6516 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6517 not defined. This bug was masked in 4.22 by the effect that was fixed in
6520 2. Some messages that were rejected after a message id was allocated were
6521 shown as "incomplete" by exigrep. It no longer does this for messages that
6522 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6524 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6525 have allow_domain_literals set, the ID did not get logged in the <= line.
6526 Domain literals are now always recognized in Message-ID: header lines.
6528 4. The first argument for a ${extract expansion item is the key name or field
6529 number. Leading and trailing spaces in this item were not being ignored,
6530 causing some misleading effects.
6532 5. When deliver_drop_privilege was set, single queue runner processes started
6533 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6534 same command in the process it spins off) were not dropping privilege.
6536 6. When the daemon running as "exim" started a queue runner, it always
6537 re-executed Exim in the spun-off process. This is a waste of effort when
6538 deliver_drop_privilege is set. The new process now just calls the
6539 queue-runner function directly.
6545 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6548 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6549 ignored. Though the use of -oP was forcing the writing of a pid file, it
6550 was always written to the default place.
6552 3. If the message "no IP address found for host xxxx" is generated during
6553 incoming verification, it is now followed by identification of the incoming
6554 connection (so you can more easily find what provoked it).
6556 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6558 5. Added some features to "harden" Exim a bit more against certain attacks:
6560 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6561 be put in Local/Makefile. This is like the never_users runtime option,
6562 but it cannot be overridden. The default setting is "root".
6564 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6565 prefix string with which any file named in a -C command line option
6568 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6569 is retained for -C and -D only if the caller of Exim is root. Without
6570 it, the exim user may also use -C and -D and retain privilege.
6572 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6573 command line option is disabled.
6575 6. Macro names set by the -D option must start with an upper case letter, just
6576 like macro names defined in the configuration file.
6578 7. Added "dereference=" facility to LDAP.
6580 8. Two instances of the typo "uknown" in the source files are fixed.
6582 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6583 the Configure-Makefile script screwed up while processing it.
6585 10. Incorporated PCRE 4.4.
6587 11. The SMTP synchronization check was not operating right at the start of an
6588 SMTP session. For example, it could not catch a HELO sent before the client
6589 waited for the greeting. There is now a check for outstanding input at the
6590 point when the greeting is written. Because of the duplex, asynchronous
6591 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6592 way, but not yet received, when the check is performed.
6594 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6595 on TCP/IP sockets, because this apparently causes some broken clients to
6598 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6599 unchanged) from the Cygwin maintainer.
6601 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6602 was supported instead of testing for mbx. Effectively a typo.
6604 15. The spa authenticator server code was not checking that the input it
6605 received was valid base64.
6607 16. The debug output line for the "set" modifier in ACLs was not showing the
6608 name of the variable that was being set.
6610 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6612 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6613 lookup on its own. Something else was needed to trigger the lookup. For
6614 example, a match in host_lookup or the need for a host name in a host list.
6615 Now, if $sender_host_name is referenced and the host name has not yet been
6616 looked up, a lookup is performed. If the lookup fails, the variable remains
6617 empty, and $host_lookup_failed is set to "1".
6619 19. Added "eqi" as a case-independent comparison operator.
6621 20. The saslauthd authentication condition could segfault if neither service
6622 nor realm was specified.
6624 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6625 error message that was logged was misleading, and incoming SMTP
6626 connections were dropped. The message is now more accurate, and temporary
6627 errors are given to SMTP connections.
6629 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6630 (except RSET) with a 421 error, until QUIT is received. However, it was
6631 failing to send a response to QUIT.
6633 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6634 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6635 ACL is now run after the helo_try_verify_hosts code.
6637 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6638 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6639 case-independent, so other case variants are also recognized). Apparently
6640 some systems use these upper case variants.
6642 25. If more than two messages were waiting for the same host, and a transport
6643 filter was specified for the transport, Exim sent two messages over the
6644 same TCP/IP connection, and then failed with "socket operation on non-
6645 socket" when it tried to send the third.
6647 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6649 27. The extern definition of crypt16() in expand.c was not being excluded when
6650 the OS had its own crypt16() function.
6652 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6653 as a preferred synonym for return_size_limit, both as an option and as an
6656 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6658 30. Changed the default configuration ACL to relax the local part checking rule
6659 for addresses that are not in any local domains. For these addresses,
6660 slashes and pipe symbols are allowed within local parts, but the sequence
6661 /../ is explicitly forbidden.
6663 31. SPA server authentication was not clearing the challenge buffer before
6666 32. log_message in a "warn" ACL statement was writing to the reject log as
6667 well as to the main log, which contradicts the documentation and doesn't
6668 seem right (because no rejection is happening). So I have stopped it.
6670 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6671 However, I am unable to do any testing of this.
6673 34. Fixed an infelicity in the appendfile transport. When checking directories
6674 for a mailbox, to see if any needed to be created, it was accidentally
6675 using path names with one or more superfluous leading slashes; tracing
6676 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6678 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6679 contents are added to the log line that is written for every discarded
6680 recipient. (Previously a log_message setting was ignored.)
6682 36. The ${quote: operator now quotes the string if it is empty.
6684 37. The install script runs exim in order to find its version number. If for
6685 some reason other than non-existence or emptiness, which it checks, it
6686 could not run './exim', it was installing it with an empty version number,
6687 i.e. as "exim-". This error state is now caught, and the installation is
6690 38. An argument was missing from the function that creates an error message
6691 when Exim fails to connect to the socket for saslauthd authentication.
6692 This could cause Exim to crash, or give a corrupted message.
6694 39. Added isip, isip4, and isip6 to ${if conditions.
6696 40. The ACL variables $acl_xx are now saved with the message, and can be
6697 accessed later in routers, transports, and filters.
6699 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6700 strings in the file are not string-expanded.
6702 42. If a MAIL command specified a SIZE value that was too large to fit into an
6703 int variable, the check against message_size_limit failed. Such values are
6704 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6705 day this will have to be increased, but I don't think I want to be around
6706 when emails are that large.
6713 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6714 iconv() is not standard in FreeBSD.
6716 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6717 IPv6 enabled. The observed symptom was a segmentation fault on return from
6718 the function os_common_find_running_interfaces() in src/os.c.
6720 3. In the check_special_case() function in daemon.c I had used "errno" as an
6721 argument name, which causes warnings on some systems. This was basically a
6722 typo, since it was named "eno" in the comments!
6724 4. The code that waits for the clock to tick (at a resolution of some fraction
6725 of a second) so as to ensure message-id uniqueness was always waiting for
6726 at least one whole tick, when it could have waited for less. [This is
6727 almost certainly not relevant at current processor speeds, where it is
6728 unlikely to ever wait at all. But we try to future-proof.]
6730 5. The function that sleeps for a time interval that includes fractions of a
6731 second contained a race. It did not block SIGALRM between setting the
6732 timer, and suspending (a couple of lines later). If the interval was short
6733 and the sigsuspend() was delayed until after it had expired, the suspension
6734 never ended. On busy systems this could lead to processes getting stuck for
6737 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6738 process, before it forks any delivery processes. The open lookup caching
6739 mechanism meant that the open file or database connection was passed into
6740 the delivery process. The problem was that delivery processes always tidy
6741 up cached lookup data. This could cause a problem for the next delivery
6742 process started by the queue runner, because the external queue runner
6743 process does not know about the closure. So the next delivery process
6744 still has data in the lookup cache. In the case of a file lookup, there was
6745 no problem because closing a file descriptor in a subprocess doesn't affect
6746 the parent. However, if the lookup was caching a connection to a database,
6747 the connection was closed, and the second delivery process was likely to
6748 see errors such as "PGSQL: query failed: server closed the connection
6749 unexpectedly". The problem has been fixed by closing all cached lookups
6750 in a queue runner before running a delivery process.
6752 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6753 seem to have the "const" qualifier which it has on other OS. I've
6756 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6757 *of the same type* (client or server) with the same public name that an
6758 error should be diagnosed.
6760 9. When Exim looked up a host name for an IP address, but failed to find the
6761 original IP address when looking up the host name (a safety check), it
6762 output the message "<ip address> does not match any IP for NULL", which was
6763 confusing, to say the least. The bug was that the host name should have
6764 appeared instead of "NULL".
6766 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6767 user that is built into the binary, and the -C or -D options is used, root
6768 privilege is dropped before the configuration file is read. In addition,
6769 logging is switched to stderr instead of the normal log files. If the
6770 configuration then re-defines the Exim user, the unprivileged environment
6771 is probably not what is expected, so Exim logs a panic warning message (but
6774 However, if deliver_drop_privilege is set, the unprivileged state may well
6775 be exactly what is intended, so the warning has been cut out in that case,
6776 and Exim is allowed to try to write to its normal log files.
6782 1. smtp_return_error_details was not giving details for temporary sender
6783 or receiver verification errors.
6785 2. Diagnose a configuration error if two authenticators have the same public
6788 3. Exim used not to create the message log file for a message until the first
6789 delivery attempt. This could be confusing when incoming messages were held
6790 for policy or load reasons. The message log file is now created at the time
6791 the message is received, and an initial "Received" line is written to it.
6793 4. The automatically generated man page for command line options had a minor
6794 bug that caused no ill effects; however, a more serious problem was that
6795 the procedure for building the man page automatically didn't always
6796 operate. Consequently, release 4.20 contains an out-of-date version. This
6797 shouldn't happen again.
6799 5. When building Exim with embedded Perl support, the script that builds the
6800 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6801 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6803 6. The freeze_tell option was not being used for messages that were frozen on
6804 arrival, either by an ACL or by local_scan().
6806 7. Added the smtp_incomplete_transaction log selector.
6808 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6809 was accepting AUTH without a new EHLO.
6811 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6812 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6815 10. Logging of TCP/IP connections (when configured) now happens in the main
6816 daemon process instead of the child process, so that the TCP/IP connection
6817 count is more accurate (but it can never be perfect).
6819 11. The use of "drop" in a nested ACL was not being handled correctly in the
6820 outer ACL. Now, if condition failure induced by the nested "drop" causes
6821 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6822 or "require"), the connection is dropped.
6824 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6825 that yield "discard" can now be used with an "accept" or a "discard" verb,
6826 but an error is generated for any others (because I can't see a useful way
6827 to define what should happen).
6829 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6830 are now processed in the same way as lines in the Exim configuration file.
6831 In particular, continuation lines are supported.
6833 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6835 15. Added -ti meaning -t -i.
6837 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6838 domains, and warn by logging if others are found.
6840 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6841 returned from GIFCONF when Exim is trying to find the list of interfaces on
6842 a host. The code in os.c has been modified to copy each ifreq to an aligned
6843 structure in all cases.
6845 Also, in some cases, the returned ifreq's were being copied to a 'struct
6846 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6847 means the last couple of bytes of an IPv6 address could be chopped if the
6848 ifreq contained only a normal sockaddr (14 bytes storage).
6850 18. Named domain lists were not supported in the hosts_treat_as_local option.
6851 An entry such as +xxxx was not recognized, and was treated as a literal
6854 19. Ensure that header lines added by a DATA ACL are included in the reject log
6855 if the ACL subsequently rejects the message.
6857 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6858 MD5 (which is deprecated).
6860 21. When testing a filter file using -bf, Exim was writing a message when it
6861 took the sender from a "From " line in the message, but it was not doing so
6862 when it took $return_path from a Return-Path: header line. It now does.
6864 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6865 with a valid header line field name (a series of printing characters
6866 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6868 23. Changed "disc" in the source to "disk" to conform to the documentation and
6869 the book and for uniformity.
6871 24. Ignore Sendmail's -Ooption=value command line item.
6873 25. When execve() failed while trying to run a command in a pipe transport,
6874 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6875 could be confused with a return value of 69 from the command itself. This
6876 has been changed to 127, the value the shell returns if it is asked to run
6877 a non-existent command. The wording for the related log line suggests a
6878 non-existent command as the problem.
6880 26. If received_header_text expands to an empty string, do not add a Received:
6881 header line to the message. (Well, it adds a token one on the spool, but
6882 marks it "old" so that it doesn't get used or transmitted.)
6884 27. Installed eximstats 1.28 (addition of -nt option).
6886 28. There was no check for failure on the call to getsockname() in the daemon
6887 code. This can fail if there is a shortage of resources on the system, with
6888 ENOMEM, for example. A temporary error is now given on failure.
6890 29. Contrary to the C standard, it seems that in some environments, the
6891 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6892 program. Exim now does this explicitly; it affects the formatting of
6893 timestamps using strftime().
6895 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6896 complaints. I've now initialized all the variables, to avoid this.
6898 32. Header lines added by a system filter were not being "seen" during
6899 transport-time rewrites.
6901 33. The info_callback() function passed to OpenSSL is set up with type void
6902 (*)(SSL *, int, int), as described somewhere. However, when calling the
6903 function (actually a macro) that sets it up, the type void(*)() is
6904 expected. I've put in a cast to prevent warnings from picky compilers.
6906 34. If a DNS black list lookup found a CNAME record, but there were no A
6907 records associated with the domain it pointed at, Exim crashed.
6909 35. If a DNS black list lookup returned more than one A record, Exim ignored
6910 all but the first. It now scans all returned addresses if a particular IP
6911 value is being sought. In this situation, the contents of the
6912 $dnslist_value variable are a list of all the addresses, separated by a
6915 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6916 to accept a host name and all its aliases if the forward lookup for any of
6917 them yielded the IP address of the incoming connection. Now it accepts only
6918 those names whose forward lookup yields the correct IP address. Any other
6919 names are discarded. This closes a loophole whereby a rogue DNS
6920 administrator could create reverse DNS records to break through a
6921 wildcarded host restriction in an ACL.
6923 37. If a user filter or a system filter that ran in a subprocess used any of
6924 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6925 the wrong values were passed to the pipe command ($thisaddress had the
6926 value of $0, $0 had the value of $1, etc). This bug was introduced by
6927 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6929 38. Improved the line breaking for long SMTP error messages from ACLs.
6930 Previously, if there was no break point between 40 and 75 characters, Exim
6931 left the rest of the message alone. Two changes have been made: (a) I've
6932 reduced the minimum length to 35 characters; (b) if it can't find a break
6933 point between 35 and 75 characters, it looks ahead and uses the first one
6934 that it finds. This may give the occasional overlong line, but at least the
6935 remaining text gets split now.
6937 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6938 file descriptors might be low, and that setting 1000 would always raise it.
6939 It turns out that in some environments, the limit is already over 1000 and
6940 that lowering it causes trouble. So now Exim takes care not to decrease it.
6942 40. When delivering a message, the value of $return_path is set to $sender_
6943 address at the start of routing (routers may change the value). By an
6944 oversight, this default was not being set up when an address was tested by
6945 -bt or -bv, which affected the outcome if any router or filter referred to
6948 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6949 log only when "message" or "log_message" are set. However, if one of the
6950 conditions was an address verification, or a call to a nested ACL, the
6951 messages generated by the underlying test were being passed through. This
6952 no longer happens. The underlying message is available in $acl_verify_
6953 message for both "message" and "log_message" expansions, so it can be
6954 passed through if needed.
6956 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6957 new expansion $bh_ to give the encoded byte string without charset
6958 translation. Translation happens only if iconv() is available; HAVE_ICONV
6959 indicates this at build time. HEADERS_CHARSET gives the charset to
6960 translate to; headers_charset can change it in the configuration, and
6961 "headers charset" can change it in an individual filter file.
6963 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6964 that creates RFC 2047 encoded "words" labels them as that charset instead
6965 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6966 expansion operator; (ii) when Exim creates a From: line for a local
6967 message; (iii) when a header line is rewritten to include a "phrase" part.
6969 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6970 buggy, causing it to skip the first lines of messages whose message ID
6971 ended in 'D'. This would not have bitten before Exim release 4.14, because
6972 message IDs were unlikely to end in 'D' before then. The effect was to have
6973 incorrect size information for certain domains.
6975 45. #include "config.h" was missing at the start of the crypt16.c module. This
6976 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6979 46. If there was a timeout during a "random" callout check, Exim treated it as
6980 a failure of the random address, and carried on sending RSET and the real
6981 address. If the delay was just some slowness somewhere, the response to the
6982 original RCPT would be taken as a response to RSET and so on, causing
6983 mayhem of various kinds.
6985 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6986 when I implemented it. It didn't allow for the fact that some option values
6987 may legitimately be negative (e.g. size_addition), and it didn't even do
6988 the right test for positive values.
6990 48. Domain names in DNS records are case-independent. Exim always looks them up
6991 in lower case. Some resolvers return domain names in exactly the case they
6992 appear in the zone file, that is, they may contain uppercase letters. Not
6993 all resolvers do this - some return always lower case. Exim was treating a
6994 change of case by a resolver as a change of domain, similar to a widening
6995 of a domain abbreviation. This triggered its re-routing code and so it was
6996 trying to route what was effectively the same domain again. This normally
6997 caused routing to fail (because the router wouldn't handle the domain
6998 twice). Now Exim checks for this case specially, and just changes the
6999 casing of the domain that it ultimately uses when it transmits the message
7002 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
7005 50. If a filter generated a file delivery with a non-absolute name (possible if
7006 no home directory exists for the router), the forbid_file option was not
7009 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
7010 the existing equality matching.
7012 52. Exim was using ints instead of ino_t variables in some places where it was
7013 dealing with inode numbers.
7015 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
7016 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
7017 called TMPDIR, and if it finds it is different, it changes its value.
7019 54. The smtp_printf() function is now made available to local_scan() so
7020 additional output lines can be written before returning. There is also an
7021 smtp_fflush() function to enable the detection of a dropped connection.
7022 The variables smtp_input and smtp_batched_input are exported to
7025 55. Changed the default runtime configuration: the message "Unknown user"
7026 has been removed from the ACL, and instead placed on the localuser router,
7027 using the cannot_route_message feature. This means that any verification
7028 failures that generate their own messages won't get overridden. Similarly,
7029 the "Unrouteable address" message that was in the ACL for unverifiable
7030 relay addresses has also been removed.
7032 56. Added hosts_avoid_esmtp to the smtp transport.
7034 57. The exicyclog script was not checking for the esoteric option
7035 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7036 will work only if exicyclog is run under the appropriate euid.
7038 58. Following a discussion on the list, the rules by which Exim recognises line
7039 endings on incoming messages have been changed. The -dropcr and drop_cr
7040 options are now no-ops, retained only for backwards compatibility. The
7041 following line terminators are recognized: LF CRLF CR. However, special
7042 processing applies to CR:
7044 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7045 nor a local message in the state where . is a terminator.
7047 (ii) If a bare CR is encountered in a header line, an extra space is added
7048 after the line terminator so as not to end the header. The reasoning
7049 behind this is that bare CRs in header lines are most likely either
7050 to be mistakes, or people trying to play silly games.
7052 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7053 was being incorrectly given as 18 bytes larger than it should have been.
7054 This is a VOB (very old bug).
7056 60. This may never have affected anything current, but just in case it has:
7057 When the local host is found other than at the start of a list of hosts,
7058 the local host, those with the same MX, and any that follow, are discarded.
7059 When the list in question was part of a longer list of hosts, the following
7060 hosts (not currently being processed) were also being discarded. This no
7061 longer happens. I'm not sure if this situation could ever has previously
7064 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7067 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7068 header. This is something that is recommended in a new Internet Draft, and
7069 is something that is documented as being done by Sendmail. There are two
7070 possible values. For messages generated by the autoreply transport, Exim
7073 Auto-Submitted: auto-replied
7075 whereas for all other generated messages (e.g. bounces) it adds
7077 Auto-Submitted: auto-generated
7079 63. The "personal" condition in filters now includes a test for the
7080 Auto-Submitted: header. If it contains the string "auto-" the message it
7081 not considered personal.
7083 64. Added rcpt_include_affixes as a generic transport option.
7085 65. Added queue_only_override (default true).
7087 66. Added the syslog_duplication option.
7089 67. If what should have been the first header line of a message consisted of
7090 a space followed by a colon, Exim was mis-interpreting it as a header line.
7091 It isn't of course - it is syntactically invalid and should therefore be
7092 treated as the start of the message body. The misbehaviour could have
7093 caused a number of strange effects, including loss of data in subsequent
7094 header lines, and spool format errors.
7096 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7097 client host had authenticated. This control can now be exercised by an ACL
7098 for more flexibility.
7100 69. By default, callouts do not happen when testing with -bh. There is now a
7101 variant, -bhc, which does actually run the callout code, including
7102 consulting and updating the callout cache.
7104 70. Added support for saslauthd authentication, courtesy of Alexander
7107 71. If statvfs() failed on the spool or log directories while checking their
7108 size for availability, Exim confusingly gave the error "space shortage".
7109 Furthermore, in debugging mode it crashed with a floating point exception.
7110 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7111 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7112 serious problem, Exim now writes to the main and panic logs when this
7113 happens, with details of the failure. It then refuses to accept the
7114 incoming message, giving the message "spool directory problem" or "log
7115 directory problem" with a 421 code for SMTP messages.
7117 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7118 0, 1, and 2 exist, because some OS complain for execs without them (see
7119 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7120 descriptors. However, the code omitted to check that the open succeeded,
7121 causing mysterious errors if for some reason the permissions on /dev/null
7122 got screwed. Now Exim writes a message to the main and panic logs, and
7123 bombs out if it can't open /dev/null.
7125 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7126 interact so that it is all more flexible. It is supposed to remain
7127 backwards compatible. Also added extra_local_interfaces.
7129 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7130 to bomb out with an assertion failure - to the client this appears as a
7131 connection drop. This problem occurs in the part of the code that was taken
7132 from the Samba project. Fortunately, the assertion is in a very simple
7133 function, so I have fixed this by reproducing the function inline in the
7134 one place where it is called, and arranging for authentication to fail
7135 instead of killing the process with assert().
7137 75. The SPA client code was not working when the server requested OEM rather
7138 than Unicode encoding.
7140 76. Added code to make require_files with a specific uid setting more usable in
7141 the case where statting the file as root fails - usually a non-root-mounted
7142 NFS file system. When this happens and the failure is EACCES, Exim now
7143 forks a subprocess and does the per-uid checking as the relevant uid.
7145 77. Added process_log_path.
7147 78. If log_file_path was not explicitly set, a setting of check_log_space or
7148 check_log_inodes was ignored.
7150 79. If a space check for the spool or log partitions fails, the incident is now
7151 logged. Of course, in the latter case the data may get lost...
7153 80. Added the %p formatting code to string_format() so that it can be used to
7154 print addresses in debug_print(). Adjusted all the address printing in the
7155 debugging in store.c to use %p rather than %d.
7157 81. There was a concern that a line of code in smtp_in.c could overflow a
7158 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7159 initially expressed, the concern was not well-founded, because trailing
7160 spaces are removed early. However, if the trailing spaces were followed by
7161 a NULL, they did not get removed, so the overflow was possible. Two fixes
7164 (a) I re-wrote the offending code in a cleaner fashion.
7165 (b) If an incoming SMTP command contains a NULL character, it is rejected
7168 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7169 runs initgroups(), so that if the Exim user is in any additional groups,
7170 they will be used during message reception.
7176 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob