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
207 JH/01 Rework error string handling in TLS interface so that the caller in
208 more cases is responsible for logging. This permits library-sourced
209 string to be attached to addresses during delivery, and collapses
210 pairs of long lines into single ones.
212 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
213 during configuration. Wildcards are allowed and expanded.
215 JH/02 Rework error string handling in DKIM to pass more info back to callers.
216 This permits better logging.
218 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
219 do not close it down and have the child transport start it up again on
220 the passed-on TCP connection. Instead, proxy the child (and any
221 subsequent ones) for TLS via a unix-domain socket channel. Logging is
222 affected: the continued delivery log lines do not have any DNSSEC, TLS
223 Certificate or OCSP information. TLS cipher information is still logged.
225 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
226 identical IP addresses on different listening ports. Will also affect
229 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
230 add noisy ifdef guards to special-case this sillyness.
231 Patch from Bernd Kuhls.
233 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
234 than 255 are no longer allowed.
236 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
237 Disable the session-cache too, which might reduce our load. Since we
238 currrectly use a new context for every connection, both as server and
239 client, there is no benefit for these.
240 GnuTLS appears to not support tickets server-side by default (we don't
241 call gnutls_session_ticket_enable_server()) but client side is enabled
242 by default on recent versions (3.1.3 +) unless the PFS priority string
245 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
246 <https://reproducible-builds.org/specs/source-date-epoch/>.
248 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
249 the check for any unsuccessful recipients did not notice the limit, and
250 erroneously found still-pending ones.
252 JH/08 Pipeline CHUNKING command and data together, on kernels that support
253 MSG_MORE. Only in-clear (not on TLS connections).
255 JH/09 Avoid using a temporary file during transport using dkim. Unless a
256 transport-filter is involved we can buffer the headers in memory for
257 creating the signature, and read the spool data file once for the
258 signature and again for transmission.
260 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
261 4.77 as the kernel support then wasn't solid, having issues in 64bit
262 mode. Now, it's been long enough. Add support for FreeBSD also.
264 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
265 case where the routing stage had gathered several addresses to send to
266 a host before calling the transport for the first, we previously failed
267 to close down TLS in the old transport process before passing the TCP
268 connection to the new process. The new one sent a STARTTLS command
269 which naturally failed, giving a failed delivery and bloating the retry
270 database. Investigation and fix prototype from Wolfgang Breyha.
272 JH/12 Fix check on SMTP command input synchronisation. Previously there were
273 false-negatives in the check that the sender had not preempted a response
274 or prompt from Exim (running as a server), due to that code's lack of
275 awareness of the SMTP input buffering.
277 PP/04 Add commandline_checks_require_admin option.
278 Exim drops privileges sanely, various checks such as -be aren't a
279 security problem, as long as you trust local users with access to their
280 own account. When invoked by services which pass untrusted data to
281 Exim, this might be an issue. Set this option in main configuration
282 AND make fixes to the calling application, such as using `--` to stop
285 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
286 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
287 responses to those, into a single TLS record each way (this usually means
288 a single packet). As a side issue, smtp_enforce_sync now works on TLS
291 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
292 affects you only if you're dancing at the edge of the param size limits.
293 If you are, and this message makes sense to you, then: raise the
294 configured limit or use OpenSSL 1.1. Nothing we can do for older
297 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
298 cmdline element to get the documented default one. Previously it was
301 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
302 are unsuable for non-message-specific reasons - eg. connection timeout,
303 banner-time rejection.
305 JH/16 Fix logging of delivery remote port, when specified by router, under
308 PP/06 Repair manualroute's ability to take options in any order, even if one
309 is the name of a transport.
312 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
314 JH/17 Change the list-building routines interface to use the expanding-string
315 triplet model, for better allocation and copying behaviour.
317 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
318 Previously it was constructed the first time a possibly-matching string
319 was met in the configuration file input during startup; now it is done
322 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
323 compatible one, to avoid the (poorly documented) possibility of a config
324 file in the working directory redirecting the DB files, possibly correpting
325 some existing file. CVE-2017-10140 assigned for BDB.
327 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
328 cache-hot. Previously, although the result was properly cached, the
329 initial verify call returned a defer.
331 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
332 the main verify for receipient in uncached-mode.
334 JH/22 Retire historical build files to an "unsupported" subdir. These are
335 defined as "ones for which we have no current evidence of testing".
337 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
338 if present. Previously it was ignored.
340 JH/24 Start using specified-initialisers in C structure init coding. This is
341 a C99 feature (it's 2017, so now considered safe).
343 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
344 if was a fixed-sized field and bitmask ops via macros; it is now more
347 PP/07 GitHub PR 56: Apply MariaDB build fix.
348 Patch provided by Jaroslav Škarvada.
350 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
351 during Coverity cleanups [4.87 JH/47]
352 Diagnosis and fix provided by Michael Fischer v. Mollard.
354 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
355 the right size to place the terminating semicolon on its own folded
356 line, the header hash was calculated to an incorrect value thanks to
357 the (relaxed) space the fold became.
359 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
360 and confused the parent.
362 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
363 which could crash as a result. This could lead to undeliverable messages.
365 JH/28 Logging: "next input sent too soon" now shows where input was truncated
368 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
369 matters on fast-turnover and PID-randomising systems, which were getting
370 out-of-order delivery.
372 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
373 a possibly-overlapping copy. The symptom was that "Remote host closed
374 connection in response to HELO" was logged instead of the actual 4xx
377 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
378 Previously only that bufferd was discarded, resulting in SYMTP command
381 JH/32 DKIM: when a message has multiple signatures matching an identity given
382 in dkim_verify_signers, run the dkim acl once for each. Previously only
383 one run was done. Bug 2189.
385 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
386 "panic the current process" to "deliberately defer". The panic log is
387 still written with the problem list name; the mail and reject logs now
388 get a temp-reject line for the message that was being handled, saying
389 something like "domains check lookup or other defer". The SMTP 451
390 message is still "Temporary local problem".
392 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
393 A crafted sequence of BDAT commands could result in in-use memory beeing
394 freed. CVE-2017-16943.
396 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
397 from SMTP input. Previously it was always done; now only done for DATA
398 and not BDAT commands. CVE-2017-16944.
400 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
401 to the message (such as an overlong header line). Previously this was
402 not done and we did not exit BDAT mode. Followon from the previous item
403 though a different problem.
409 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
410 than -2003 did; needs libidn2 in addition to libidn.
412 JH/02 The path option on a pipe transport is now expanded before use.
414 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
415 Patch provided by "Björn", documentation fix added too.
417 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
418 missing a wire-to-host endian conversion.
420 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
421 close after a BDAT command line could be taken as a following command,
422 giving a synch failure. Fix by only checking for synch immediately
423 before acknowledging the chunk.
425 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
426 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
427 macro. Patches provided by Josh Soref.
429 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
430 Previously we did not; the RFC seems ambiguous and VRFY is not listed
431 by IANA as a service extension. However, John Klensin suggests that we
434 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
435 the dkim code may be unix-mode line endings rather than smtp wire-format
436 CRLF, so prepend a CR to any bare LF.
438 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
439 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
441 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
442 the first were themselves being wrongly included in the feed into dkim
443 processing; with most chunk sizes in use this resulted in an incorrect
444 body hash calculated value.
446 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
447 DKIM signature block, for verification. Although advised against by
448 standards it is specifically not ruled illegal.
450 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
452 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
453 missing a body hash (the bh= tag).
455 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
456 It seems that HAProxy sends the Proxy Protocol information in clear and
457 only then does a TLS startup, so do the same.
459 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
460 TCP connections (such as for Spamd) unless the daemon successfully set
461 Fast Open mode on its listening sockets. This fixes breakage seen on
462 too-old kernels or those not configured for Fast Open, at the cost of
463 requiring both directions being enabled for TFO, and TFO never being used
464 by non-daemon-related Exim processes.
466 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
467 endings, at least on the first header line. Try to canonify any that get
468 past that check, despite the cost.
470 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
471 now limited to an arbitrary five deep, while parsing addresses with the
472 strip_excess_angle_brackets option enabled.
474 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
475 instead leave the unprompted TLS handshake in socket buffer for the
476 TLS library to consume.
478 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
480 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
482 JH/16 Drop variables when they go out of scope. Memory management drops a whole
483 region in one operation, for speed, and this leaves assigned pointers
484 dangling. Add checks run only under the testsuite which checks all
485 variables at a store-reset and panics on a dangling pointer; add code
486 explicitly nulling out all the variables discovered. Fixes one known
487 bug: a transport crash, where a dangling pointer for $sending_ip_address
488 originally assigned in a verify callout, is re-used.
490 PP/06 Drop '.' from @INC in various Perl scripts.
492 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
494 PP/08 Reduce a number of compilation warnings under clang; building with
495 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
496 should be warning-free.
498 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
500 HS/01 Fix portability problems introduced by PP/08 for platforms where
501 realloc(NULL) is not equivalent to malloc() [SunOS et al].
503 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
504 chunk. This allows us to accept broken chunked messages. We need a more
505 general solution here.
507 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
508 already-broken messages in the queue.
510 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
512 JH/19 Fix reference counting bug in routing-generated-address tracking.
518 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
519 supports it and a size is available (ie. the sending peer gave us one).
521 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
522 years of being deprecated). The replacements are the ACLs
523 acl_smtp_mime and acl_not_smtp_mime.
525 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
526 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
527 or even an in-clear connection were permitted. Now, if the host lookup
528 was dnssec and dane was requested then the host is only used if the
529 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
530 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
531 if one fails this test.
532 This means that a poorly-configured remote DNS will make it incommunicado;
533 but it protects against a DNS-interception attack on it.
535 JH/04 Bug 1810: make continued-use of an open smtp transport connection
536 non-noisy when a race steals the message being considered.
538 JH/05 If main configuration option tls_certificate is unset, generate a
539 self-signed certificate for inbound TLS connections.
541 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
542 in rewrites and routers.
544 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
545 and logged a warning sing 4.83; now they are a configuration file error.
547 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
548 (lacking @domain). Apply the same qualification processing as RCPT.
550 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
552 JH/10 Support ${sha256:} applied to a string (as well as the previous
555 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
556 a cutthrough deliver is pending, as we always want to make a connection.
557 This also avoids re-routing the message when later placing the cutthrough
558 connection after a verify cache hit.
559 Do not update it with the verify result either.
561 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
562 when routing results in more than one destination address.
564 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
565 signing (which inhibits the cutthrough capability). Previously only
566 the presence of an option was tested; now an expansion evaluating as
567 empty is permissible (obviously it should depend only on data available
568 when the cutthrough connection is made).
570 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
571 the relevant preceding SMTP command did not note the pipelining mode.
573 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
574 Previously they were not counted.
576 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
577 as one having no matching records. Previously we deferred the message
578 that needed the lookup.
580 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
581 distinguished as "(=".
583 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
584 for missing MX records. Previously it only worked for missing A records.
586 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
588 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
589 after the data-go-ahead and data-ack. Patch from Jason Betts.
591 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
592 even for a "none" policy. Patch from Tony Meyer.
594 JH/22 Fix continued use of a connection for further deliveries. If a port was
595 specified by a router, it must also match for the delivery to be
598 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
599 When one of the recipients of a message was unsuitable for the connection
600 (has no matching addresses), we lost track of needing to mark it
601 deferred. As a result mail would be lost.
603 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
605 JH/25 Decoding ACL controls is now done using a binary search; the source code
606 takes up less space and should be simpler to maintain. Merge the ACL
607 condition decode tables also, with similar effect.
609 JH/26 Fix problem with one_time used on a redirect router which returned the
610 parent address unchanged. A retry would see the parent address marked as
611 delivered, so not attempt the (identical) child. As a result mail would
614 JH/27 Fix a possible security hole, wherein a process operating with the Exim
615 UID can gain a root shell. Credit to http://www.halfdog.net/ for
616 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
619 JH/28 Enable {spool,log} filesystem space and inode checks as default.
620 Main config options check_{log,spool}_{inodes,space} are now
621 100 inodes, 10MB unless set otherwise in the configuration.
623 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
624 Previously it only applied to the main-section connection policy
627 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
629 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
630 by me. Added RFC7919 DH primes as an alternative.
632 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
633 are not in the system include path.
635 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
636 GnuTLS, when a session startup failed (eg because the client disconnected)
637 Exim did stdio operations after fclose. This was exposed by a recent
638 change which nulled out the file handle after the fclose.
640 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
641 signed directly by the cert-signing cert, rather than an intermediate
642 OCSP-signing cert. This is the model used by LetsEncrypt.
644 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
646 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
647 an incoming connection.
649 HS/02 Bug 1802: Do not half-close the connection after sending a request
652 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
653 fallback to "prime256v1".
655 JH/34 SECURITY: Use proper copy of DATA command in error message.
656 Could leak key material. Remotely exploitable. CVE-2016-9963.
662 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
663 and 3.4.4 - once the server is enabled to respond to an OCSP request
664 it does even when not requested, resulting in a stapling non-aware
665 client dropping the TLS connection.
667 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
668 support variable-length bit vectors. No functional change.
670 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
671 The I= interface field on outgoing lines is now after the H= remote
672 host field, same as incoming lines. There is a separate
673 outgoing_interface log selector which allows you to disable the
676 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
677 If not running log_selector +smtp_connection the mainlog would be held
678 open indefinitely after a "too many connections" event, including to a
679 deleted file after a log rotate. Leave the per net connection logging
680 leaving it open for efficiency as that will be quickly detected by the
681 check on the next write.
683 HS/01 Bug 1671: Fix post transport crash.
684 Processing the wait-<transport> messages could crash the delivery
685 process if the message IDs didn't exist for some reason. When
686 using 'split_spool_directory=yes' the construction of the spool
687 file name failed already, exposing the same netto behaviour.
689 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
690 mime_regex ACL conditions.
692 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
693 to DSN fail messages (bounces): remote IP, remote greeting, remote response
694 to HELO, local diagnostic string.
696 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
697 log line to debug. Even when configured with a tls authenticator many
698 client connections are expected to not authenticate in this way, so
699 an authenticate fail is not an error.
701 HS/02 Add the Exim version string to the process info. This way exiwhat
702 gives some more detail about the running daemon.
704 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
705 matter for fast-change records such as DNSBLs.
707 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
708 retry record, even if constant. There may be multiple transports with
709 different interface settings and the retry behaviour needs to be kept
712 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
714 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
716 JH/10 Bug 840: fix log_defer_output option of pipe transport
718 JH/11 Bug 830: use same host for all RCPTS of a message, even under
719 hosts_randomize. This matters a lot when combined with mua_wrapper.
721 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
722 ${quote_pgsql:<string>} operator.
724 JH/13 Bug 1708: avoid misaligned access in cached lookup.
726 JH/14 Change header file name for freeradius-client. Relevant if compiling
727 with Radius support; from the Gentoo tree and checked under Fedora.
729 JH/15 Bug 1712: Introduce $prdr_requested flag variable
731 JH/16 Bug 1714: Permit an empty string as expansion result for transport
732 option transport_filter, meaning no filtering.
734 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
736 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
737 defaults to "*" (all hosts). The variable is now available when not built
738 with TLS, default unset, mainly to enable keeping the testsuite sane.
739 If a server certificate is not supplied (via tls_certificate) an error is
740 logged, and clients will find TLS connections fail on startup. Presumably
741 they will retry in-clear.
742 Packagers of Exim are strongly encouraged to create a server certificate
743 at installation time.
745 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
746 with the $config_file variable.
748 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
749 in transport context, after the attempt, and per-recipient. The latter type
750 is per host attempted. The event data is the error message, and the errno
751 information encodes the lookup type (A vs. MX) used for the (first) host,
752 and the trailing two digits of the smtp 4xx response.
754 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
755 to write to mainlog (or rejectlog, paniclog) in the window between file
756 creation and permissions/ownership being changed. Particularly affects
757 installations where exicyclog is run as root, rather than exim user;
758 result is that the running daemon panics and dies.
760 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
762 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
763 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
764 "pri" and "weight". Note that the previous implicit priority given by the
765 list order is no longer honoured.
767 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
770 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
771 by defining SUPPORT_SOCKS.
773 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
774 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
775 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
776 variables are renamed to proxy_{local,external}_{address,port}.
778 JH/27 Move Internationalisation support from Experimental to mainline, enabled
779 for a build by defining SUPPORT_I18N
781 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
782 of the query string, and make ${quote_redis:} do that quoting.
784 JH/29 Move Events support from Experimental to mainline, enabled by default
785 and removable for a build by defining DISABLE_EVENT.
787 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
789 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
790 cached by the daemon.
792 JH/32 Move Redis support from Experimental to mainline, enabled for a build
793 by defining LOOKUP_REDIS. The libhiredis library is required.
795 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
796 keys are given for lookup.
798 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
799 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
800 only supported when built with TLS support. The PolarSSL SHA routines
801 are still used when the TLS library is too old for convenient support.
803 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
804 openssl_options), for security. OpenSSL forces this from version 1.1.0
805 server-side so match that on older versions.
807 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
808 allocation for $value could be released as the expansion processing
809 concluded, but leaving the global pointer active for it.
811 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
812 and to use the domains and local_parts ACL conditions.
814 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
815 incorrectly not doubled on cutthrough transmission, hence seen as a
816 body-termination at the receiving system - resulting in truncated mails.
817 Commonly the sender saw a TCP-level error, and retransmitted the message
818 via the normal store-and-forward channel. This could result in duplicates
819 received - but deduplicating mailstores were liable to retain only the
820 initial truncated version.
822 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
824 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
826 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
827 we're in there, support oversigning also; bug 1309.
829 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
831 HS/04 Add support for keep_environment and add_environment options.
833 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
834 either intentional arithmetic overflow during PRNG, or testing config-
837 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
838 delivery resulted in actual delivery. Cancel cutthrough before DATA
841 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
842 rejects a recipient: pass the reject to the originator.
844 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
845 Many were false-positives and ignorable, but it's worth fixing the
848 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
849 for the new environment-manipulation done at startup. Move the routines
850 from being local to tls.c to being global via the os.c file.
852 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
853 an extract embedded as result-arg for a map, the first arg for extract
854 is unavailable so we cannot tell if this is a numbered or keyed
855 extraction. Accept either.
861 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
864 JH/02 The smtp transport option "multi_domain" is now expanded.
866 JH/03 The smtp transport now requests PRDR by default, if the server offers
869 JH/04 Certificate name checking on server certificates, when exim is a client,
870 is now done by default. The transport option tls_verify_cert_hostnames
871 can be used to disable this per-host. The build option
872 EXPERIMENTAL_CERTNAMES is withdrawn.
874 JH/05 The value of the tls_verify_certificates smtp transport and main options
875 default to the word "system" to access the system default CA bundle.
876 For GnuTLS, only version 3.0.20 or later.
878 JH/06 Verification of the server certificate for a TLS connection is now tried
879 (but not required) by default. The verification status is now logged by
880 default, for both outbound TLS and client-certificate supplying inbound
883 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
886 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
887 Status Notification (bounce) messages are now MIME format per RFC 3464.
888 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
889 under the control of the dsn_advertise_hosts option, and routers may
890 have a dsn_lasthop option.
892 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
893 default, modifiable by a malware= option. The list separator for
894 the options can now be changed in the usual way. Bug 68.
896 JH/10 The smtp_receive_timeout main option is now expanded before use.
898 JH/11 The incoming_interface log option now also enables logging of the
899 local interface on delivery outgoing connections.
901 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
902 if the interface and destination host and port all match.
904 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
907 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
908 Patch from Andrew Lewis.
910 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
911 now supports optional time-restrictions, weighting, and priority
912 modifiers per server. Patch originally by <rommer@active.by>.
914 JH/16 The spamd_address main option now supports a mixed list of local
915 and remote servers. Remote servers can be IPv6 addresses, and
916 specify a port-range.
918 JH/17 Bug 68: The spamd_address main option now supports an optional
919 timeout value per server.
921 JH/18 Bug 1581: Router and transport options headers_add/remove can
922 now have the list separator specified.
924 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
927 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
930 JH/21 Support for the A6 type of dns record is withdrawn.
932 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
933 rather than the verbs used.
935 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
936 from 255 to 1024 chars.
938 JH/24 Verification callouts now attempt to use TLS by default.
940 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
941 are generic router options now. The defaults didn't change.
943 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
944 Original patch from Alexander Shikoff, worked over by JH.
946 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
947 files if ZCAT_COMMAND is not executable.
949 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
951 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
953 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
954 Normally benign, it bites when the pair was led to by a CNAME;
955 modern usage is to not canonicalize the domain to a CNAME target
956 (and we were inconsistent anyway for A-only vs AAAA+A).
958 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
960 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
961 when evaluating $sender_host_dnssec.
963 JH/31 Check the HELO verification lookup for DNSSEC, adding new
964 $sender_helo_dnssec variable.
966 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
968 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
970 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
972 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
973 documented as working, but never had. Support all but $spam_report.
975 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
976 added for tls authenticator.
978 HS/03 Add perl_taintmode main config option
984 TL/01 When running the test suite, the README says that variables such as
985 no_msglog_check are global and can be placed anywhere in a specific
986 test's script, however it was observed that placement needed to be near
987 the beginning for it to behave that way. Changed the runtest perl
988 script to read through the entire script once to detect and set these
989 variables, reset to the beginning of the script, and then run through
990 the script parsing/test process like normal.
992 TL/02 The BSD's have an arc4random API. One of the functions to induce
993 adding randomness was arc4random_stir(), but it has been removed in
994 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
995 function when detected.
997 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
998 cause callback expansion.
1000 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
1001 syntax errors in an expansion can be treated as a string instead of
1002 logging or causing an error, due to the internal use of bool_lax
1003 instead of bool when processing it.
1005 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
1006 server certificates when making smtp deliveries.
1008 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
1010 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
1012 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
1014 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
1015 Merged patch from Sebastian Wiedenroth.
1017 JH/05 Fix results-pipe from transport process. Several recipients, combined
1018 with certificate use, exposed issues where response data items split
1019 over buffer boundaries were not parsed properly. This eventually
1020 resulted in duplicates being sent. This issue only became common enough
1021 to notice due to the introduction of connection certificate information,
1022 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1024 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1025 size buffer was used, resulting in syntax errors when an expansion
1028 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1029 version 3.3.6 or later.
1031 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1032 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1033 both become "event_action", the variables become $event_name, $event_data
1034 and $event_defer_errno. There is a new variable $verify_mode, usable in
1035 routers, transports and related events. The tls:cert event is now also
1036 raised for inbound connections, if the main configuration event_action
1039 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1040 early OCSP support, but no stapling (appears to be less than 1.0.0).
1042 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1043 server certificate names available under the smtp transport option
1044 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1047 JH/10 Time-related extraction expansions from certificates now use the main
1048 option "timezone" setting for output formatting, and are consistent
1049 between OpenSSL and GnuTLS compilations. Bug 1541.
1051 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1052 encoded parameter in the incoming message. Bug 1558.
1054 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1055 include certificate info, eximon was claiming there were spoolfile
1058 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1060 JH/14 Log delivery-related information more consistently, using the sequence
1061 "H=<name> [<ip>]" wherever possible.
1063 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1064 are problematic for Debian distribution, omit them from the release
1067 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1069 JH/16 Fix string representation of time values on 64bit time_t architectures.
1072 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1073 output list separator was used.
1078 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1079 checkers that were complaining about end of non-void function with no
1082 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1083 This was a regression introduced in 4.83 by another bugfix.
1085 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1087 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1088 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1094 TF/01 Correctly close the server side of TLS when forking for delivery.
1096 When a message was received over SMTP with TLS, Exim failed to clear up
1097 the incoming connection properly after forking off the child process to
1098 deliver the message. In some situations the subsequent outgoing
1099 delivery connection happened to have the same fd number as the incoming
1100 connection previously had. Exim would try to use TLS and fail, logging
1101 a "Bad file descriptor" error.
1103 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1104 utilities have not been installed.
1106 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1107 temporary space as the ACL may create new global variables.
1109 TL/01 LDAP support uses per connection or global context settings, depending
1110 upon the detected version of the libraries at build time.
1112 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1113 to extract and use the src ip:port in logging and expansions as if it
1114 were a direct connection from the outside internet. PPv2 support was
1115 updated based on HAProxy spec change in May 2014.
1117 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1119 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1120 Properly escape header and check for NULL return.
1122 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1125 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1127 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1128 characters in header names, implemented as a verify condition.
1129 Contributed by Michael Fischer v. Mollard.
1131 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1132 results permerror and temperror. Previous values are deprecated but
1133 still accepted. In a future release, err_perm and err_temp will be
1134 completely removed, which will be a backward incompatibility if the
1135 ACL tests for either of these two old results. Patch contributed by
1136 user bes-internal on the mailing list.
1138 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1140 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1141 selectors, in both main and reject logs.
1143 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1146 JH/07 Add malware type "sock" for talking to simple daemon.
1148 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1150 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1151 routers/transports under cutthrough routing.
1153 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1154 numbers. Touch up "bool" conditional to keep the same definition.
1156 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1158 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1160 JH/12 Expand items in router/transport headers_add or headers_remove lists
1161 individually rather than the list as a whole. Bug 1452.
1163 Required for reasonable handling of multiple headers_ options when
1164 they may be empty; requires that headers_remove items with embedded
1165 colons must have them doubled (or the list-separator changed).
1167 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1168 view the policy declared in the DMARC record. Currently, $dmarc_status
1169 is a combined value of both the record presence and the result of the
1172 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1174 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1175 dnslookup router and the smtp transport (applying to the forward
1178 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1179 of ldap servers used for a specific lookup. Patch provided by Heiko
1182 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1183 New variable $lookup_dnssec_authenticated for observability.
1185 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1186 Patch submitted by Lars Timman.
1188 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1190 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1191 Requires trusted mode and valid format message id, aborts otherwise.
1192 Patch contributed by Heiko Schlichting.
1194 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1195 certextract with support for various fields. Bug 1358.
1197 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1198 is requested by default, modifiable by smtp transport option
1201 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1202 operate on certificate variables to give certificate fingerprints
1203 Also new ${sha256:cert_variable}.
1205 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1207 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1208 Christian Aistleitner.
1210 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1212 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1213 file. Patch from Wolfgang Breyha.
1215 JH/25 Expand the coverage of the delivery $host and $host_address to
1216 client authenticators run in verify callout. Bug 1476.
1218 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1219 align with daemon_smtp_ports. Bug 72.
1221 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1222 support and error reporting did not work properly.
1224 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1225 and is readable. Patch from Andrew Colin Kissa.
1227 TL/14 Enhance documentation of ${run expansion and how it parses the
1228 commandline after expansion, particularly in the case when an
1229 unquoted variable expansion results in an empty value.
1231 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1233 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1234 by renaming away from it. Addresses GH issue 15, reported by
1237 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1239 TL/15 SECURITY: prevent double expansion in math comparison functions
1240 (can expand unsanitized data). Not remotely exploitable.
1247 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1249 PP/02 Make -n do something, by making it not do something.
1250 When combined with -bP, the name of an option is not output.
1252 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1255 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1256 $sender_host_name and config options to manage this, and basic check
1259 PP/05 DSCP support for outbound connections and control modifier for inbound.
1261 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1262 (Only plugin which currently uses this is kerberos4, which nobody should
1263 be using, but we should make it available and other future plugins might
1264 conceivably use it, even though it would break NAT; stuff *should* be
1265 using channel bindings instead).
1267 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1268 name; added for Sendmail compatibility; requires admin caller.
1269 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1270 just ignore it); requires trusted caller.
1271 Also parse but ignore: -Ac -Am -X<logfile>
1274 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1276 TL/02 Add +smtp_confirmation as a default logging option.
1278 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1279 Patch by Magnus Holmgren from 2007-02-20.
1281 TL/04 Bugzilla 1281 - Spec typo.
1282 Bugzilla 1283 - Spec typo.
1283 Bugzilla 1290 - Spec grammar fixes.
1285 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1287 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1289 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1290 mailing list by by Dmitry Isaikin.
1292 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1294 JH/02 Support "G" suffix to numbers in ${if comparisons.
1296 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1298 NM/01 Bugzilla 1197 - Spec typo
1299 Bugzilla 1196 - Spec examples corrections
1301 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1303 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1304 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1307 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1308 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1310 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1311 "acl {{name}{arg}...}", and optional args on acl condition
1314 JH/05 Permit multiple router/transport headers_add/remove lines.
1316 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1318 JH/07 Avoid using a waiting database for a single-message-only transport.
1319 Performance patch from Paul Fisher. Bugzilla 1262.
1321 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1324 JH/09 Add $headers_added variable, with content from use of ACL modifier
1325 add_header (but not yet added to the message). Bugzilla 199.
1327 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1328 Pulled from Bugzilla 817 by Wolfgang Breyha.
1330 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1332 (nb: this is the same fix as in Exim 4.80.1)
1334 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1337 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1340 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1342 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1343 advertise SMTP AUTH mechanism to us, instead of a generic
1344 protocol violation error. Also, make Exim more robust to bad
1345 data from the Dovecot auth socket.
1347 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1349 When a queue runner is handling a message, Exim first routes the
1350 recipient addresses, during which it prunes them based on the retry
1351 hints database. After that it attempts to deliver the message to
1352 any remaining recipients. It then updates the hints database using
1355 So if a recipient address works intermittently, it can get repeatedly
1356 deferred at routing time. The retry hints record remains fresh so the
1357 address never reaches the final cutoff time.
1359 This is a fairly common occurrence when a user is bumping up against
1360 their storage quota. Exim had some logic in its local delivery code
1361 to deal with this. However it did not apply to per-recipient defers
1362 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1364 This change adds a proper retry rule check during routing so that the
1365 final cutoff time is checked against the message's age. We only do
1366 this check if there is an address retry record and there is not a
1367 domain retry record; this implies that previous attempts to handle
1368 the address had the retry_use_local_parts option turned on. We use
1369 this as an approximation for the destination being like a local
1370 delivery, as in LMTP.
1372 I suspect this new check makes the old local delivery cutoff check
1373 redundant, but I have not verified this so I left the code in place.
1375 TF/02 Correct gecos expansion when From: is a prefix of the username.
1377 Test 0254 submits a message to Exim with the header
1381 When I ran the test suite under the user fanf2, Exim expanded
1382 the header to contain my full name, whereas it should have added
1383 a Resent-Sender: header. It erroneously treats any prefix of the
1384 username as equal to the username.
1386 This change corrects that bug.
1388 GF/01 DCC debug and logging tidyup
1389 Error conditions log to paniclog rather than rejectlog.
1390 Debug lines prefixed by "DCC: " to remove any ambiguity.
1392 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1394 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1395 Bug spotted by Jeremy Harris; was flawed since initial commit.
1396 Would have resulted in OCSP responses post-SNI triggering an Exim
1397 NULL dereference and crash.
1399 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1401 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1402 Bug detection, analysis and fix by Samuel Thibault.
1403 Bugzilla 1331, Debian bug #698092.
1405 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1407 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1408 Server implementation by Todd Lyons, client by JH.
1409 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1410 config variable "prdr_enable" controls whether the server
1411 advertises the facility. If the client requests PRDR a new
1412 acl_data_smtp_prdr ACL is called once for each recipient, after
1413 the body content is received and before the acl_smtp_data ACL.
1414 The client is controlled by both of: a hosts_try_prdr option
1415 on the smtp transport, and the server advertisement.
1416 Default client logging of deliveries and rejections involving
1417 PRDR are flagged with the string "PRDR".
1419 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1420 fclose(). Diagnosis by Todd Lyons.
1422 PP/17 Update configure.default to handle IPv6 localhost better.
1423 Patch by Alain Williams (plus minor tweaks).
1426 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1427 This is now consistent with GnuTLS, and is now documented: the
1428 previous undocumented portable approach to treating the option as
1429 unset was to force an expansion failure. That still works, and
1430 an empty string is now equivalent.
1432 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1433 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1434 not performing validation itself.
1436 PP/20 Added force_command boolean option to pipe transport.
1437 Patch from Nick Koston, of cPanel Inc.
1439 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1442 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1444 PP/21 Fix eximon continuous updating with timestamped log-files.
1445 Broken in a format-string cleanup in 4.80, missed when I repaired the
1446 other false fix of the same issue.
1447 Report and fix from Heiko Schlichting.
1450 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1451 Report from Prashanth Katuri.
1453 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1454 It's SecureTransport, so affects any MacOS clients which use the
1455 system-integrated TLS libraries, including email clients.
1457 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1458 using a MIME ACL for non-SMTP local injection.
1459 Report and assistance in diagnosis by Warren Baker.
1461 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1463 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1465 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1466 last id that failed so it may be referenced in subsequent ACL's.
1468 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1471 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1472 ldap library initialization, allowing self-signed CA's to be
1473 used. Also properly sets require_cert option later in code by
1474 using NULL (global ldap config) instead of ldap handle (per
1475 session). Bug diagnosis and testing by alxgomz.
1477 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1478 the src/util/ subdirectory.
1480 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1481 renamed to Transport Post Delivery Action by Jeremy Harris, as
1484 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1485 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1486 redis_servers = needs to be configured which will be used by the redis
1487 lookup. Patch from Warren Baker, of The Packet Hub.
1489 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1491 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1492 hostname or reverse DNS when processing a host list. Used suggestions
1493 from multiple comments on this bug.
1495 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1497 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1498 Missed a few lines, added it to make the runtest require no keyboard
1501 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1502 contains upper case chars. Make router use caseful_local_part.
1504 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1505 support when GnuTLS has been built with p11-kit.
1511 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1513 This, or similar/improved, will also be change PP/11 of 4.82.
1519 PP/01 Handle short writes when writing local log-files.
1520 In practice, only affects FreeBSD (8 onwards).
1521 Bugzilla 1053, with thanks to Dmitry Isaikin.
1523 NM/01 Bugzilla 949 - Documentation tweak
1525 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1528 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1530 PP/02 Implemented gsasl authenticator.
1532 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1534 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1535 `pkg-config foo` for cflags/libs.
1537 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1538 with rest of GSASL and with heimdal_gssapi.
1540 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1541 `pkg-config foo` for cflags/libs for the TLS implementation.
1543 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1544 properties get this fed in as external SSF. A number of robustness
1545 and debugging improvements to the cyrus_sasl authenticator.
1547 PP/08 cyrus_sasl server now expands the server_realm option.
1549 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1550 Patch by Jeremy Harris.
1552 PP/10 Added dbmjz lookup type.
1554 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1556 PP/12 MAIL args handles TAB as well as SP, for better interop with
1557 non-compliant senders.
1558 Analysis and variant patch by Todd Lyons.
1560 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1561 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1562 Patch from Dirk Mueller <dmueller@suse.com>
1564 PP/13 tls_peerdn now print-escaped for spool files.
1565 Observed some $tls_peerdn in wild which contained \n, which resulted
1566 in spool file corruption.
1568 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1569 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1570 or write after TLS renegotiation, which otherwise led to messages
1573 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1574 as a tracking header (ie: a signed header comes before the signature).
1575 Patch from Wolfgang Breyha.
1577 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1578 comma-sep list; embedded commas doubled.
1580 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1582 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1584 Report and patch from Dmitry Banschikov.
1586 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1587 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1588 failures. We appear to now support TLS1.1+ with Exim.
1590 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1591 lets Exim select keys and certificates based upon TLS SNI from client.
1592 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1593 before an outbound SMTP session. New log_selector, +tls_sni.
1595 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1596 NULL dereference. Report and patch from Alun Jones.
1598 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1599 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1600 Not seeing resolver debug output on NetBSD, but suspect this is a
1601 resolver implementation change.
1603 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1604 Left warnings. Added "eximon gdb" invocation mode.
1606 PP/21 Defaulting "accept_8bitmime" to true, not false.
1608 PP/22 Added -bw for inetd wait mode support.
1610 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1611 locate the relevant includes and libraries. Made this the default.
1613 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1614 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1616 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1617 This may cause build issues on older platforms.
1619 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1620 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1621 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1622 Added SNI support via GnuTLS too.
1623 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1625 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1627 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1628 Applied second patch from Janne, implementing suggestion to default
1629 multiple-strings-in-record handling to match SPF spec.
1631 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1633 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1634 read-only, out of scope).
1635 Patch from Wolfgang Breyha, report from Stuart Northfield.
1637 PP/29 Fix three issues highlighted by clang analyser static analysis.
1638 Only crash-plausible issue would require the Cambridge-specific
1639 iplookup router and a misconfiguration.
1640 Report from Marcin Mirosław.
1642 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1644 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1645 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1646 As part of this, removing so much warning spew let me fix some minor
1647 real issues in debug logging.
1649 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1650 assignment on my part. Fixed.
1652 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1653 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1654 Janne Snabb (who went above and beyond: thank you).
1656 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1657 string otherwise requires a connection and a bunch more work and it's
1658 relatively easy to get wrong. Should also expose TLS library linkage
1661 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1662 64-bit ${eval} (JH/03).
1664 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1665 GNU libc to support some of the 64-bit stuff, should not lead to
1666 conflicts. Defined before os.h is pulled in, so if a given platform
1667 needs to override this, it can.
1669 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1670 protection layer was required, which is not implemented.
1671 Bugzilla 1254, patch from Wolfgang Breyha.
1673 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1674 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1675 tls_dhparam take prime identifiers. Also unbreak combination of
1676 OpenSSL+DH_params+TLSSNI.
1678 PP/39 Disable SSLv2 by default in OpenSSL support.
1684 PP/01 Solaris build fix for Oracle's LDAP libraries.
1685 Bugzilla 1109, patch from Stephen Usher.
1687 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1689 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1692 TF/02 Fix a couple more cases where we did not log the error message
1693 when unlink() failed. See also change 4.74-TF/03.
1695 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1696 lock up or crash if it happened to be inside a call to libc when it
1697 got a SIGUSR1 from exiwhat.
1699 The SIGUSR1 handler appends the current process status to the process
1700 log which is later printed by exiwhat. It used to use the general
1701 purpose logging code to do this, but several functions it calls are
1702 not safe for signals.
1704 The new output code in the SIGUSR1 handler is specific to the process
1705 log, and simple enough that it's easy to inspect for signal safety.
1706 Removing some special cases also simplifies the general logging code.
1707 Removing the spurious timestamps from the process log simplifies
1710 TF/04 Improved ratelimit ACL condition.
1712 The /noupdate option has been deprecated in favour of /readonly which
1713 has clearer semantics. The /leaky, /strict, and /readonly update modes
1714 are mutually exclusive. The update mode is no longer included in the
1715 database key; it just determines when the database is updated. (This
1716 means that when you upgrade Exim will forget old rate measurements.)
1718 Exim now checks that the per_* options are used with an update mode that
1719 makes sense for the current ACL. For example, when Exim is processing a
1720 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1721 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1722 must specify per_mail/readonly. If you omit the update mode it defaults to
1723 /leaky where that makes sense (as before) or /readonly where required.
1725 The /noupdate option is now undocumented but still supported for
1726 backwards compatibility. It is equivalent to /readonly except that in
1727 ACLs where /readonly is required you may specify /leaky/noupdate or
1728 /strict/noupdate which are treated the same as /readonly.
1730 A useful new feature is the /count= option. This is a generalization
1731 of the per_byte option, so that you can measure the throughput of other
1732 aggregate values. For example, the per_byte option is now equivalent
1733 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1735 The per_rcpt option has been generalized using the /count= mechanism
1736 (though it's more complicated than the per_byte equivalence). When it is
1737 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1738 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1739 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1740 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1741 Note that using per_rcpt with a non-readonly update mode in more than
1742 one ACL will cause the recipients to be double-counted. (The per_mail
1743 and per_byte options don't have this problem.)
1745 The handling of very low rates has changed slightly. If the computed rate
1746 is less than the event's count (usually one) then this event is the first
1747 after a long gap. In this case the rate is set to the same as this event's
1748 count, so that the first message of a spam run is counted properly.
1750 The major new feature is a mechanism for counting the rate of unique
1751 events. The new per_addr option counts the number of different
1752 recipients that someone has sent messages to in the last time period. It
1753 behaves like per_rcpt if all the recipient addresses are different, but
1754 duplicate recipient addresses do not increase the measured rate. Like
1755 the /count= option this is a general mechanism, so the per_addr option
1756 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1757 example, measure the rate that a client uses different sender addresses
1758 with the options per_mail/unique=$sender_address. There are further
1759 details in the main documentation.
1761 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1763 TF/06 Removed a few PCRE remnants.
1765 TF/07 Automatically extract Exim's version number from tags in the git
1766 repository when doing development or release builds.
1768 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1769 Bugzilla 879. Patch from Paul Fisher.
1771 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1772 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1775 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1777 PP/05 Variable $av_failed, true if the AV scanner deferred.
1778 Bugzilla 1078. Patch from John Horne.
1780 PP/06 Stop make process more reliably on build failure.
1781 Bugzilla 1087. Patch from Heiko Schlittermann.
1783 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1784 Bugzilla 1089. Patch from Heiko Schlittermann.
1786 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1787 Bugzilla 1131. Patch from Holger Weiß.
1789 PP/09 Handle IPv6 addresses with SPF.
1790 Bugzilla 860. Patch from Wolfgang Breyha.
1792 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1794 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1797 PP/11 match_* no longer expand right-hand-side by default.
1798 New compile-time build option, EXPAND_LISTMATCH_RHS.
1799 New expansion conditions, "inlist", "inlisti".
1801 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1803 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1805 PP/14 fix log_write() format string regression from TF/03.
1806 Bugzilla 1152. Patch from Dmitry Isaikin.
1812 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1814 PP/02 Harmonised TLS library version reporting; only show if debugging.
1815 Layout now matches that introduced for other libraries in 4.74 PP/03.
1817 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1819 PP/04 New "dns_use_edns0" global option.
1821 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1824 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1825 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1827 TK/01 Updated PolarSSL code to 0.14.2.
1828 Bugzilla 1097. Patch from Andreas Metzler.
1830 PP/07 Catch divide-by-zero in ${eval:...}.
1831 Fixes bugzilla 1102.
1833 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1836 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1837 format-string attack -- SECURITY: remote arbitrary code execution.
1839 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1840 time unintentionally subject to list matching rules, letting the header
1841 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1842 arbitrary string expansion). This allowed for information disclosure.
1844 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1845 INT_MIN/-1 -- value coerced to INT_MAX.
1851 NM/01 Workaround for PCRE version dependency in version reporting
1854 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1855 This fixes portability to compilers other than gcc, notably
1856 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1858 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1859 makefiles for portability to HP-UX and POSIX correctness.
1861 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1862 Also via indented variable definition in the Makefile.
1863 (Debugging by Oliver Heesakkers).
1865 PP/02 Restore caching of spamd results with expanded spamd_address.
1866 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1868 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1869 Improves build reliability. Fix from: Frank Elsner
1871 NM/02 Fix wide character breakage in the rfc2047 coding
1872 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1874 NM/03 Allow underscore in dnslist lookups
1875 Fixes bug 1026. Patch from Graeme Fowler
1877 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1878 Code patches from Adam Ciarcinski of NetBSD.
1880 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1883 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1884 is logged, to avoid truncation. Patch from John Horne.
1886 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1887 Patch from Jakob Hirsch.
1889 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1890 SQL string expansion failure details.
1891 Patch from Andrey Oktyabrski.
1893 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1894 Patch from Simon Arlott.
1896 PP/09 New lookups functionality failed to compile on old gcc which rejects
1897 extern declarations in function scope.
1898 Patch from Oliver Fleischmann
1900 PP/10 Use sig_atomic_t for flags set from signal handlers.
1901 Check getgroups() return and improve debugging.
1902 Fixed developed for diagnosis in bug 927 (which turned out to be
1905 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1906 Patch from Mark Zealey.
1908 PP/12 Bugzilla 1056: Improved spamd server selection.
1909 Patch from Mark Zealey.
1911 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1912 Based on patch from Heiko Schlittermann.
1914 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1915 Patch from Uwe Doering, sign-off by Michael Haardt.
1917 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1918 variable declaration deep within a block. Bug and patch from
1921 PP/15 lookups-Makefile IRIX compatibility coercion.
1923 PP/16 Make DISABLE_DKIM build knob functional.
1925 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1926 Patch by Simon Arlott
1928 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1929 variable argument macros. Our copy now differs from upstream.
1935 TF/01 Failure to get a lock on a hints database can have serious
1936 consequences so log it to the panic log.
1938 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1939 controlled using the smtp_confirmation log selector.
1941 TF/03 Include the error message when we fail to unlink a spool file.
1943 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1944 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1945 for maintaining out-of-tree patches for some time.
1947 PP/01 Bugzilla 139: Documentation and portability issues.
1948 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1949 Handle per-OS dynamic-module compilation flags.
1951 PP/02 Let /dev/null have normal permissions.
1952 The 4.73 fixes were a little too stringent and complained about the
1953 permissions on /dev/null. Exempt it from some checks.
1954 Reported by Andreas M. Kirchwitz.
1956 PP/03 Report version information for many libraries, including
1957 Exim version information for dynamically loaded libraries. Created
1958 version.h, now support a version extension string for distributors
1959 who patch heavily. Dynamic module ABI change.
1961 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1962 privilege escalation vulnerability whereby the Exim run-time user
1963 can cause root to append content of the attacker's choosing to
1966 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1969 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1970 If dropping privileges for untrusted macros, we disabled normal logging
1971 on the basis that it would fail; for the Exim run-time user, this is not
1972 the case, and it resulted in successful deliveries going unlogged.
1973 Fixed. Reported by Andreas Metzler.
1979 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1980 only prepend for the Resent-* case. Fixes regression introduced in
1981 Exim 4.70 by NM/22 for Bugzilla 607.
1983 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1984 increasing numbers of administrators be bitten by this.
1986 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1988 PP/03 Bugzilla 994: added openssl_options main configuration option.
1990 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1992 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1994 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1996 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1997 without a peer certificate, leading to a segfault because of an
1998 assumption that peers always have certificates. Be a little more
1999 paranoid. Problem reported by Martin Tscholak.
2001 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
2002 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
2003 NB: ClamAV planning to remove STREAM in "middle of 2010".
2004 CL also introduces -bmalware, various -d+acl logging additions and
2005 more caution in buffer sizes.
2007 PP/09 Implemented reverse_ip expansion operator.
2009 PP/10 Bugzilla 937: provide a "debug" ACL control.
2011 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
2013 PP/12 Bugzilla 973: Implement --version.
2015 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2017 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2019 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2021 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2022 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2023 ignore trailing whitespace.
2025 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2027 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2030 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2031 Notification from Dr Andrew Aitchison.
2033 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2034 ExtendedDetectionInfo response format.
2035 Notification from John Horne.
2037 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2040 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2041 XSL and documented dependency on system catalogs, with examples of how
2044 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2047 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2048 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2049 configuration file which is writeable by the Exim user or group.
2051 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2052 of configuration files to cover files specified with the -C option if
2053 they are going to be used with root privileges, not just the default
2056 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2057 option (effectively making it always true).
2059 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2060 files to be used while preserving root privileges.
2062 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2063 that rogue child processes cannot use them.
2065 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2066 run-time user, instead of root.
2068 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2069 Exim run-time user without dropping privileges.
2071 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2072 result string, instead of calling string_vformat() twice with the same
2075 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2076 for other users. Others should always drop root privileges if they use
2077 -C on the command line, even for a whitelisted configure file.
2079 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2081 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2087 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2088 $data_path, and $header_path variables; fixed documentation bugs and
2091 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2092 exipick to access non-standard spools, including the "frozen" queue
2095 NM/01 Bugzilla 965: Support mysql stored procedures.
2096 Patch from Alain Williams
2098 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2100 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2101 Patch from Andreas Metzler
2103 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2104 Patch from Kirill Miazine
2106 NM/05 Bugzilla 671: Added umask to procmail example.
2108 JJ/03 installed exipick 20100323.0, fixing doc bug
2110 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2111 directory. Notification and patch from Dan Rosenberg.
2113 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2115 TK/02 Improve log output when DKIM signing operation fails.
2117 MH/01 Treat the transport option dkim_domain as a colon separated
2118 list, not as a single string, and sign the message with each element,
2119 omitting multiple occurences of the same signer.
2121 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2122 Bugzilla 985, 986. Patch by Simon Arlott
2124 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2125 Patch by Simon Arlott
2127 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2128 MBX locking. Notification from Dan Rosenberg.
2134 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2136 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2138 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2140 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2142 NM/04 Bugzilla 915: Flag broken perl installation during build.
2148 TK/01 Added patch by Johannes Berg that expands the main option
2149 "spamd_address" if it starts with a dollar sign.
2151 TK/02 Write list of recipients to X-Envelope-Sender header when building
2152 the mbox-format spool file for content scanning (suggested by Jakob
2155 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2156 (http://www.dcc-servers.net/) support via dccifd. Activated by
2157 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2159 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2160 by Mark Daniel Reidel <mr@df.eu>.
2162 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2163 When building exim an external PCRE library is now needed -
2164 PCRE is a system library on the majority of modern systems.
2165 See entry on PCRE_LIBS in EDITME file.
2167 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2168 conversation. Added nologin parameter to request.
2169 Patch contributed by Kirill Miazine.
2171 TF/01 Do not log submission mode rewrites if they do not change the address.
2173 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2175 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2176 log files in place. Contributed by Roberto Lima.
2178 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2180 TF/03 Bugzilla 615: When checking the local_parts router precondition
2181 after a local_part_suffix or local_part_prefix option, Exim now
2182 does not use the address's named list lookup cache, since this
2183 contains cached lookups for the whole local part.
2185 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2186 Robert Millan. Documentation is in experimental-spec.txt.
2188 TF/04 Bugzilla 668: Fix parallel build (make -j).
2190 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2192 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2193 Patch provided by Jan Srzednicki.
2195 TF/05 Leading white space used to be stripped from $spam_report which
2196 wrecked the formatting. Now it is preserved.
2198 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2199 that they are available at delivery time.
2201 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2203 TF/08 TLS error reporting now respects the incoming_interface and
2204 incoming_port log selectors.
2206 TF/09 Produce a more useful error message if an SMTP transport's hosts
2207 setting expands to an empty string.
2209 NM/06 Bugzilla 744: EXPN did not work under TLS.
2210 Patch provided by Phil Pennock.
2212 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2213 Patch provided by Richard Godbee.
2215 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2216 acl_smtp_notquit, added index entry.
2218 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2219 Patch provided by Eugene Bujak.
2221 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2222 accept(). Patch provided by Maxim Dounin.
2224 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2225 Patch provided by Phil Pennock.
2227 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2229 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2230 Patch provided by Brad "anomie" Jorsch.
2232 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2233 Patch provided by Dean Brooks.
2235 TK/05 Add native DKIM support (does not depend on external libraries).
2237 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2238 Patch provided by Graeme Fowler.
2240 NM/16 Bugzilla 851: Documentation example syntax fix.
2242 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2244 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2247 NM/19 Bugzilla 745: TLS version reporting.
2248 Patch provided by Phil Pennock.
2250 NM/20 Bugzilla 167: bool: condition support.
2251 Patch provided by Phil Pennock.
2253 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2254 clients. Patch provided by Phil Pennock.
2256 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2257 Patch provided by Brad "anomie" Jorsch.
2259 NM/23 Bugzilla 687: Fix misparses in eximstats.
2260 Patch provided by Heiko Schlittermann.
2262 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2263 Patch provided by Heiko Schlittermann.
2265 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2266 plus update to original patch.
2268 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2270 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2271 Patch provided by David Brownlee.
2273 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2275 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2277 NM/30 Bugzilla 888: TLS documentation bugfixes.
2279 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2281 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2282 Unlike the original bugzilla I have changed all shell scripts in src tree.
2284 NM/33 Bugzilla 898: Transport filter timeout fix.
2285 Patch by Todd Rinaldo.
2287 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2288 Patch by Serge Demonchaux.
2290 NM/35 Bugzilla 39: Base64 decode bug fixes.
2291 Patch by Jakob Hirsch.
2293 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2295 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2297 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2299 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2305 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2306 ALT-N's libdkim that I have put here:
2307 http://duncanthrax.net/exim-experimental/
2309 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2310 were called 'true' and it seems that C99 defines that as a reserved
2311 keyword to be used with 'bool' variable types. That means you could
2312 not include C99-style headers which use bools without triggering
2313 build errors in sieve.c.
2315 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2316 as mailq or other aliases. Changed the --help handling significantly
2317 to do whats expected. exim_usage() emits usage/help information.
2319 SC/01 Added the -bylocaldomain option to eximstats.
2321 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2323 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2325 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2331 PH/01 Another patch from the Sieve maintainer.
2333 PH/02 When an IPv6 address is converted to a string for single-key lookup
2334 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2335 dots are used instead of colons so that keys in lsearch files need not
2336 contain colons. This was done some time before quoting was made available
2337 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2338 (notated using the quote facility) so as to distinguish them from IPv4
2339 keys. This meant that lookups for IP addresses in host lists did not work
2340 for iplsearch lookups.
2342 This has been fixed by arranging for IPv6 addresses to be expressed with
2343 colons if the lookup type is iplsearch. This is not incompatible, because
2344 previously such lookups could never work.
2346 The situation is now rather anomalous, since one *can* have colons in
2347 ordinary lsearch keys. However, making the change in all cases is
2348 incompatible and would probably break a number of configurations.
2350 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2353 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2354 conversion specification without a maximum field width, thereby enabling
2355 a rogue spamd server to cause a buffer overflow. While nobody in their
2356 right mind would setup Exim to query an untrusted spamd server, an
2357 attacker that gains access to a server running spamd could potentially
2358 exploit this vulnerability to run arbitrary code as the Exim user.
2360 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2361 $primary_hostname instead of what libspf2 thinks the hosts name is.
2363 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2364 a directory entry by the name of the lookup key. Previously, if a
2365 symlink pointed to a non-existing file or a file in a directory that
2366 Exim lacked permissions to read, a lookup for a key matching that
2367 symlink would fail. Now it is enough that a matching directory entry
2368 exists, symlink or not. (Bugzilla 503.)
2370 PH/03 The body_linecount and body_zerocount variables are now exported in the
2373 PH/04 Added the $dnslist_matched variable.
2375 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2376 This means they are set thereafter only if the connection becomes
2379 PH/06 Added the client_condition to authenticators so that some can be skipped
2380 by clients under certain conditions.
2382 PH/07 The error message for a badly-placed control=no_multiline_responses left
2383 "_responses" off the end of the name.
2385 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2387 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2388 (without spaces) instead of just copying the configuration text.
2390 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2392 PH/11 Added $max_received_linelength.
2394 PH/12 Added +ignore_defer and +include_defer to host lists.
2396 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2397 way in which PCRE > 7.0 is built.
2399 PH/14 Implemented queue_only_load_latch.
2401 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2402 MAIL command. The effect was to mangle the value on 64-bit systems.
2404 PH/16 Another patch from the Sieve maintainer.
2406 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2408 PH/18 If a system quota error occurred while trying to create the file for
2409 a maildir delivery, the message "Mailbox is full" was not appended to the
2410 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2411 only to a quota excession during the actual writing of the file.
2413 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2414 characters?) which causes problems in log lines. The DN values are now
2415 passed through string_printing() before being added to log lines.
2417 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2418 and InterBase are left for another time.)
2420 PH/21 Added message_body_newlines option.
2422 PH/22 Guard against possible overflow in moan_check_errorcopy().
2424 PH/23 POSIX allows open() to be a macro; guard against that.
2426 PH/24 If the recipient of an error message contained an @ in the local part
2427 (suitably quoted, of course), incorrect values were put in $domain and
2428 $local_part during the evaluation of errors_copy.
2434 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2435 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2438 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2439 issue a MAIL command.
2441 PH/02 In an ACL statement such as
2443 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2445 if a client was not listed at all, or was listed with a value other than
2446 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2447 the condition was not true (as it should be), so access was not denied.
2448 The bug was that the ! inversion was incorrectly passed on to the second
2449 item. This has been fixed.
2451 PH/03 Added additional dnslists conditions == and =& which are different from
2452 = and & when the dns lookup returns more than one IP address.
2454 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2455 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2457 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2458 FSYNC, which compiles an option called disable_fsync that allows for
2459 bypassing fsync(). The documentation is heavily laced with warnings.
2461 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2463 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2464 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2465 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2466 including adding "make clean"; (3) Added -fPIC when compiling the test
2467 dynamically loaded module, to get rid of a warning.
2469 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2470 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2471 = 0s. The bug is otherwise harmless.
2473 PH/07 There was a bug in the dovecot authenticator such that the value of
2474 $auth1 could be overwritten, and so not correctly preserved, after a
2475 successful authentication. This usually meant that the value preserved by
2476 the server_setid option was incorrect.
2478 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2480 PH/09 Installed PCRE release 7.0.
2482 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2483 run for batched SMTP input. It is now run at the start of every message
2484 in the batch. While fixing this I discovered that the process information
2485 (output by running exiwhat) was not always getting set for -bs and -bS
2486 input. This is fixed, and it now also says "batched" for BSMTP.
2488 PH/11 Added control=no_pipelining.
2490 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2491 patch, slightly modified), and move the expansion of helo_data till after
2492 the connection is made in the smtp transport (so it can use these
2495 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2497 PH/14 Added log_selector = +pid.
2499 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2501 PH/16 Add ${if forany and ${if forall.
2503 PH/17 Added dsn_from option to vary the From: line in DSNs.
2505 PH/18 Flush SMTP output before performing a callout, unless control =
2506 no_callout_flush is set.
2508 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2509 was true (the default) a successful delivery failed to delete the retry
2510 item, thus causing premature timeout of the address. The bug is now
2513 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2515 PH/21 Long custom messages for fakedefer and fakereject are now split up
2516 into multiline responses in the same way that messages for "deny" and
2517 other ACL rejections are.
2519 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2520 with slight modification.
2522 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2523 draft, changing the syntax and factoring some duplicate code".
2525 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2526 for deliveries of the second and subsequent messages over the same SMTP
2529 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2530 ${reduce, with only minor "tidies".
2532 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2534 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2535 expansion side effects.
2537 PH/27 When a message times out after an over-quota error from an Exim-imposed
2538 quota, the bounce message says "mailbox is full". This message was not
2539 being given when it was a system quota that was exceeded. It now should
2542 MH/03 Made $recipients available in local_scan(). local_scan() already has
2543 better access to the recipient list through recipients_list[], but
2544 $recipients can be useful in postmaster-provided expansion strings.
2546 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2547 in the case of a MAIL command with additional options following the
2548 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2549 were accidentally chopped off.
2551 PH/29 SMTP synchronization checks are implemented when a command is read -
2552 there is a check that no more input is waiting when there shouldn't be
2553 any. However, for some commands, a delay in an ACL can mean that it is
2554 some time before the response is written. In this time, more input might
2555 arrive, invalidly. So now there are extra checks after an ACL has run for
2556 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2557 pipelining has not been advertised.
2559 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2561 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2562 returned in the SMTP response when smtp_return_error_details was false.
2563 This has been fixed.
2565 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2566 instead of the C I/O that was originally supplied, because problems were
2567 reported on Solaris.
2569 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2570 Exim which did not show up earlier: it was assuming that a call to
2571 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2572 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2573 was a macro that became an assignment, so it seemed to work. This has
2574 changed to a proper function call with a void return, hence the compile
2575 error. Exim's code has been fixed.
2577 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2580 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2582 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2584 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2585 because it turns out to be too big or there is a timeout) there is no
2586 "Completed" line in the log. When some messages of this type were
2587 selected by exigrep, they were listed as "not completed". Others were
2588 picked up by some special patterns. I have improved the selection
2589 criteria to be more general.
2591 PH/38 The host_find_failed option in the manualroute router can now be set
2592 to "ignore", to completely ignore a host whose IP address cannot be
2593 found. If all hosts are ignored, the behaviour is controlled by the new
2594 host_all_ignored option.
2596 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2597 homing or because of multiple MX records with /mx) generated more than
2598 one IP address, and the following item turned out to be the local host,
2599 all the secondary addresses of the first item were incorrectly removed
2600 from the list, along with the local host and any following hosts (which
2601 is what is supposed to happen).
2603 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2604 whoever called Exim into the -H file. In the case of the daemon it was
2605 behaving confusingly. When first started, it used values for whoever
2606 started the daemon, but after a SIGHUP it used the Exim user (because it
2607 calls itself on a restart). I have changed the code so that it now always
2610 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2611 message are rejected with the same error (e.g. no authentication or bad
2612 sender address), and a DATA command is nevertheless sent (as can happen
2613 with PIPELINING or a stupid MUA), the error message that was given to the
2614 RCPT commands is included in the rejection of the DATA command. This is
2615 intended to be helpful for MUAs that show only the final error to their
2618 PH/42 Another patch from the Sieve maintainer.
2620 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2621 Eximstats - Fixed some broken HTML links and added missing column headers
2623 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2624 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2626 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2628 PH/43 Yet another patch from the Sieve maintainer.
2630 PH/44 I found a way to check for a TCP/IP connection going away before sending
2631 the response to the final '.' that terminates a message, but only in the
2632 case where the client has not sent further data following the '.'
2633 (unfortunately, this is allowed). However, in many cases there won't be
2634 any further data because there won't be any more messages to send. A call
2635 to select() can be used: if it shows that the input is "ready", there is
2636 either input waiting, or the socket has been closed. An attempt to read
2637 the next input character can distinguish the two cases. Previously, Exim
2638 would have sent an OK response which the client would never have see.
2639 This could lead to message repetition. This fix should cure that, at
2640 least in a lot of common cases.
2642 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2643 advertised in response to EHLO.
2649 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2650 fixed by 4.65/MH/01 (is this a record?) are fixed:
2652 (i) An empty string was always treated as zero by the numeric comparison
2653 operators. This behaviour has been restored.
2655 (ii) It is documented that the numeric comparison operators always treat
2656 their arguments as decimal numbers. This was broken in that numbers
2657 starting with 0 were being interpreted as octal.
2659 While fixing these problems I realized that there was another issue that
2660 hadn't been noticed. Values of message_size_limit (both the global option
2661 and the transport option) were treated as octal if they started with 0.
2662 The documentation was vague. These values are now always treated as
2663 decimal, and I will make that clear in the documentation.
2669 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2670 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2673 MH/01 Don't check that the operands of numeric comparison operators are
2674 integers when their expansion is in "skipping" mode (fixes bug
2675 introduced by 4.64-PH/07).
2677 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2678 child addresses, Exim now panics and dies. Previously, because the count
2679 is held in a short int, deliveries were likely to be lost. As such a
2680 large number of recipients for a single message is ridiculous
2681 (performance will be very, very poor), I have chosen to impose a limit
2682 rather than extend the field.
2688 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2689 leftover -K file (the existence of which was triggered by #402).
2690 While we were at it, introduced process PID as part of the -K
2691 filename. This should rule out race conditions when creating
2694 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2695 processing considerably. Previous code took too long for large mails,
2696 triggering a timeout which in turn triggers #401.
2698 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2699 in the DK code in transports.c. sendfile() is not really portable,
2700 hence the _LINUX specificness.
2702 TF/01 In the add_headers option to the mail command in an Exim filter,
2703 there was a bug that Exim would claim a syntax error in any
2704 header after the first one which had an odd number of characters
2707 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2708 callout verification, Exim cached a "reject" for the entire domain. This
2709 is correct for most verifications, but it is not correct for a recipient
2710 verification with use_sender or use_postmaster set, because in that case
2711 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2712 case of MAIL FROM:<> rejection from other early rejections (e.g.
2713 rejection of HELO). When verifying a recipient using a non-null MAIL
2714 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2715 Whatever the result of the callout, the value of the domain cache is
2716 left unchanged (for any other kind of callout, getting as far as trying
2717 RCPT means that the domain itself is ok).
2719 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2722 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2723 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2724 session, the daemon ignores ECONNECT errors and logs others; it now
2725 ignores EPIPE as well.
2727 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2728 (quoted-printable decoding).
2730 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2731 later the small subsequent patch to fix an introduced bug.
2733 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2735 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2737 PH/08 An error is now given if message_size_limit is specified negative.
2739 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2740 to be given (somewhat) arbitrary names.
2742 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2745 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2746 miscellaneous code fixes
2748 PH/10 Added the log_reject_target ACL modifier to specify where to log
2751 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2752 hostname. This is wrong, because it relates to the incoming message (and
2753 probably the interface on which it is arriving) and not to the outgoing
2754 callout (which could be using a different interface). This has been
2755 changed to use the value of the helo_data option from the smtp transport
2756 instead - this is what is used when a message is actually being sent. If
2757 there is no remote transport (possible with a router that sets up host
2758 addresses), $smtp_active_hostname is used.
2760 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2761 tweaks were necessary in order to get it to work (see also 21 below):
2762 (a) The code assumed that strncpy() returns a negative number on buffer
2763 overflow, which isn't the case. Replaced with Exim's string_format()
2765 (b) There were several signed/unsigned issues. I just did the minimum
2766 hacking in of casts. There is scope for a larger refactoring.
2767 (c) The code used strcasecmp() which is not a standard C function.
2768 Replaced with Exim's strcmpic() function.
2769 (d) The code set only $1; it now sets $auth1 as well.
2770 (e) A simple test gave the error "authentication client didn't specify
2771 service in request". It would seem that Dovecot has changed its
2772 interface. Fortunately there's a specification; I followed it and
2773 changed what the client sends and it appears to be working now.
2775 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2778 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2779 address A is aliased to B and C, where B exists and C does not. Without
2780 -v the output is "A verified" because verification stops after a
2781 successful redirection if more than one address is generated. However,
2782 with -v the child addresses are also verified. Exim was outputting "A
2783 failed to verify" and then showing the successful verification for C,
2784 with its parentage. It now outputs "B failed to verify", showing B's
2785 parentage before showing the successful verification of C.
2787 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2788 look up a TXT record in a specific list after matching in a combined
2791 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2792 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2793 they consult the DNS. I had assumed they would set it the way they
2794 wanted; and indeed my experiments on Linux seem to show that in some
2795 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2796 To be on the safe side, however, I have now made the interface to
2797 host_find_byname() similar to host_find_bydns(), with an argument
2798 containing the DNS resolver options. The host_find_byname() function now
2799 sets these options at its start, just as host_find_bydns() does. The smtp
2800 transport options dns_qualify_single and dns_search_parents are passed to
2801 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2802 of host_find_byname() use the default settings of RES_DEFNAMES
2803 (qualify_single) but not RES_DNSRCH (search_parents).
2805 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2806 spool_read_header() do less string testing, by means of a preliminary
2807 switch on the second character of optional "-foo" lines. (This is
2808 overdue, caused by the large number of possibilities that now exist.
2809 Originally there were few.) While I was there, I also converted the
2810 str(n)cmp tests so they don't re-test the leading "-" and the first
2811 character, in the hope this might squeeze out yet more improvement.
2813 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2814 flag allowing group syntax was set by the header_syntax check but not
2815 turned off, possible causing trouble later; (2) The flag was not being
2816 set at all for the header_verify test, causing "group"-style headers to
2817 be rejected. I have now set it in this case, and also caused header_
2818 verify to ignore an empty address taken from a group. While doing this, I
2819 came across some other cases where the code for allowing group syntax
2820 while scanning a header line wasn't quite right (mostly, not resetting
2821 the flag correctly in the right place). These bugs could have caused
2822 trouble for malformed header lines. I hope it is now all correct.
2824 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2825 with the "reply" argument non-NULL. The code, however (which originally
2826 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2827 but it didn't always do it. This confused somebody who was copying the
2828 code for some other use. I have removed all the tests.
2830 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2831 feature that was used to support insecure browsers during the U.S. crypto
2832 embargo. It requires special client support, and Exim is probably the
2833 only MTA that supported it -- and would never use it because real RSA is
2834 always available. This code has been removed, because it had the bad
2835 effect of slowing Exim down by computing (never used) parameters for the
2836 RSA_EXPORT functionality.
2838 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2839 authenticator to fail if there's a tab character in the incoming data
2840 (there should never be unless someone is messing about, as it's supposed
2841 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2842 if the connection is using TLS or if the remote IP is the same as the
2843 local IP, and the "valid-client-cert option" if a client certificate has
2846 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2847 authenticators. This can be used for authorization after authentication
2848 succeeds. (In the case of plaintext, it servers for both authentication
2851 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2852 if any retry times were supplied.
2854 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2855 connection, where there is no client IP address to check. In this
2856 situation, the verify now always succeeds.
2858 PH/25 Applied John Jetmore's -Mset patch.
2860 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2862 PH/27 In a string expansion for a processed (not raw) header when multiple
2863 headers of the same name were present, leading whitespace was being
2864 removed from all of them, but trailing whitespace was being removed only
2865 from the last one. Now trailing whitespace is removed from each header
2866 before concatenation. Completely empty headers in a concatenation (as
2867 before) are ignored.
2869 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2870 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2872 PH/29 [Removed. This was a change that I later backed out, and forgot to
2873 correct the ChangeLog entry (that I had efficiently created) before
2874 committing the later change.]
2876 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2877 address errors (4xx response to RCPT) over the same connection as other
2878 messages routed to the same hosts. Such deliveries are always "forced",
2879 so retry times are not inspected. This resulted in far too many retries
2880 for the affected addresses. The effect occurred only when there were more
2881 hosts than the hosts_max_try setting in the smtp transport when it had
2882 the 4xx errors. Those hosts that it had tried were not added to the list
2883 of hosts for which the message was waiting, so if all were tried, there
2884 was no problem. Two fixes have been applied:
2886 (i) If there are any address or message errors in an SMTP delivery, none
2887 of the hosts (tried or untried) are now added to the list of hosts
2888 for which the message is waiting, so the message should not be a
2889 candidate for sending over the same connection that was used for a
2890 successful delivery of some other message. This seems entirely
2891 reasonable: after all the message is NOT "waiting for some host".
2892 This is so "obvious" that I'm not sure why it wasn't done
2893 previously. Hope I haven't missed anything, but it can't do any
2894 harm, as the worst effect is to miss an optimization.
2896 (ii) If, despite (i), such a delivery is accidentally attempted, the
2897 routing retry time is respected, so at least it doesn't keep
2898 hammering the server.
2900 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2901 in ${readsocket because some servers need this prod.
2903 PH/32 Added some extra debug output when updating a wait-xxx database.
2905 PH/33 The hint "could be header name not terminated by colon", which has been
2906 given for certain expansion errors for a long time, was not being given
2907 for the ${if def:h_colon_omitted{... case.
2909 PH/34 The spec says: "With one important exception, whenever a domain list is
2910 being scanned, $domain contains the subject domain." There was at least
2911 one case where this was not true.
2913 PH/35 The error "getsockname() failed: connection reset by peer" was being
2914 written to the panic log as well as the main log, but it isn't really
2915 panic-worthy as it just means the connection died rather early on. I have
2916 removed the panic log writing for the ECONNRESET error when getsockname()
2919 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2920 runs only) independently of the message's sender address. This meant
2921 that, if the 4xx error was in fact related to the sender, a different
2922 message to the same recipient with a different sender could confuse
2923 things. In particular, this can happen when sending to a greylisting
2924 server, but other circumstances could also provoke similar problems.
2925 I have changed the default so that the retry time for these errors is now
2926 based a combination of the sender and recipient addresses. This change
2927 can be overridden by setting address_retry_include_sender=false in the
2930 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2931 remote server are returned as part of bounce messages. This was not
2932 happening for LMTP over a pipe (the lmtp transport), but now it is the
2933 same for both kinds of LMTP.
2935 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2936 in header lines that were in fact CNAMEs. This is no longer the case.
2938 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2939 and queue runs started by the daemon processed all messages. This has
2940 been fixed so that -R and -S can now usefully be given with -q<time>.
2942 PH/40 Import PCRE release 6.7 (fixes some bugs).
2944 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2946 PH/42 Give an error if -q is specified more than once.
2948 PH/43 Renamed the variables $interface_address and $interface_port as
2949 $received_ip_address and $received_port, to make it clear that these
2950 values apply to message reception, and not to the outgoing interface when
2951 a message is delivered. (The old names remain recognized, of course.)
2953 PH/44 There was no timeout on the connect() call when using a Unix domain
2954 socket in the ${readsocket expansion. There now is.
2956 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2957 be meaningful with "accept".
2959 SC/01 Eximstats V1.43
2960 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2962 SC/02 Eximstats V1.44
2963 Use a glob alias rather than an array ref in the generated
2964 parser. This improves both readability and performance.
2966 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2967 Collect SpamAssassin and rejection statistics.
2968 Don't display local sender or destination tables unless
2969 there is data to show.
2970 Added average volumes into the top table text output.
2972 SC/04 Eximstats V1.46
2973 Collect data on the number of addresses (recipients)
2974 as well as the number of messages.
2976 SC/05 Eximstats V1.47
2977 Added 'Message too big' to the list of mail rejection
2978 reasons (thanks to Marco Gaiarin).
2980 SC/06 Eximstats V1.48
2981 Mainlog lines which have GMT offsets and are too short to
2982 have a flag are now skipped.
2984 SC/07 Eximstats V1.49 (Alain Williams)
2985 Added the -emptyok flag.
2987 SC/08 Eximstats V1.50
2988 Fixes for obtaining the IP address from reject messages.
2990 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2991 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2992 whitespace changes from 4.64-PH/27
2994 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2997 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2998 are found, allow negative numbers in numeric criteria)
3000 JJ/06 exipick.20061117.2, added new $message_body_missing variable
3002 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
3003 to match changes made in 4.64-PH/43
3005 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
3007 PH/47 Put in an explicit test for a DNS lookup of an address record where the
3008 "domain" is actually an IP address, and force a failure. This locks out
3009 those revolvers/nameservers that support "A-for-A" lookups, in
3010 contravention of the specifications.
3012 PH/48 When a host name was looked up from an IP address, and the subsequent
3013 forward lookup of the name timed out, the host name was left in
3014 $sender_host_name, contrary to the specification.
3016 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3017 restricted to single-key lookups, Exim was not diagnosing an error if
3018 * or *@ was used with a query-style lookup.
3020 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3022 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3023 long ago, but noone interested enough thought of it. Let's just say that
3024 the "1.1" means that there are some new functions that weren't there at
3025 some point in the past.
3027 PH/51 Error processing for expansion failure of helo_data from an smtp
3028 transport during callout processing was broken.
3030 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3031 tested/used via the -bh/-bhc/-bs options.
3033 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3034 bug, fixed in subsequent PCRE releases).
3036 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3037 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3039 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3045 SC/01 Use a glob alias rather than an array ref in eximstats generated
3046 parser. This improves both readability and performance.
3048 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3049 Don't display local sender or destination tables in eximstats unless
3050 there is data to show.
3051 Added average volumes into the eximstats top table text output.
3053 SC/03 Collect data on the number of addresses (recipients) as well
3054 as the number of messages in eximstats.
3056 TF/01 Correct an error in the documentation for the redirect router. Exim
3057 does (usually) call initgroups() when daemonizing.
3059 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3060 with consistent privilege compared to when running as a daemon.
3062 TF/03 Note in the spec that $authenticated_id is not set for local
3063 submissions from trusted users.
3065 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3066 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3068 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3069 by adding some example configuration directives to the default
3070 configuration file. A little bit of work is required to uncomment the
3071 directives and define how usernames and passwords are checked, but
3072 there is now a framework to start from.
3074 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3075 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3076 without this. I don't know how relevant this is to other LDAP libraries.
3078 PH/02 Add the verb name to the "unknown ACL verb" error.
3080 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3082 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3084 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3085 directory not expanded when it should be if an expanded home directory
3086 was set for the address (which is overridden by the transport).
3088 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3091 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3092 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3093 because it is too late at that time, and has no effect.
3095 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3096 security issue with \' (bugzilla #107). I could not use the
3097 PQescapeStringConn() function, because it needs a PGconn value as one of
3100 PH/08 When testing addresses using -bt, indicate those final addresses that
3101 are duplicates that would not cause an additional delivery. At least one
3102 person was confused, thinking that -bt output corresponded to deliveries.
3103 (Suppressing duplicates isn't a good idea as you lose the information
3104 about possibly different redirections that led to the duplicates.)
3106 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3107 systems where poll() doesn't work, in particular OS X.
3109 PH/10 Added more information to debugging output for retry time not reached.
3111 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3112 operations in malware.c.
3114 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3117 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3118 syslog_duplication set false, log lines that would normally be written
3119 both the the main log and to the reject log were not written to syslog at
3122 PH/14 In the default configuration, change the use of "message" in ACL warn
3123 statements to "add_header".
3125 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3126 not followed by a command (e.g. "seen endif").
3128 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3129 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3132 PH/17 Added extra conditions to the default value of delay_warning_condition
3136 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3137 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3138 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3141 The Auto-Submitted: and various List- headers are standardised, whereas I
3142 don't think Precedence: ever was.
3144 PH/18 Refactored debugging code in route_finduser() to show more information,
3145 in particular, the error code if getpwnam() issues one.
3147 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3148 This is apparently needed in addition to the PH/07 change above to avoid
3149 any possible encoding problems.
3151 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3152 but not after initializing Perl.
3154 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3155 output them only if debugging. By default they are written stderr,
3156 apparently, which is not desirable.
3158 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3161 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3164 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3166 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3167 authenticated or an ident call has been made. Suppress the default
3168 values for $authenticated_id and $authenticated_sender (but permit -oMai
3169 and -oMas) when testing with -bh.
3171 PH/24 Re-jigged the order of the tests in the default configuration so that the
3172 tests for valid domains and recipients precede the DNS black list and CSA
3173 tests, on the grounds that those ones are more expensive.
3175 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3176 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3177 This bug exists in every version of Exim that I still have, right back to
3180 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3181 However, an attempt to turn on case-sensitivity in a regex key by
3182 including (?-i) didn't work because the subject string was already
3183 lowercased, and the effects were non-intuitive. It turns out that a
3184 one-line patch can be used to allow (?-i) to work as expected.
3190 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3191 other effects) broke the use of negated acl sub-conditions.
3193 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3196 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3197 "Deny" causes Exim to reject the incoming connection with a 554 error.
3198 Unfortunately, if there is a major crisis, such as a disk failure,
3199 tcp-wrappers gives "deny", whereas what one would like would be some
3200 kind of temporary error. A kludge has been added to help with this.
3201 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3202 554 error is used if errno is still zero or contains ENOENT (which occurs
3203 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3206 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3208 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3209 errors. Otherwise a message that provokes a temporary error (when other
3210 messages do not) can cause a whole host to time out.
3212 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3213 addresses were routed directly to files or pipes from a redirect router.
3214 File deliveries just didn't batch; pipe deliveries might have suffered
3217 PH/06 A failure to get a lock for a hints database would erroneously always say
3218 "Failed to get write lock", even when it was really a read lock.
3220 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3221 of 0600. This has been changed to use the value of the lockfile_mode
3222 option (which defaults to 0600).
3224 PH/08 Applied small patch from the Sieve maintainer.
3226 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3227 folder from quota calculations, a direct delivery into this folder messed
3228 up the contents of the maildirsize file. This was because the regex was
3229 used only to exclude .Trash (or whatever) when the size of the mailbox
3230 was calculated. There was no check that a delivery was happening into an
3231 excluded directory. This bug has been fixed by ignoring all quota
3232 processing for deliveries into excluded directories.
3234 PH/10 Added the maildirfolder_create_regex option to appendfile.
3240 PH/01 The code for finding all the local interface addresses on a FreeBSD
3241 system running IPv6 was broken. This may well have applied to all BSD
3242 systems, as well as to others that have similar system calls. The broken
3243 code found IPv4 interfaces correctly, but gave incorrect values for the
3244 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3245 that it would not match correctly against @[] and not recognize the IPv6
3248 PH/02 The ipliteral router was not recognizing addresses of the form user@
3249 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3251 PH/03 Added disable_ipv6.
3253 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3254 decoded form, because it is most often used to construct To: headers
3255 lines in autoreplies, and the decoded form may well be syntactically
3256 invalid. However, $reply_address has leading white space removed, and all
3257 newlines turned into spaces so that the autoreply transport does not
3260 PH/05 If group was specified without a user on a router, and no group or user
3261 was specified on a transport, the group from the router was ignored.
3263 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3264 for visible compile-time settings that can be used to change these
3265 numbers, for those that want even more. Backwards compatibility with old
3266 spool files has been maintained. However, going back to a previous Exim
3267 release will lost any variables that are in spool files.
3269 PH/07 Two small changes when running in the test harness: increase delay when
3270 passing a TCP/IP connection to a new process, in case the original
3271 process has to generate a bounce, and remove special handling of
3272 127.0.0.2 (sic), which is no longer necessary.
3274 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3275 be the same on different OS.
3277 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3280 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3281 whether --show-vars was specified or not
3283 JJ/02 exipick: Added support for new ACL variable spool format introduced
3286 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3287 syntactically invalid From: or Reply-to: line, and a filter used this to
3288 generate an autoreply, and therefore failed to obtain an address for the
3289 autoreply, Exim could try to deliver to a non-existent relative file
3290 name, causing unrelated and misleading errors. What now happens is that
3291 it logs this as a hard delivery error, but does not attempt to create a
3294 PH/11 The exinext utility has a -C option for testing purposes, but although
3295 the given file was scanned by exinext itself; it wasn't being passed on
3296 when Exim was called.
3298 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3299 an end-of-file indication when reading a command response.
3301 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3302 compiled. In many other places in Exim, IPv6 addresses are always
3303 recognized, so I have changed this. It also means that IPv4 domain
3304 literals of the form [IPV4:n.n.n.n] are now always recognized.
3306 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3307 used if the router is not running as root, for example, when verifying at
3308 ACL time, or when using -bh. The debugging output from this situation was
3309 non-existent - all you got was a failure to exec. I have made two
3312 (a) Failures to set uid/gid, the current directory, or a process leader
3313 in a subprocess such as that created by queryprogram now generate
3314 suitable debugging output when -d is set.
3316 (b) The queryprogram router detects when it is not running as root,
3317 outputs suitable debugging information if -d is set, and then runs
3318 the subprocess without attempting to change uid/gid.
3320 PH/15 Minor change to Makefile for building test_host (undocumented testing
3323 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3324 additional section of a DNS packet that returns MX or SRV records.
3325 Instead, it always explicitly searches for A/AAAA records. This avoids
3326 major problems that occur when a DNS server includes only records of one
3327 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3328 fixed another bug: if SRV records were looked up and the corresponding
3329 address records were *not* found in the additional section, the port
3330 values from the SRV records were lost.
3332 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3333 using the correct key (the original address) when searching the retry
3334 rules in order to find which one to use for generating the retry hint.
3336 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3337 adding the default one. Similarly, if it contains a Reply-To: header, the
3338 errors_reply_to option, if set, is not used.
3340 PH/19 When calculating a retry time, Exim used to measure the "time since
3341 failure" by looking at the "first failed" field in the retry record. Now
3342 it does not use this if it is later than than the arrival time of the
3343 message. Instead it uses the arrival time. This makes for better
3344 behaviour in cases where some deliveries succeed, thus re-setting the
3345 "first failed" field. An example is a quota failure for a huge message
3346 when small messages continue to be delivered. Without this change, the
3347 "time since failure" will always be short, possible causing more frequent
3348 delivery attempts for the huge message than are intended.
3349 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3351 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3352 $1, $2, $3) because the numerical variables can be reset during some
3353 expansion items (e.g. "match"), thereby losing the authentication data.
3355 PH/21 Make -bV show the size of off_t variables so that the test suite can
3356 decide whether to run tests for quotas > 2G.
3358 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3359 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3360 filecount value is greater than 2G or if a quota value is greater than 2G
3361 on a system where the size of off_t is not greater than 4, a panic error
3364 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3365 never match. The debug and -bh output now contains an explicit error
3366 message indicating a malformed IPv4 address or mask.
3368 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3369 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3370 PH/23 above applies.
3372 PH/25 Do not write to syslog when running in the test harness. The only
3373 occasion when this arises is a failure to open the main or panic logs
3374 (for which there is an explicit test).
3376 PH/26 Added the /no_tell option to "control=freeze".
3378 PH/27 If a host name lookup failed very early in a connection, for example, if
3379 the IP address matched host_lookup and the reverse lookup yielded a name
3380 that did not have a forward lookup, an error message of the form "no IP
3381 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3382 could be logged. Now it outputs the IP address instead of "NULL".
3384 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3385 allows the sender and the authenticated sender to be set when
3386 submitting a message from within Exim. Since child_open_exim() is
3387 documented for local_scan(), the new function should be too.
3389 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3390 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3391 results in an empty string is now treated as unset.
3393 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3395 PH/31 Added sender_verify_fail logging option.
3397 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3398 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3399 tidied the source and removed it altogether.
3401 PH/33 When a queue run was abandoned because the load average was too high, a
3402 log line was always written; now it is written only if the queue_run log
3403 selector is set. In addition, the log line for abandonment now contains
3404 information about the queue run such as the pid. This is always present
3405 in "start" and "stop" lines but was omitted from the "abandon" line.
3407 PH/34 Omit spaces between a header name and the colon in the error message that
3408 is given when verify = headers_syntax fails (if there are lots of them,
3409 the message gets confusing).
3411 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3412 names, as there are now some PTR records that contain slashes. This check
3413 is only to protect against broken name servers that fall over on strange
3414 characters, so the fact that it applies to all lookups doesn't matter.
3416 PH/36 Now that the new test suite is complete, we can remove some of the
3417 special code in Exim that was needed for the old test suite. For example,
3418 sorting DNS records because real resolvers return them in an arbitrary
3419 order. The new test suite's fake resolver always returns records in the
3422 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3423 bounces) except when queue_only is set, to avoid logging races between
3424 the different processes.
3426 PH/38 Panic-die if .include specifies a non-absolute path.
3428 PH/39 A tweak to the "H" retry rule from its user.
3430 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3431 a label. They prevented compilation on older perls.
3433 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3434 a warning to be raised on newish perls.
3436 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3437 on queue. Changes to match documented behaviour of showing count of
3438 messages matching specified criteria.
3440 PH/40 Changed the default ident timeout from 30s to 5s.
3442 PH/41 Added support for the use of login_cap features, on those BSD systems
3443 that have them, for controlling the resources used by pipe deliveries.
3445 PH/42 The content-scanning code uses fopen() to create files in which to put
3446 message data. Previously it was not paying any attention to the mode of
3447 the files. Exim runs with umask(0) because the rest of the code creates
3448 files with open(), and sets the required mode explicitly. Thus, these
3449 files were ending up world-writeable. This was not a big issue, because,
3450 being within the spool directory, they were not world-accessible. I have
3451 created a function called modefopen, which takes an additional mode
3452 argument. It sets umask(777), creates the file, chmods it to the required
3453 mode, then resets the umask. All the relevant calls to fopen() in the
3454 content scanning code have been changed to use this function.
3456 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3457 to 24 hours. This avoids potential overflow problems when processing G
3458 and H retry rules. I suspect nobody ever tinkers with this value.
3460 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3462 PH/45 When the plaintext authenticator is running as a client, the server's
3463 challenges are checked to ensure they are valid base64 strings. By
3464 default, the authentication attempt is cancelled if an invalid string is
3465 received. Setting client_ignore_invalid_base64 true ignores these errors.
3466 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3467 they are received. Thus, the responses can be made to depend on the
3468 challenges. If an invalid string is ignored, an empty string is placed in
3471 PH/46 Messages that are created by the autoreply transport now contains a
3472 References: header, in accordance with RFCs 2822 and 3834.
3474 PH/47 Added authenticated_sender_force to the smtp transport.
3476 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3478 PH/49 Installed latest patch from the Sieve maintainer.
3480 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3481 was also set, the appendfile transport was unnecessarily scanning a
3482 directory of message files (e.g. for maildir delivery) to find the count
3483 of files (along with the size), even though it did not need this
3484 information. It now does the scan only if it needs to find either the
3485 size of the count of files.
3487 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3489 PH/52 Two bugs concerned with error handling when the smtp transport is
3492 (i) Exim was not creating retry information for temporary errors given
3493 for individual recipients after the DATA command when the smtp transport
3494 was used in LMTP mode. This meant that they could be retried too
3495 frequently, and not timed out correctly.
3497 (ii) Exim was setting the flag that allows error details to be returned
3498 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3499 recipients that were returned after the DATA command.
3501 PH/53 This is related to PH/52, but is more general: for any failing address,
3502 when detailed error information was permitted to be returned to the
3503 sender, but the error was temporary, then after the final timeout, only
3504 "retry timeout exceeded" was returned. Now it returns the full error as
3505 well as "retry timeout exceeded".
3507 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3508 do this, and (what is worse) MTAs that accept it.
3510 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3511 will now be deprecated.
3513 PH/56 New os.c-cygwin from the Cygwin maintainer.
3515 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3516 formats (previously only available in exim formats via -bpr, -bpru,
3517 and -bpra. Now also available in native and exiqgrep formats)
3519 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3520 with very large, slow to parse queues
3522 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3524 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3526 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3527 responses to authentication challenges, though it was showing the
3528 challenges; (ii) I've removed the CR characters from the debug output for
3531 PH/58 Allow for the insertion of a newline as well as a space when a string
3532 is turned into more than one encoded-word during RFC 2047 encoding. The
3533 Sieve code now uses this.
3535 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3536 data_4xx, lost_connection, tls_required.
3538 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3539 message was being sent as an SMTP response.
3541 PH/61 Add -l and -k options to exicyclog.
3543 PH/62 When verifying, if an address was redirected to one new address, so that
3544 verification continued, and the new address failed or deferred after
3545 having set something in $address_data, the value of $address_data was not
3546 passed back to the ACL. This was different to the case when no
3547 redirection occurred. The value is now passed back in both cases.
3549 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3550 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3551 use login_cap.h, so on its own it isn't the distinguishing feature. The
3552 new name refers directly to the setclassresources() function.
3554 PH/65 Added configuration files for NetBSD3.
3556 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3558 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3559 is preferred over IPv4.
3561 PH/68 The bounce_return_message and bounce_return_body options were not being
3562 honoured for bounces generated during the reception of non-SMTP messages.
3563 In particular, this applied to messages rejected by the ACL. This bug has
3564 been fixed. However, if bounce_return_message is true and bounce_return_
3565 body is false, the headers that are returned for a non-SMTP message
3566 include only those that have been read before the error was detected.
3567 (In the case of an ACL rejection, they have all been read.)
3569 PH/69 The HTML version of the specification is now built in a directory called
3570 spec_html instead of spec.html, because the latter looks like a path with
3571 a MIME-type, and this confuses some software.
3573 PH/70 Catch two compiler warnings in sieve.c.
3575 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3576 function verify_get_ident() calls ip_connect() to connect a socket, but
3577 if the "connect()" function timed out, ip_connect() used to close the
3578 socket. However, verify_get_ident() also closes the socket later, and in
3579 between Exim writes to the log, which may get opened at this point. When
3580 the socket was closed in ip_connect(), the log could get the same file
3581 descriptor number as the socket. This naturally causes chaos. The fix is
3582 not to close the socket in ip_connect(); the socket should be closed by
3583 the function that creates it. There was only one place in the code where
3584 this was missing, in the iplookup router, which I don't think anybody now
3585 uses, but I've fixed it anyway.
3587 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3588 well as to direct DNS lookups. Otherwise the handling of names in host
3589 lists is inconsistent and therefore confusing.
3595 PH/01 Two changes to the default runtime configuration:
3597 (1) Move the checks for relay_from_hosts and authenticated clients from
3598 after to before the (commented out) DNS black list checks.
3600 (2) Add control=submission to the relay_from_hosts and authenticated
3601 clients checks, on the grounds that messages accepted by these
3602 statements are most likely to be submissions.
3604 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3606 (1) Generate an error if the third argument for the ${prvs expansion is
3609 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3612 (3) Reset the variables that are obtained from the first argument of
3613 ${prvscheck and used in the second argument before leaving the code,
3614 because their memory is reclaimed, so using them afterwards may do
3617 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3618 one (it's much easier than Tom thought :-).
3620 (5) Because of (4), we can now allow for the use of $prvscheck_result
3621 inside the third argument.
3623 PH/03 For some reason, the default setting of PATH when running a command from
3624 a pipe transport was just "/usr/bin". I have changed it to
3627 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3628 anything to be listed in the output from -bV.
3630 PH/05 When a filter generated an autoreply, the entire To: header line was
3631 quoted in the delivery log line, like this:
3633 => >A.N.Other <ano@some.domain> <original@ddress> ...
3635 This has been changed so that it extracts the operative address. There
3636 may be more than one such address. If so, they are comma-separated, like
3639 => >ano@some.domain,ona@other.domain <original@ddress> ...
3641 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3642 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3643 not being looked up in the rDNS to get a host name, Exim was showing the
3644 IP address twice in Received: lines, even though the IP addresses were
3645 identical. For example:
3647 Received: from [1.2.3.4] (helo=[1.2.3.4])
3649 However, if the real host name was known, it was omitting the HELO data
3650 if it matched the actual IP address. This has been tidied up so that it
3651 doesn't show the same IP address twice.
3653 PH/07 When both +timestamp and +memory debugging was on, the value given by
3654 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3655 called by the debug printing, thereby overwriting the timestamp buffer.
3656 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3658 PH/08 When the original message was included in an autoreply transport, it
3659 always said "this is a copy of the message, including all the headers",
3660 even if body_only or headers_only was set. It now gives an appropriate
3663 PH/09 Applied a patch from the Sieve maintainer which:
3665 o fixes some comments
3666 o adds the (disabled) notify extension core
3667 o adds some debug output for the result of if/elsif tests
3668 o points to the current vacation draft in the documentation
3669 and documents the missing references header update
3673 o fixes a bug in processing the envelope test (when testing
3674 multiple envelope elements, the last element determined the
3677 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3678 Electronic Mail") by including:
3680 Auto-submitted: auto-generated
3682 in the messages that it generates (bounce messages and others, such as
3683 warnings). In the case of bounce messages for non-SMTP messages, there was
3684 also a typo: it was using "Auto_submitted" (underscore instead of
3685 hyphen). Since every message generated by Exim is necessarily in response
3686 to another message, thes have all been changed to:
3688 Auto-Submitted: auto-replied
3690 in accordance with these statements in the RFC:
3692 The auto-replied keyword:
3694 - SHOULD be used on messages sent in direct response to another
3695 message by an automatic process,
3697 - MUST NOT be used on manually-generated messages,
3699 - MAY be used on Delivery Status Notifications (DSNs) and Message
3700 Disposition Notifications (MDNs),
3702 - MUST NOT be used on messages generated by automatic or periodic
3703 processes, except for messages which are automatic responses to
3706 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3707 to the default Received: header definition.
3709 PH/12 Added log selector acl_warn_skipped (default on).
3711 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3712 variables because (a) they are in the wrong storage pool and (b) even if
3713 they were copied, it wouldn't work properly because of the caching.
3715 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3716 checking when decoding. Apparently there are clients that generate
3717 overlong encoded strings. Why am I not surprised?
3719 PH/15 If the first argument of "${if match_address" was not empty, but did not
3720 contain an "@" character, Exim crashed. Now it writes a panic log message
3721 and treats the condition as false.
3723 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3725 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3726 extension "envelope-auth". The code is finished and in agreement with
3727 other implementations, but there is no documentation so far and in fact,
3728 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3729 not changing the active code.
3731 Print executed "if" and "elsif" statements when debugging is used. This
3732 helps a great deal to understand what a filter does.
3734 Document more things not specified clearly in RFC3028. I had all this
3735 sorted out, when out of a sudden new issues came to my mind. Oops."
3737 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3740 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3741 canonical form (as documented). However, after a host name lookup from
3742 the IP address, check_host() was doing a simple string comparison with
3743 addresses acquired from the DNS when checking that the found name did
3744 have the original IP as one of its addresses. Since any found IPv6
3745 addresses are likely to be in abbreviated form, the comparison could
3746 fail. Luckily, there already exists a function for doing the comparison
3747 by converting both addresses to binary, so now that is used instead of
3748 the text comparison.
3750 PH/20 There was another similar case to PH/19, when a complete host name was
3751 given in a host list; looking up its IP address could give an abbreviated
3752 form, whereas the current host's name might or might not be abbreviated.
3753 The same fix has been applied.
3759 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3760 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3763 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3764 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3766 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3768 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3769 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3770 Also, for "Auto-Submitted", treat anything other than "no" as
3771 non-personal, in accordance with RFC 3834. (Previously it treated
3772 anything starting "auto-" as non-personal.)
3774 TF/01 The control=submission/name=... option had a problem with syntax
3775 errors if the name included a slash character. The /name= option
3776 now slurps the rest of the string, so it can include any characters
3777 but it must come last in the list of options (after /sender_retain
3780 PH/05 Some modifications to the interface to the fake nameserver for the new
3788 TK/01 Added the "success_on_redirect" address verification option. See
3789 NewStuff for rationale and an example.
3791 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3793 PH/02 Patch to exigrep to allow it to work on syslog lines.
3795 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3796 fread() to skip over the body file's header line, because in Cygwin the
3797 header line is locked and is inaccessible.
3799 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3800 co-exist for some time) to make it clear that it is the Exim ID that is
3801 referenced, not the Message-ID: header line.
3803 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3804 string_format() function, because snprintf() does not exist on all
3807 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3808 ${stat: expansion item.
3810 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3811 protocol synchronization error", to keep the pedants happy.
3813 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3814 well as for IRIX systems, when gcc is being used. See the host.c source
3817 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3819 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3822 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3823 transport and to the smtp transport in LMTP mode.
3825 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3827 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3828 run a filter in a subprocess. This could lead to confusion in subsequent
3829 lookups in the parent process. There should also be a search_tidyup() at
3830 the end of the subprocess.
3832 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3833 only if the host matched helo_try_verify_hosts, which caused the
3834 verification to occur when the EHLO/HELO command was issued. The ACL just
3835 tested the remembered result. Now, if a previous verification attempt has
3836 not happened, "verify = helo" does it there and then.
3838 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3840 TK/03 Fix log output including CR from clamd.
3842 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3843 exist provoked a memory error which could cause a segfault.
3845 PH/15 Installed PCRE 6.2
3847 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3849 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3850 of the problem. Specifically, suggested +O2 rather than +O1 for the
3853 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3855 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3856 record, the port was not correctly logged when the outgoing_port log
3857 selector was set (it logged the transort's default port).
3859 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3860 fallback_hosts, and "hosts" in the smtp transport.
3862 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3863 host errors such as "Connection refused".
3865 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3866 authentication with radiusclient 0.4.9:
3868 - Error returned from rc_read_config was caught wrongly
3869 - Username/password not passed on to radius server due to wrong length.
3871 The presumption is that some radiusclient API changes for 4.51/PH/17
3872 were not taken care of correctly. The code is still untested by me (my
3873 Linux distribution still has 0.3.2 of radiusclient), but it was
3874 contributed by a Radius user.
3876 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3877 expanding the "port" option of the smtp transport.
3879 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3880 while reading a MIME header. Thanks to Tom Hughes for a patch.
3882 PH/24 Include config.h inside local_scan.h so that configuration settings are
3885 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3886 that in an ACL for RCPT (for example), you can examine exactly what was
3889 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3890 commands, but it was not correctly comparing the address with the actual
3891 client host address. Thus, it would show the EHLO address in Received:
3892 header lines when this was not necessary.
3894 PH/27 Added the % operator to ${eval:}.
3896 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3897 it should be ignoring failures (because with -C, for example, it has lost
3898 privilege). It wasn't ignoring creation failures other than "already
3901 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3904 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3905 because an input line was too long, either on its own, or by virtue of
3906 too many continuations, the temporary file was not being removed, and the
3907 return code was incorrect.
3909 PH/31 Missing "BOOL" in function definition in filtertest.c.
3911 PH/32 Applied Sieve patches from the maintainer.
3913 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3915 PH/33 Added "verify = not_blind".
3917 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3918 Local/Makefile (with some defaults set). These are used in built scripts
3919 such as exicyclog, but they have never been used in the exim_install
3920 script (though there are many overriding facilities there). I have
3921 arranged that the exim_install script now takes note of these two
3924 PH/35 Installed configuration files for Dragonfly.
3926 PH/36 When a locally submitted message by a trusted user did not contain a
3927 From: header, and the sender address was obtained from -f or from an SMTP
3928 MAIL command, and the trusted user did not use -F to supply a sender
3929 name, $originator_name was incorrectly used when constructing a From:
3930 header. Furthermore, $originator_name was used for submission mode
3931 messages from external hosts without From: headers in a similar way,
3932 which is clearly wrong.
3934 PH/37 Added control=suppress_local_fixups.
3936 PH/38 When log_selector = +received_sender was set, and the addition of the
3937 sender made the log line's construction buffer exactly full, or one byte
3938 less than full, an overflow happened when the terminating "\n" was
3941 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3942 when the result of a list match is failure because a DNS lookup failed.
3944 PH/40 RM_COMMAND is now used in the building process.
3946 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3947 the "build-* directories that it finds.
3949 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3950 address in a domain literal was a prefix of an interface address.
3952 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3953 when verifying a sender address, unless rewrite_headers is false.
3955 PH/44 Wrote a long comment about why errors_to addresses are verified as
3956 recipients, not senders.
3958 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3959 the ratelimit ACL was added.
3961 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3963 PH/46 Added extra information about PostgreSQL errors to the error string.
3965 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3966 avoiding the need to install special zones in a real server. This is
3967 backwards compatible; if it can't find the fake resolver, it drops back.
3968 Thus, both old and new test suites can be run.
3970 TF/02 Added util/ratelimit.pl
3972 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3973 clock is set back in time.
3975 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3976 Candler <B.Candler@pobox.com>.
3978 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3979 OK for addresses that are the result of redirections.
3981 PH/48 A number of further additions for the benefit of the new test suite,
3982 including a fake gethostbyname() that interfaces to the fake DNS resolver
3985 TF/06 The fix for widen_domains has also been applied to qualify_single and
3986 search_parents which are the other dnslookup options that can cause
3989 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3992 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3994 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3995 service" if it encountered an MX record with an empty target hostname.
3996 The message is now "an MX or SRV record indicated no SMTP service".
3998 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3999 if the DNS of the sending site is misconfigured. This is quite a
4000 common situation. This change restores the behaviour of treating a
4001 helo verification defer as a failure.
4003 PH/51 If self=fail was set on a router, the bounce message did not include the
4004 actual error message.
4010 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
4012 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
4013 command itself ended in error, the underlying message about the transport
4014 filter timeout was being overwritten with the pipe command error. Now the
4015 underlying error message should be appended to the second error message.
4017 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4019 PH/02 Reduce the amount of output that "make" produces by default. Full output
4020 can still be requested.
4022 PH/03 The warning log line about a condition test deferring for a "warn" verb
4023 was being output only once per connection, rather than after each
4024 occurrence (because it was using the same function as for successful
4025 "warn" verbs). This seems wrong, so I have changed it.
4027 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4028 it should not have, which might have caused a crash in the right
4029 circumstances, but probably never did.
4031 PH/04 Installed a modified version of Tony Finch's patch to make submission
4032 mode fix the return path as well as the Sender: header line, and to
4033 add a /name= option so that you can make the user's friendly name appear
4036 TF/03 Added the control = fakedefer ACL modifier.
4038 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4039 Mark Lowes for thorough testing.
4041 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4043 TK/03 Merged latest SRS patch from Miles Wilton.
4045 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4046 with the definition in sysexits.h (which is #included earlier).
4047 Fortunately, Exim does not actually use EX_OK. The code used to try to
4048 preserve the sysexits.h value, by assuming that macro definitions were
4049 scanned for macro replacements. I have been disabused of this notion,
4050 so now the code just undefines EX_OK before #including unistd.h.
4052 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4053 in the smtp transport. When a block could not be written in a single
4054 write() function, the timeout was being re-applied to each part-write.
4055 This seems wrong - if the receiver was accepting one byte at a time it
4056 would take for ever. The timeout is now adjusted when this happens. It
4057 doesn't have to be particularly precise.
4059 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4060 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4062 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4063 without a domain if the check to <postmaster@domain> fails.
4065 SC/01 Eximstats: added -xls and the ability to specify output files
4066 (patch written by Frank Heydlauf).
4068 SC/02 Eximstats: use FileHandles for outputting results.
4070 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4072 SC/04 Eximstats: fixed display of large numbers with -nvr option
4074 SC/05 Eximstats: fixed merging of reports with empty tables.
4076 SC/06 Eximstats: added the -include_original_destination flag
4078 SC/07 Eximstats: removed tabs and trailing whitespace.
4080 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4082 TK/06 MBOX spool code: Add real "From " MBOX separator line
4083 so the .eml file is really in mbox format (even though
4084 most programs do not really care). Patch from Alex Miller.
4086 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4087 The latter is generated from $received_to and is only set if the
4088 message has one envelope recipient. SA can use these headers,
4089 obviously out-of-the-box. Patch from Alex Miller.
4091 PH/08 The ${def test on a variable was returning false if the variable's
4092 value was "0", contrary to what the specification has always said!
4093 The result should be true unless the variable is empty.
4095 PH/09 The syntax error of a character other than { following "${if
4096 def:variable_name" (after optional whitespace) was not being diagnosed.
4097 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4098 accidental colon was present, for example, could give incorrect results.
4100 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4101 result is used (not including appendfile, where other changes are about
4104 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4105 This involved changing a lot of size variables from int to off_t. It
4106 should work with maildirs and everything.
4108 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4109 spamd dying while we are connected to it.
4111 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4114 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4115 function for BDB 4.3.
4117 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4119 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4120 the sources for PCRE have been re-arranged and more files are now
4123 PH/15 The code I had for printing potentially long long variables in PH/11
4124 above was not the best (it lost precision). The length of off_t variables
4125 is now inspected at build time, and an appropriate printing format (%ld
4126 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4127 to be "long long int" or "long int". This is needed for the internal
4128 formatting function string_vformat().
4130 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4131 the configuration file to be ":syslog", then the script "guesses" where
4132 the logs files are, rather than using the compiled in default. In our
4133 case the guess is not the same as the compiled default, so the script
4134 suddenly stopped working when I started to use syslog. The patch checks
4135 to see if log_file_path is "". If so, it attempts to read it from exim
4136 with no configuration file to get the compiled in version, before it
4137 falls back to the previous guessing code."
4139 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4140 implementing BATV in an Exim configuration. See NewStuff for the gory
4143 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4144 Makefile that are specific to HP-UX.
4146 PH/18 If the "use_postmaster" option was set for a recipient callout together
4147 with the "random" option, the postmaster address was used as the MAIL
4148 FROM address for the random test, but not for the subsequent recipient
4149 test. It is now used for both.
4151 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4152 patch removes a few documentation additions to RFC 3028, because the
4153 latest draft now contains them. It adds the new en;ascii-case comparator
4154 and a new error check for 8bit text in MIME parts. Comparator and
4155 require names are now matched exactly. I enabled the subaddress
4156 extension, but it is not well tested yet (read: it works for me)."
4158 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4159 rework some of the code of TK/09 above to avoid the hardwired use of
4160 "%lld" and "long long". Replaced the call to snprintf() with a call to
4163 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4164 records point to non-existent hosts", "retry timeout exceeded", and
4165 "retry time not reached for any host after a long failure period".
4167 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4168 experimental DomainKeys support:
4170 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4171 (2) On an error such as an illegally used "control", the wrong name for
4172 the control was given.
4174 These problems did NOT occur unless DomainKeys support was compiled.
4176 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4178 PH/24 Added ${if match_ip condition.
4180 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4181 databases so that it will be absolutely obvious if a crash occurs in the
4182 DB library. This is a regular occurrence (often caused by mis-matched
4185 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4186 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4187 picked up on a user's system that detects such things. There doesn't seem
4188 to be a gcc warning option for this - only an attribute that has to be
4189 put on the function's prototype. It seems that in Fedora Core 4 they have
4190 set this on a number of new functions. No doubt there will be more in due
4193 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4194 specify a transport. However, if an address that was verified by such a
4195 router was the subject of a callout, Exim crashed because it tried to
4196 read the rcpt_include_affixes from the non-existent transport. Now it
4197 just assumes that the setting of that option is false. This bug was
4198 introduced by 4.51/PH/31.
4200 PH/28 Changed -d+all to exclude +memory, because that information is very
4201 rarely of interest, but it makes the output a lot bigger. People tend to
4202 do -d+all out of habit.
4204 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4205 code in os-type was giving problems when libc.so lives in lib64, like on
4208 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4209 aren't the modern standard, and it seems that some systems' include files
4210 don't always have them. Exim was already checking for some of the newer
4211 ones like T_AAAA, and defining it itself. I've added checks for all the
4212 record types that Exim uses.
4214 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4215 not automatically generating a new one, as it is supposed to. This
4216 prevented TLS from working. If the file did exist, but contained invalid
4217 data, a new version was generated, as expected. It was only the case of a
4218 non-existent file that was broken.
4220 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4221 with a change in libdomainkeys > 0.64.
4223 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4224 from DNS. If the selector record carries the flag, it now has
4225 precedence over the domain-wide flag.
4227 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4229 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4230 the use of a port name that isn't defined in /etc/services) occurred, the
4231 message was deferred as in a normal delivery, and thus remained on the
4232 spool, instead of being failed because of the mua_wrapper setting. This
4233 is now fixed, and I tidied up some of the mua_wrapper messages at the
4236 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4237 the messages in a hash of arrays rather than using individual hashes.
4238 This is a bit cleaner and results in dramatic memory savings, albeit
4239 at a slight CPU cost.
4241 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4242 as requested by Marc Sherman.
4244 SC/10 Eximstats: added histograms for user specified patterns as requested
4247 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4249 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4250 fopen() in the content-scanning modules that did not already have it.
4256 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4257 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4259 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4261 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4263 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4264 to test Sieve filters that use "vacation".
4266 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4267 that changes the way the GnuTLS parameters are stored in the cache file.
4268 The new format can be generated externally. For backward compatibility,
4269 if the data in the cache doesn't make sense, Exim assumes it has read an
4270 old-format file, and it generates new data and writes a new file. This
4271 means that you can't go back to an older release without removing the
4274 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4275 work if there are any delivery delays because "one_time" forces the
4276 parent to be marked "delivered", so its unseen clone is never tried
4277 again. For this reason, Exim now forbids the simultaneous setting of
4280 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4281 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4282 entry if you want to know the details. The fix, however, neglected to
4283 consider the case when local delivery batching is involved. The test for
4284 "previously delivered" was not happening when checking to see if an
4285 address could be batched with a previous (undelivered) one; under
4286 certain circumstances this could lead to multiple deliveries to the same
4289 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4290 in its include files, and this causes problems building Exim.
4292 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4293 header_syntax) but Exim was just ignoring anything given after a slash.
4294 In particular, this caused confusion with an attempt to use "verify =
4295 reverse_host_lookup/defer_ok". An error is now given when options are
4296 supplied for verify items that do not have them. (Maybe reverse_host_
4297 lookup should have a defer_ok option, but that's a different point.)
4299 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4300 defined by RFC 821) to 2048, because there were problems with some AUTH
4301 commands, and RFC 1869 says the size should be increased for extended
4302 SMTP commands that take arguments.
4304 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4307 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4308 "unknown" error; now it says that the functionality isn't in the binary.
4310 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4311 an address' error message when a string expansion fails (syntax or
4312 whatever). Otherwise the password may appear in the log. Following change
4313 PH/42 below, there is no longer a chance of it appearing in a bounce
4316 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4318 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4319 of its addresses was ever tried. (Bugzilla bug #2.)
4321 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4322 the result incorrectly in the debug output. (It correctly added a newline
4323 to what was transported.)
4325 TF/01 Added $received_time.
4327 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4328 commented out examples of how to interface to a virus scanner and to
4329 SpamAssassin. Also added commented examples of av_scanner and
4330 spamd_address settings.
4332 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4333 and controls are allowed in which ACLs. There were a couple of minor
4334 errors. Some of the entries in the conditions table (which is a table of
4335 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4336 negation of where the condition IS allowed.
4338 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4340 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4341 header file does not have a version number, so I've had to invent a new
4342 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4343 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4344 radiusclient), but it was contributed by a Radius user.
4346 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4347 files or directories, for OpenSSL.
4349 PH/19 When an Exim process that is running as root has to create an Exim log
4350 file, it does so in a subprocess that runs as exim:exim so as to get the
4351 ownership right at creation (otherwise, other Exim processes might see
4352 the file with the wrong ownership). There was no test for failure of this
4353 fork() call, which would lead to the process getting stuck as it waited
4354 for a non-existent subprocess. Forks do occasionally fail when resources
4355 run out. I reviewed all the other calls to fork(); they all seem to check
4358 PH/20 When checking for unexpected SMTP input at connect time (before writing
4359 the banner), Exim was not dealing correctly with a non-positive return
4360 from the read() function. If the client had disconnected by this time,
4361 the result was a log entry for a synchronization error with an empty
4362 string after "input=" when read() returned zero. If read() returned -1
4363 (an event I could not check), uninitialized data bytes were printed.
4364 There were reports of junk text (parts of files, etc) appearing after
4367 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4369 PH/22 Added support for macro redefinition, and (re)definition in between
4370 driver and ACL definitions.
4372 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4373 forgetting to use the resulting value; it was using the unexpanded value.
4375 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4376 hadn't been configured. The fix is from Juergen Kreileder, who
4377 understands it better than I do:
4379 "Here's what I see happening with three configured cyrus_sasl
4380 authenticators configured (plain, login, cram-md5):
4382 On startup auth_cyrus_sasl_init() gets called for each of these.
4383 This means three calls to sasl_listmech() without a specified mech_list.
4384 => SASL tests which mechs of all available mechs actually work
4385 => three warnings about OTP not working
4386 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4388 With the patch, sasl_listmech() also gets called three times. But now
4389 SASL's mech_list option is set to the server_mech specified in the the
4390 authenticator. Or in other words, the answer from sasl_listmech()
4391 gets limited to just the mech you're testing for (which is different
4393 => the return list contains just 'plain' or 'login', 'cram-md5' or
4394 nothing depending on the value of ob->server_mech.
4396 I've just tested the patch: Authentication still works fine,
4397 unavailable mechs specified in the exim configuration are still
4398 caught, and the auth.log warnings about OTP are gone."
4400 PH/25 When debugging is enabled, the contents of the command line are added
4401 to the debugging output, even when log_selector=+arguments is not
4404 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4405 answer is "GNU", and only if the return is "GNU/something" is the answer
4408 PH/27 $acl_verify_message is now set immediately after the failure of a
4409 verification in an ACL, and so is available in subsequent modifiers. In
4410 particular, the message can be preserved by coding like this:
4412 warn !verify = sender
4413 set acl_m0 = $acl_verify_message
4415 Previously, $acl_verify_message was set only while expanding "message"
4416 and "log_message" when a very denied access.
4418 PH/28 Modified OS/os.c-Linux with
4420 -#ifndef OS_LOAD_AVERAGE
4421 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4423 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4424 nomenclature these days.)
4426 PH/29 Installed patch from the Sieve maintainer that adds the options
4427 sieve_useraddress and sieve_subaddress to the redirect router.
4429 PH/30 In these circumstances:
4430 . Two addresses routed to the same list of hosts;
4431 . First host does not offer TLS;
4432 . First host accepts first address;
4433 . First host gives temporary error to second address;
4434 . Second host offers TLS and a TLS session is established;
4435 . Second host accepts second address.
4436 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4437 and peerdn, if requested) that were in fact used only for the second
4440 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4441 attention to any local part prefix or suffix that was matched by the
4442 router that accepted the address. It now behaves in the same way as it
4443 does for delivery: the affixes are removed from the local part unless
4444 rcpt_include_affixes is set on the transport.
4446 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4447 timeout during the DATA phase of an incoming message.
4449 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4450 applied a patch sanctioned by the Sieve maintainer.
4452 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4453 the uid or gid is negative. A case of a negative gid caused this to be
4454 noticed. The fix allows for either to be negative.
4456 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4457 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4460 PH/36 The change PH/12 above was broken. Fixed it.
4462 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4463 the grounds that routing the same address twice would always produce the
4464 same answer. This might have been true once, but it is certainly no
4465 longer true now. Routing a child address may depend on the previous
4466 routing that produced that child. Some complicated redirection strategies
4467 went wrong when messages had multiple recipients, and made Exim's
4468 behaviour dependent on the order in which the addresses were given.
4470 I have moved the duplicate checking until after the routing is complete.
4471 Exim scans the addresses that are assigned to local and remote
4472 transports, and removes any duplicates. This means that more work will be
4473 done, as duplicates will always all be routed, but duplicates are
4474 presumably rare, so I don't expect this is of any significance.
4476 For deliveries to pipes, files, and autoreplies, the duplicate checking
4477 still happens during the routing process, since they are not going to be
4480 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4481 It corrects a timeout issue with spamd. This is Ian's comment: "The
4482 background is that sometimes spamd either never reads data from a
4483 connection it has accepted, or it never writes response data. The exiscan
4484 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4485 blindly assumes that writes won't block so it may never time out."
4487 PH/39 Allow G after quota size as well as K and M.
4489 PH/40 The value set for $authenticated_id in an authenticator may not contain
4490 binary zeroes or newlines because the value is written to log lines and
4491 to spool files. There was no check on this. Now the value is run through
4492 the string_printing() function so that such characters are converted to
4493 printable escape sequences.
4495 PH/41 $message_linecount is a new variable that contains the total number of
4496 lines in the message. Compare $body_linecount, which is the count for the
4499 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4500 bounce and delay warning messages, except in certain special cases, which
4503 (a) An SMTP error message from a remote host;
4504 (b) A message specified in a :fail: redirection;
4505 (c) A message specified in a "fail" command in a system filter;
4506 (d) A message specified in a FAIL return from the queryprogram router;
4507 (e) A message specified by the cannot_route_message router option.
4509 In these cases only, Exim does include the error details in bounce and
4510 warning messages. There are also a few cases where bland messages such
4511 as "unrouteable address" or "local delivery error" are given.
4513 PH/43 $value is now also set for the "else" part of a ${run expansion.
4515 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4516 being worked on, but at least Exim now implements the latest version to
4519 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4520 process to complete was treated as a delivery failure, a timeout while
4521 writing the message to the pipe was logged, but erroneously treated as a
4522 successful delivery. Such timeouts include transport filter timeouts. For
4523 consistency with the overall process timeout, these timeouts are now
4524 treated as errors, giving rise to delivery failures by default. However,
4525 there is now a new Boolean option for the pipe transport called
4526 timeout_defer, which, if set TRUE, converts the failures into defers for
4527 both kinds of timeout. A transport filter timeout is now identified in
4530 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4531 systems where "make" and "gmake" are different, calling "gmake" at top
4532 level broke things. I've arranged for the value of $(MAKE) to be passed
4533 from the Makefile to this script so that it can call the same version of
4537 A note about Exim versions 4.44 and 4.50
4538 ----------------------------------------
4540 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4541 changes of various kinds. As a consequence, a big documentation update was
4542 needed. This delayed the release for rather longer than seemed good, especially
4543 in the light of a couple of (minor) security issues. Therefore, the changes
4544 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4545 release. So 4.44 and 4.50 are in effect two different branches that both start
4548 I have left the 4.50 change log unchanged; it contains all the changes since
4549 4.43. The change log for 4.44 is below; many of its items are identical to
4550 those for 4.50. This seems to be the most sensible way to preserve the
4551 historical information.
4557 1. Minor wording change to the doc/README.SIEVE file.
4559 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4560 computation of the current number of files was incorrect.
4562 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4563 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4566 4. Give more explanation in the error message when the command for a transport
4567 filter fails to execute.
4569 5. There are several places where Exim runs a non-Exim command in a
4570 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4571 was being done only for the command run by the queryprogram router. It is
4572 now done for all such subprocesses. The other cases are: ${run, transport
4573 filters, and the commands run by the lmtp and pipe transports.
4575 6. Added CONFIGURE_GROUP build-time option.
4577 7. Some older OS have a limit of 256 on the maximum number of file
4578 descriptors. Exim was using setrlimit() to set 1000 as a large value
4579 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4580 systems. I've change it so that if it can't get 1000, it tries for 256.
4582 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4583 was an oversight, and furthermore, ever since the addition of extra
4584 controls (e.g. 4.43/32), the checks on when to allow different forms of
4585 "control" were broken. There should now be diagnostics for all cases when a
4586 control that does not make sense is encountered.
4588 9. Added the /retain_sender option to "control=submission".
4590 10. $recipients is now available in the predata ACL (oversight).
4592 11. Tidy the search cache before the fork to do a delivery from a message
4593 received from the command line. Otherwise the child will trigger a lookup
4594 failure and thereby defer the delivery if it tries to use (for example) a
4595 cached ldap connection that the parent has called unbind on.
4597 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4598 of $address_data from the recipient verification was clobbered by the
4599 sender verification.
4601 13. The value of address_data from a sender verification is now available in
4602 $sender_address_data in subsequent conditions in the ACL statement.
4604 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4606 15. Added a new option "connect=<time>" to callout options, to set a different
4609 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4610 was its contents. (It was OK if the option was not defined at all.)
4612 17. A "Completed" log line is now written for messages that are removed from
4613 the spool by the -Mrm option.
4615 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4616 information about exactly what failed.
4618 19. Added -dd to debug only the daemon process.
4620 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4621 handles timeouts, both on the server side and network timeouts. Renamed the
4622 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4624 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4625 It is now set to "smtps".
4627 22. $host_address is now set to the target address during the checking of
4628 ignore_target_hosts.
4630 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4631 being passed; this would have caused $sender_host_name to have been used if
4632 matching the list had actually called for a host name (not very likely,
4633 since this list is usually IP addresses). A host name is now passed as
4636 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4637 code that specifies a non-restarting handler (typically sigaction() in
4638 modern systems) in an attempt to fix a rare and obscure crash bug.
4640 25. Narrowed the window for a race in the daemon that could cause it to ignore
4641 SIGCHLD signals. This is not a major problem, because they are used only to
4642 wake it up if nothing else does.
4644 26. A malformed maildirsize file could cause Exim to calculate negative values
4645 for the mailbox size or file count. Odd effects could occur as a result.
4646 The maildirsize information is now recalculated if the size or filecount
4649 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4650 support for a long time. Removed HAVE_SYS_VFS_H.
4652 28. Installed the latest version of exipick from John Jetmore.
4654 29. In an address list, if the pattern was not a regular expression, an empty
4655 subject address (from a bounce message) matched only if the pattern was an
4656 empty string. Non-empty patterns were not even tested. This was the wrong
4657 because it is perfectly reasonable to use an empty address as part of a
4658 database query. An empty address is now tested by patterns that are
4659 lookups. However, all the other forms of pattern expect the subject to
4660 contain a local part and a domain, and therefore, for them, an empty
4661 address still always fails if the pattern is not itself empty.
4663 30. Exim went into a mad DNS loop when attempting to do a callout where the
4664 host was specified on an smtp transport, and looking it up yielded more
4665 than one IP address.
4667 31. Re-factored the code for checking spool and log partition space into a
4668 function that finds that data and another that does the check. The former
4669 is then used to implement four new variables: $spool_space, $log_space,
4670 $spool_inodes, and $log_inodes.
4672 32. The RFC2047 encoding function was originally intended for short strings
4673 such as real names; it was not keeping to the 75-character limit for
4674 encoded words that the RFC imposes. It now respects the limit, and
4675 generates multiple encoded words if necessary. To be on the safe side, I
4676 have increased the buffer size for the ${rfc2047: expansion operator from
4679 33. It is now permitted to omit both strings after an "if" condition; if the
4680 condition is true, the result is "true". As before, when the second string
4681 is omitted, a false condition yields an empty string. This makes it less
4682 cumbersome to write custom ACL and router conditions.
4684 34. Failure to deliver a bounce message always caused it to be frozen, even if
4685 there was an errors_to setting on the router. The errors_to setting is now
4688 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4689 canonical form (fully expanded) before being placed in
4690 $sender_host_address.
4692 36. The table in the code that translates DNS record types into text (T_A to
4693 "A" for instance) was missing entries for NS and CNAME. It is just possible
4694 that this could have caused confusion if both these types were looked up
4695 for the same domain, because the text type is used as part of Exim's
4696 per-process caching. But the chance of anyone hitting this buglet seems
4699 37. The dnsdb lookup has been extended in a number of ways.
4701 (1) There is a new type, "zns", which walks up the domain tree until it
4702 finds some nameserver records. It should be used with care.
4704 (2) There is a new type, "mxh", which is like "mx" except that it returns
4705 just the host names, not the priorities.
4707 (3) It is now possible to give a list of domains (or IP addresses) to be
4708 looked up. The behaviour when one of the lookups defers can be
4709 controlled by a keyword.
4711 (4) It is now possible to specify the separator character for use when
4712 multiple records are returned.
4714 38. The dnslists ACL condition has been extended: it is now possible to supply
4715 a list of IP addresses and/or domains to be looked up in a particular DNS
4718 39. Added log_selector=+queue_time_overall.
4720 40. When running the queue in the test harness, wait just a tad after forking a
4721 delivery process, to get repeatability of debugging output.
4723 41. Include certificate and key file names in error message when GnuTLS fails
4724 to set them up, because the GnuTLS error message doesn't include the name
4725 of the failing file when there is a problem reading it.
4727 42. Allow both -bf and -bF in the same test run.
4729 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4731 44. The "Exiscan patch" is now merged into the mainline Exim source.
4733 45. Sometimes the final signoff response after QUIT could fail to get
4734 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4735 < 0 before doing a fflush(). This bug looks as though it goes back to the
4736 introduction of TLS in release 3.20, but "sometimes" must have been rare
4737 because the tests only now provoked it.
4739 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4740 (this can affect the format of dates).
4742 47. exim_tidydb, when checking for the continued existence of a message for
4743 which it has found a message-specific retry record, was not finding
4744 messages that were in split spool directories. Consequently, it was
4745 deleting retry records that should have stayed in existence.
4747 48. Steve fixed some bugs in eximstats.
4749 49. The SPA authentication driver was not abandoning authentication and moving
4750 on to the next authenticator when an expansion was forced to fail,
4751 contradicting the general specification for all authenticators. Instead it
4752 was generating a temporary error. It now behaves as specified.
4754 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4755 (the order specifies the preference for clients). The order is now AES256,
4756 AES128, 3DES, ARCFOUR128.
4758 51. Small patch to Sieve code - explicitly set From: when generating an
4761 52. Exim crashed if a remote delivery caused a very long error message to be
4762 recorded - for instance if somebody sent an entire SpamAssassin report back
4763 as a large number of 550 error lines. This bug was coincidentally fixed by
4764 increasing the size of one of Exim's internal buffers (big_buffer) that
4765 happened as part of the Exiscan merge. However, to be on the safe side, I
4766 have made the code more robust (and fixed the comments that describe what
4769 53. Now that there can be additional text after "Completed" in log lines (if
4770 the queue_time_overall log selector is set), a one-byte patch to exigrep
4771 was needed to allow it to recognize "Completed" as not the last thing in
4774 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4775 patch that reportedly fixes this has been added. I am not expert enough to
4776 create a test for it. This is what the patch creator wrote:
4778 "I found a little strange behaviour of ldap code when working with
4779 Windows 2003 AD Domain, where users was placed in more than one
4780 Organization Units. When I tried to give exim partial DN, the exit code
4781 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4782 But simultaneously result of request was absolutely normal ldap result,
4783 so I produce this patch..."
4785 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4786 REFERENCE, so I have modified the code to exclude the patch when that macro
4789 55. Some experimental protocols are using DNS PTR records for new purposes. The
4790 keys for these records are domain names, not reversed IP addresses. The
4791 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4792 leaves it alone. Component reversal etc. now happens only for IP addresses.
4795 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4797 57. Double the size of the debug message buffer (to 2048) so that more of very
4798 long debug lines gets shown.
4800 58. The exicyclog utility now does better if the number of log files to keep
4801 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4803 59. Two changes related to the smtp_active_hostname option:
4805 (1) $smtp_active_hostname is now available as a variable.
4806 (2) The default for smtp_banner uses $smtp_active_hostname instead
4807 of $primary_hostname.
4809 60. The host_aton() function is supposed to be passed a string that is known
4810 to be a valid IP address. However, in the case of IPv6 addresses, it was
4811 not checking this. This is a hostage to fortune. Exim now panics and dies
4812 if the condition is not met. A case was found where this could be provoked
4813 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4814 components; fortuitously, this particular loophole had already been fixed
4815 by change 4.50/55 above.
4817 If there are any other similar loopholes, the new check in host_aton()
4818 itself should stop them being exploited. The report I received stated that
4819 data on the command line could provoke the exploit when Exim was running as
4820 exim, but did not say which command line option was involved. All I could
4821 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4822 running as the user.
4825 61. There was a buffer overflow vulnerability in the SPA authentication code
4826 (which came originally from the Samba project). I have added a test to the
4827 spa_base64_to_bits() function which I hope fixes it.
4830 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4831 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4833 63. The daemon start-up calls getloadavg() while still root for those OS that
4834 need the first call to be done as root, but it missed one case: when
4835 deliver_queue_load_max is set with deliver_drop_privilege. This is
4836 necessary for the benefit of the queue runner, because there is no re-exec
4837 when deliver_drop_privilege is set.
4839 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4840 This has been fixed.
4842 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4843 was in use, was not putting the data itself into the right store pool;
4844 consequently, it could be overwritten for a subsequent message in the same
4845 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4848 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4850 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4851 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4852 the function were treating the return as a boolean value, which happened to
4853 work because 0=false and not-0=true, but is not correct code.
4855 68. The host_aton() function was not handling scoped IPv6 addresses (those
4856 with, for example, "%eth0" on the end) correctly.
4858 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4859 negated items (that is, ~something) in unsigned ints. Some compilers
4860 apparently mutter when there is no cast.
4862 70. If an address verification called from an ACL failed, and did not produce a
4863 user-specific message (i.e. there was only a "system" message), nothing was
4864 put in $acl_verify_message. In this situation, it now puts the system
4867 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4868 session; change 4.31/43 added the unwanted input to the log line - except
4869 that it did not do this in the start of session case. It now does.
4871 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4872 This is wrong and can cause the other end to generate a synchronization
4873 error if it is another Exim or anything else that does the synchronization
4874 check. A QUIT command is no longer sent after a timeout.
4876 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4877 during host lookups.
4879 74. The defer_ok option of callout verification was not working if it was used
4880 when verifying addresses in header lines, that is, for this case:
4882 verify = header_sender/callout=defer_ok
4884 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4885 those file descriptors could be used for SMTP connections. If anything
4886 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4887 could be sent to the SMTP client, causing chaos. The daemon now opens
4888 stdin, stdout, and stderr to /dev/null when it puts itself into the
4891 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4892 log by default. The user can override this with suitable Perl magic.
4894 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4895 the log message when discard triggers, was not working for the DATA ACL or
4896 for the non-SMTP ACL.
4898 78. Error message wording change in sieve.c.
4900 79. If smtp_accept_max_per_host was set, the number of connections could be
4901 restricted to fewer than expected, because the daemon was trying to set up
4902 a new connection before checking whether the processes handling previous
4903 connections had finished. The check for completed processes is now done
4904 earlier. On busy systems, this bug wouldn't be noticed because something
4905 else would have woken the daemon, and it would have reaped the completed
4908 80. If a message was submitted locally by a user whose login name contained one
4909 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4910 It caused a spool format error. I have fixed the spool reading code. A
4911 related problem was that the "from" clause in the Received: line became
4912 illegal because of the space(s). It is now covered by ${quote_local_part.
4914 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4917 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4919 83. Patch from Sieve maintainer to fix unterminated string problem in
4920 "vacation" handling.
4922 84. Some minor changes to the Linux configuration files to help with other
4923 OS variants using glibc.
4925 85. One more patch for Sieve to update vacation handling to latest spec.
4928 ----------------------------------------------------
4929 See the note above about the 4.44 and 4.50 releases.
4930 ----------------------------------------------------
4936 1. Change 4.43/35 introduced a bug that caused file counts to be
4937 incorrectly computed when quota_filecount was set in an appendfile
4940 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4941 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4944 3. Give more explanation in the error message when the command for a transport
4945 filter fails to execute.
4947 4. There are several places where Exim runs a non-Exim command in a
4948 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4949 was being done only for the command run by the queryprogram router. It is
4950 now done for all such subprocesses. The other cases are: ${run, transport
4951 filters, and the commands run by the lmtp and pipe transports.
4953 5. Some older OS have a limit of 256 on the maximum number of file
4954 descriptors. Exim was using setrlimit() to set 1000 as a large value
4955 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4956 systems. I've change it so that if it can't get 1000, it tries for 256.
4958 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4959 was an oversight, and furthermore, ever since the addition of extra
4960 controls (e.g. 4.43/32), the checks on when to allow different forms of
4961 "control" were broken. There should now be diagnostics for all cases when a
4962 control that does not make sense is encountered.
4964 7. $recipients is now available in the predata ACL (oversight).
4966 8. Tidy the search cache before the fork to do a delivery from a message
4967 received from the command line. Otherwise the child will trigger a lookup
4968 failure and thereby defer the delivery if it tries to use (for example) a
4969 cached ldap connection that the parent has called unbind on.
4971 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4972 of $address_data from the recipient verification was clobbered by the
4973 sender verification.
4975 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4976 was its contents. (It was OK if the option was not defined at all.)
4978 11. A "Completed" log line is now written for messages that are removed from
4979 the spool by the -Mrm option.
4981 12. $host_address is now set to the target address during the checking of
4982 ignore_target_hosts.
4984 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4985 being passed; this would have caused $sender_host_name to have been used if
4986 matching the list had actually called for a host name (not very likely,
4987 since this list is usually IP addresses). A host name is now passed as
4990 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4991 code that specifies a non-restarting handler (typically sigaction() in
4992 modern systems) in an attempt to fix a rare and obscure crash bug.
4994 15. Narrowed the window for a race in the daemon that could cause it to ignore
4995 SIGCHLD signals. This is not a major problem, because they are used only to
4996 wake it up if nothing else does.
4998 16. A malformed maildirsize file could cause Exim to calculate negative values
4999 for the mailbox size or file count. Odd effects could occur as a result.
5000 The maildirsize information is now recalculated if the size or filecount
5003 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5004 support for a long time. Removed HAVE_SYS_VFS_H.
5006 18. Updated exipick to current release from John Jetmore.
5008 19. Allow an empty sender to be matched against a lookup in an address list.
5009 Previously the only cases considered were a regular expression, or an
5012 20. Exim went into a mad DNS lookup loop when doing a callout where the
5013 host was specified on the transport, if the DNS lookup yielded more than
5016 21. The RFC2047 encoding function was originally intended for short strings
5017 such as real names; it was not keeping to the 75-character limit for
5018 encoded words that the RFC imposes. It now respects the limit, and
5019 generates multiple encoded words if necessary. To be on the safe side, I
5020 have increased the buffer size for the ${rfc2047: expansion operator from
5023 22. Failure to deliver a bounce message always caused it to be frozen, even if
5024 there was an errors_to setting on the router. The errors_to setting is now
5027 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5028 canonical form (fully expanded) before being placed in
5029 $sender_host_address.
5031 24. Updated eximstats to version 1.33
5033 25. Include certificate and key file names in error message when GnuTLS fails
5034 to set them up, because the GnuTLS error message doesn't include the name
5035 of the failing file when there is a problem reading it.
5037 26. Expand error message when OpenSSL has problems setting up cert/key files.
5040 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5041 (this can affect the format of dates).
5043 28. exim_tidydb, when checking for the continued existence of a message for
5044 which it has found a message-specific retry record, was not finding
5045 messages that were in split spool directories. Consequently, it was
5046 deleting retry records that should have stayed in existence.
5048 29. eximstats updated to version 1.35
5049 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5050 1.35 - bugfix such that pie charts by volume are generated correctly
5052 30. The SPA authentication driver was not abandoning authentication and moving
5053 on to the next authenticator when an expansion was forced to fail,
5054 contradicting the general specification for all authenticators. Instead it
5055 was generating a temporary error. It now behaves as specified.
5057 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5058 (the order specifies the preference for clients). The order is now AES256,
5059 AES128, 3DES, ARCFOUR128.
5061 31. Small patch to Sieve code - explicitly set From: when generating an
5064 32. Exim crashed if a remote delivery caused a very long error message to be
5065 recorded - for instance if somebody sent an entire SpamAssassin report back
5066 as a large number of 550 error lines. This bug was coincidentally fixed by
5067 increasing the size of one of Exim's internal buffers (big_buffer) that
5068 happened as part of the Exiscan merge. However, to be on the safe side, I
5069 have made the code more robust (and fixed the comments that describe what
5072 33. Some experimental protocols are using DNS PTR records for new purposes. The
5073 keys for these records are domain names, not reversed IP addresses. The
5074 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5075 leaves it alone. Component reversal etc. now happens only for IP addresses.
5078 34. The host_aton() function is supposed to be passed a string that is known
5079 to be a valid IP address. However, in the case of IPv6 addresses, it was
5080 not checking this. This is a hostage to fortune. Exim now panics and dies
5081 if the condition is not met. A case was found where this could be provoked
5082 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5083 components; fortuitously, this particular loophole had already been fixed
5084 by change 4.50/55 or 4.44/33 above.
5086 If there are any other similar loopholes, the new check in host_aton()
5087 itself should stop them being exploited. The report I received stated that
5088 data on the command line could provoke the exploit when Exim was running as
5089 exim, but did not say which command line option was involved. All I could
5090 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5091 running as the user.
5094 35. There was a buffer overflow vulnerability in the SPA authentication code
5095 (which came originally from the Samba project). I have added a test to the
5096 spa_base64_to_bits() function which I hope fixes it.
5099 36. The daemon start-up calls getloadavg() while still root for those OS that
5100 need the first call to be done as root, but it missed one case: when
5101 deliver_queue_load_max is set with deliver_drop_privilege. This is
5102 necessary for the benefit of the queue runner, because there is no re-exec
5103 when deliver_drop_privilege is set.
5105 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5106 was in use, was not putting the data itself into the right store pool;
5107 consequently, it could be overwritten for a subsequent message in the same
5108 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5111 38. Sometimes the final signoff response after QUIT could fail to get
5112 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5113 < 0 before doing a fflush(). This bug looks as though it goes back to the
5114 introduction of TLS in release 3.20, but "sometimes" must have been rare
5115 because the tests only now provoked it.
5121 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5122 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5123 Now it returns an integer. A number of calls were still expecting a T/F
5124 return. Fortuitously, in all cases, the tests worked in OK situations,
5125 which is the norm. However, things would have gone wrong on any write
5126 failures on the smtp file descriptor. This function is used when sending
5127 messages over SMTP and also when doing verify callouts.
5129 2. When Exim is called to do synchronous delivery of a locally submitted
5130 message (the -odf or -odi options), it no longer closes stderr before doing
5133 3. Implemented the mua_wrapper option.
5135 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5137 5. Implemented the functions header_remove(), header_testname(),
5138 header_add_at_position(), and receive_remove_recipient(), and exported them
5141 6. If an ACL "warn" statement specified the addition of headers, Exim already
5142 inserted X-ACL-Warn: at the start if there was no header name. However, it
5143 was not making this test for the second and subsequent header lines if
5144 there were newlines in the string. This meant that an invalid header could
5145 be inserted if Exim was badly configured.
5147 7. Allow an ACL "warn" statement to add header lines at the start or after all
5148 the Received: headers, as well as at the end.
5150 8. Added the rcpt_4xx retry error code.
5152 9. Added postmaster_mailfrom=xxx to callout verification option.
5154 10. Added mailfrom=xxxx to the callout verification option, for verify=
5157 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5158 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5160 12. Inserted some casts to stop certain compilers warning when using pointer
5161 differences as field lengths or precisions in printf-type calls (mostly
5162 affecting debugging statements).
5164 13. Added optional readline() support for -be (dynamically loaded).
5166 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5167 same clock tick as a message's arrival, so that its received time was the
5168 same as the "first fail" time on the retry record, and that message
5169 remained on the queue past the ultimate address timeout, every queue runner
5170 would try a delivery (because it was past the ultimate address timeout) but
5171 after another failure, the ultimate address timeout, which should have then
5172 bounced the address, did not kick in. This was a "< instead of <=" error;
5173 in most cases the first failure would have been in the next clock tick
5174 after the received time, and all would be well.
5176 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5177 being recognized when the domain list was tested by the match_domain
5178 condition in an expansion string.
5180 16. Added the ${str2b64: operator.
5182 17. Exim was always calling setrlimit() to set a large limit for the number of
5183 processes, without checking whether the existing limit was already
5184 adequate. (It did check for the limit on file descriptors.) Furthermore,
5185 errors from getrlimit() and setrlimit() were being ignored. Now they are
5186 logged to the main and panic logs, but Exim does carry on, to try to do its
5187 job under whatever limits there are.
5189 18. Imported PCRE 5.0.
5191 19. Trivial typo in log message " temporarily refused connection" (the leading
5194 20. If the log selector return_path_on_delivery was set and an address was
5195 redirected to /dev/null, the delivery process crashed because it assumed
5196 that a return path would always be set for a "successful" delivery. In this
5197 case, the whole delivery is bypassed as an optimization, and therefore no
5200 21. Internal re-arrangement: the function for sending a challenge and reading
5201 a response while authentication was assuming a zero-terminated challenge
5202 string. It's now changed to take a pointer and a length, to allow for
5203 binary data in such strings.
5205 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5207 23. Exim was not respecting finduser_retries when seeking the login of the
5208 uid under which it was called; it was always trying 10 times. (The default
5209 setting of finduser_retries is zero.) Also, it was sleeping after the final
5210 failure, which is pointless.
5212 24. Implemented tls_on_connect_ports.
5214 25. Implemented acl_smtp_predata.
5216 26. If the domain in control=submission is set empty, Exim assumes that the
5217 authenticated id is a complete email address when it generates From: or
5218 Sender: header lines.
5220 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5221 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5222 chown and chgrp in /bin and hostname in /usr/bin.
5224 28. Exim was keeping the "process log" file open after each use, just as it
5225 does for the main log. This opens the possibility of it remaining open for
5226 long periods when the USR1 signal hits a daemon. Occasional processlog
5227 errors were reported, that could have been caused by this. Anyway, it seems
5228 much more sensible not to leave this file open at all, so that is what now
5231 29. The long-running daemon process does not normally write to the log once it
5232 has entered its main loop, and it closes the log before doing so. This is
5233 so that log files can straightforwardly be renamed and moved. However,
5234 there are a couple of unusual error situations where the daemon does write
5235 log entries, and I had neglected to close the log afterwards.
5237 30. The text of an SMTP error response that was received during a remote
5238 delivery was being truncated at 512 bytes. This is too short for some of
5239 the long messages that one sometimes sees. I've increased the limit to
5242 31. It is now possible to make retry rules that apply only when a message has a
5243 specific sender, in particular, an empty sender.
5245 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5246 it possible to be selective about when SMTP synchronization is enforced.
5248 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5250 32. Implemented hosts_connection_nolog.
5252 33. Added an ACL for QUIT.
5254 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5257 35. Added mailbox_size and mailbox_filecount to appendfile.
5259 36. Added control = no_multiline_responses to ACLs.
5261 37. There was a bug in the logic of the code that waits for the clock to tick
5262 in the case where the clock went backwards by a substantial amount such
5263 that the microsecond fraction of "now" was more than the microsecond
5264 fraction of "then" (but the whole seconds number was less).
5266 38. Added support for the libradius Radius client library this is found on
5267 FreeBSD (previously only the radiusclient library was supported).
5273 1. When certain lookups returned multiple values in the form name=value, the
5274 quoting of the values was not always being done properly. Specifically:
5275 (a) If the value started with a double quote, but contained no whitespace,
5277 (b) If the value contained whitespace other than a space character (i.e.
5278 tabs or newlines or carriage returns) it was not quoted.
5279 This fix has been applied to the mysql and pgsql lookups by writing a
5280 separate quoting function and calling it from the lookup code. The fix
5281 should probably also be applied to nisplus, ibase and oracle lookups, but
5282 since I cannot test any of those, I have not disturbed their existing code.
5284 2. A hit in the callout cache for a specific address caused a log line with no
5285 reason for rejecting RCPT. Now it says "Previous (cached) callout
5286 verification failure".
5288 3. There was an off-by-one bug in the queryprogram router. An over-long
5289 return line was truncated at 256 instead of 255 characters, thereby
5290 overflowing its buffer with the terminating zero. As well as fixing this, I
5291 have increased the buffer size to 1024 (and made a note to document this).
5293 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5294 when Exim is waiting for an SMTP response from a remote server, Exim
5295 restarts its select() call on the socket, thereby resetting its timeout.
5296 This is not a problem when such interrupts are rare. Somebody set up a cron
5297 job to run exiwhat every 2 minutes, which is less than the normal select()
5298 timeout (5 or 10 minutes). This meant that the select() timeout never
5299 kicked in because it was always reset. I have fixed this by comparing the
5300 time when an interrupt arrives with the time at the start of the first call
5301 to select(). If more time than the timeout has elapsed, the interrupt is
5302 treated as a timeout.
5304 5. Some internal re-factoring in preparation for the addition of Sieve
5305 extensions (by MH). In particular, the "personal" test is moved to a
5306 separate function, and given an option for scanning Cc: and Bcc: (which is
5307 not set for Exim filters).
5309 6. When Exim created an email address using the login of the caller as the
5310 local part (e.g. when creating a From: or Sender: header line), it was not
5311 quoting the local part when it contained special characters such as @.
5313 7. Installed new OpenBSD configuration files.
5315 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5316 try to make them clearer.
5318 9. Callout options, other than the timeout value, were being ignored when
5319 verifying sender addresses in header lines. For example, when using
5321 verify = header_sender/callout=no_cache
5323 the cache was (incorrectly) being used.
5325 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5326 only the Cygwin environment.
5328 11. When return_path_on_delivery was set as a log selector, if different remote
5329 addresses in the same message used different return paths and parallel
5330 remote delivery occurred, the wrong values would sometimes be logged.
5331 (Whenever a remote delivery process finished, the return path value from
5332 the most recently started remote delivery process was logged.)
5334 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5335 lines when AUTH and/or TLS are in use. This is the "received protocol"
5336 field. Exim used to use "asmtp" for authenticated SMTP, without any
5337 indication (in the protocol name) for TLS use. Now it follows the RFC and
5338 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5339 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5340 names appear in log lines as well as in Received: header lines.
5342 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5343 extensions, and comparison tests, and to fix some bugs.
5345 14. Changes to the "personal" filter test:
5347 (1) The test was buggy in that it was just doing the equivalent of
5348 "contains" tests on header lines. For example, if a user's address was
5349 anne@some.where, the "personal" test would incorrectly be true for
5351 To: susanne@some.where
5353 This test is now done by extracting each address from the header in turn,
5354 and checking the entire address. Other tests that are part of "personal"
5355 are now done using regular expressions (for example, to check local parts
5356 of addresses in From: header lines).
5358 (2) The list of non-personal local parts in From: addresses has been
5359 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5360 taken from the Sieve specification recommendations.
5362 (3) If the message contains any header line starting with "List-" it is
5363 treated as non-personal.
5365 (4) The test for "circular" in the Subject: header line has been removed
5366 because it now seems ill-conceived.
5368 15. Minor typos in src/EDITME comments corrected.
5370 16. Installed latest exipick from John Jetmore.
5372 17. If headers_add on a router specified a text string that was too long for
5373 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5374 of string_sprintf() is now avoided.
5376 18. $message_body_size was not set (it was always zero) when running the DATA
5377 ACL and the local_scan() function.
5379 19. For the "mail" command in an Exim filter, no default was being set for
5380 the once_repeat time, causing a random time value to be used if "once" was
5381 specified. (If the value happened to be <= 0, no repeat happened.) The
5382 default is now 0s, meaning "never repeat". The "vacation" command was OK
5383 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5384 (I found it when inspecting the code).
5386 20. There is now an overall timeout for performing a callout verification. It
5387 defaults to 4 times the callout timeout, which applies to individual SMTP
5388 commands during the callout. The overall timeout applies when there is more
5389 than one host that can be tried. The timeout is checked before trying the
5390 next host. This prevents very long delays if there are a large number of
5391 hosts and all are timing out (e.g. when the network connections are timing
5392 out). The value of the overall timeout can be changed by specifying an
5393 additional sub-option for "callout", called "maxwait". For example:
5395 verify = sender/callout=5s,maxwait=20s
5397 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5398 to the end before writing, but this should make it even safer).
5400 22. Exim was forgetting that it had advertised PIPELINING for the second and
5401 subsequent messages on an SMTP connection. It was also not resetting its
5402 memory on STARTTLS and an internal HELO.
5404 23. When Exim logs an SMTP synchronization error within a session, it now
5405 records whether PIPELINING has been advertised or not.
5407 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5408 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5409 rather than long int.
5411 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5413 26. Added the never_mail option to autoreply.
5419 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5420 crash if the getsockname() call failed; this can happen if a connection is
5421 closed very soon after it is established. The problem was simply in the
5422 order in which certain operations were done, causing Exim to try to write
5423 to the SMTP stream before it had set up the file descriptor. The bug has
5424 been fixed by making things happen in the correct order.
5430 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5431 before the connection was closed, thus losing the rejection response.
5433 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5434 some early Solaris releases, but causes trouble in current releases where
5435 socklen_t is defined.
5437 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5440 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5443 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5444 that releasing the top of it at the end releases what was used for sub-
5445 expansions (unless the block got too big). However, discard this block if
5446 the first thing is a variable or header, so that we can use its block when
5447 it is dynamic (useful for very large $message_headers, for example).
5449 6. Lookups now cache *every* query, not just the most recent. A new, separate
5450 store pool is used for this. It can be recovered when all lookup caches are
5451 flushed. Lookups now release memory at the end of their result strings.
5452 This has involved some general refactoring of the lookup sources.
5454 7. Some code has been added to the store_xxx() functions to reduce the amount
5455 of flapping under certain conditions.
5457 8. log_incoming_interface used to affect only the <= reception log lines. Now
5458 it causes the local interface and port to be added to several more SMTP log
5459 lines, for example "SMTP connection from", and rejection lines.
5461 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5463 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5465 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5466 could be overwritten at the end of the current message (or the start of a
5467 new message if it was set in a HELO ACL). The value is now preserved for
5468 the duration of the SMTP connection.
5470 12. If a transport had a headers_rewrite setting, and a matching header line
5471 contained an unqualified address, that address was qualified, even if it
5472 did not match any rewriting rules. The underlying bug was that the values
5473 of the flags that permit the existence of unqualified sender and recipient
5474 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5475 non-local messages, and by -bnq for local messages) were not being
5476 preserved with the message after it was received.
5478 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5479 "next input=" as part of the text comprising the host identity instead of
5480 the correct text. The code was using the same buffer for two different
5481 strings. However, depending on which order the printing function evaluated
5482 its arguments, the bug did not always show up. Under Linux, for example, my
5483 test suite worked just fine.
5485 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5486 doesn't work with some older versions of Perl. It has been changed to "my",
5487 which in any case is probably the better facility to use.
5489 15. A really picky compiler found some instances of statements for creating
5490 error messages that either had too many or two few arguments for the format
5493 16. The size of the buffer for calls to the DNS resolver has been increased
5494 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5495 for addresses that have a lot of PTR records. This alleviates a problem; it
5496 does not fully solve it.
5498 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5499 buffer now truncates the list and logs the incident, which is the same
5500 action as happens when Exim is looking up a host name and its aliases.
5501 Previously in this situation something unpredictable would happen;
5502 sometimes it was "internal error: store_reset failed".
5504 18. If a server dropped the connection unexpectedly when an Exim client was
5505 using GnuTLS and trying to read a response, the client delivery process
5506 crashed while trying to generate an error log message.
5508 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5509 string, for example:
5511 warn message = H1: something\nH2: something
5513 the text was added as a single header line from Exim's point of view
5514 though it ended up OK in the delivered message. However, searching for the
5515 second and subsequent header lines using $h_h2: did not work. This has been
5516 fixed. Similarly, if a system filter added multiple headers in this way,
5517 the routers could not see them.
5519 20. Expanded the error message when iplsearch is called with an invalid key to
5520 suggest using net-iplsearch in a host list.
5522 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5523 ACLs are no longer actually imposed (and a message to that effect is
5526 22. If a "gecos" field in a passwd entry contained escaped characters, in
5527 particular, if it contained a \" sequence, Exim got it wrong when building
5528 a From: or a Sender: header from that name. A second bug also caused
5529 incorrect handling when an unquoted " was present following a character
5530 that needed quoting.
5532 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5533 was not being matched caselessly.
5535 24. Arranged for all hyphens in the exim.8 source to be escaped with
5538 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5539 properly. Recipient callout cache records were still being keyed to include
5540 the sender, even when use_sender was set false. This led to far more
5541 callouts that were necessary. The sender is no longer included in the key
5542 when use_sender is false.
5544 26. Added "control = submission" modifier to ACLs.
5546 27. Added the ${base62d: operator to decode base 62 numbers.
5548 28. dnsdb lookups can now access SRV records.
5550 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5551 the configuration file.
5553 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5554 (-v) mode. This makes the output for a verbose queue run more intelligible.
5556 31. Added a use_postmaster feature to recipient callouts.
5558 32. Added the $body_zerocount variable, containing the number of binary zero
5559 bytes in the message body.
5561 33. The time of last modification of the "new" subdirectory is now used as the
5562 "mailbox time last read" when there is a quota error for a maildir
5565 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5567 35. Added +ignore_unknown as a special item in host lists.
5569 36. Code for decoding IPv6 addresses in host lists is now included, even if
5570 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5571 address was recognized as an IP address, but was then not correctly decoded
5572 into binary, causing unexpected and incorrect effects when compared with
5579 1. Very minor rewording of debugging text in manualroute to say "list of
5580 hosts" instead of "hostlist".
5582 2. If verify=header_syntax was set, and a header line with an unqualified
5583 address (no domain) and a large number of spaces between the end of the
5584 name and the colon was received, the reception process suffered a buffer
5585 overflow, and (when I tested it) crashed. This was caused by some obsolete
5586 code that should have been removed. The fix is to remove it!
5588 3. When running in the test harness, delay a bit after writing a bounce
5589 message to get a bit more predictability in the log output.
5591 4. Added a call to search_tidyup() just before forking a reception process. In
5592 theory, someone could use a lookup in the expansion of smtp_accept_max_
5593 per_host which, without the tidyup, could leave open a database connection.
5595 5. Added the variables $recipient_data and $sender_data which get set from a
5596 lookup success in an ACL "recipients" or "senders" condition, or a router
5597 "senders" option, similar to $domain_data and $local_part_data.
5599 6. Moved the writing of debug_print from before to after the "senders" test
5602 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5603 problems for message scanning, either using a data ACL, or using
5604 local_scan() because the Received: header was not generated till after they
5605 were called (in order to set the time as the time of reception completion).
5606 I have revised the way this works. The header is now generated after the
5607 body is received, but before the ACL or local_scan() are called. After they
5608 are run, the timestamp in the header is updated.
5614 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5615 before starting a queue runner without re-exec. This happened only when
5616 deliver_drop_privilege was set or when the Exim user was set to root. The
5617 effect of the bug was that timeouts during subsequent deliveries caused
5618 crashes instead of being properly handled. The handler is now left at its
5619 default (and expected) setting.
5621 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5622 message, again when deliver_drop_privilege is set or Exim is run as root.
5623 The bug described in (1) was not present in this case, but the tidying up
5624 of the other signals was missing. I have made the two cases consistent.
5626 3. The ignore_target_hosts setting on a manualroute router was being ignored
5627 for hosts that were looked up using the /MX notation.
5629 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5632 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5633 operated on the sender address. After changing the $sender_address to <>
5634 for the sender address verify, Exim was re-instated it as the original
5635 (before rewriting) address, but remembering that it had rewritten it, so it
5636 wasn't rewriting it again. This bug also had the effect of breaking the
5637 sender address verification caching when the sender address was rewritten.
5639 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5640 This has been changed so that if the ip literal address matches
5641 ignore_target_hosts, the router declines.
5643 7. Added expansion conditions match_domain, match_address, and match_local_
5644 part (NOT match_host).
5646 8. The placeholder for the Received: header didn't have a length field set.
5648 9. Added code to Exim itself and to exim_lock to test for a specific race
5649 condition that could lead to file corruption when using MBX delivery. The
5650 issue is with the lockfile that is created in /tmp. If this file is removed
5651 after a process has opened it but before that process has acquired a lock,
5652 there is the potential for a second process to recreate the file and also
5653 acquire a lock. This could lead to two Exim processes writing to the file
5654 at the same time. The added code performs the same test as UW imapd; it
5655 checks after acquiring the lock that its file descriptor still refers to
5656 the same named file.
5658 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5659 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5662 11. Added the smtp_active_hostname option. If used, this will typically be made
5663 to depend on the incoming interface address. Because $interface_address is
5664 not set up until the daemon has forked a reception process, error responses
5665 that can happen earlier (such as "too many connections") no longer contain
5668 12. If an expansion in a condition on a "warn" statement fails because a lookup
5669 defers, the "warn" statement is abandoned, and the next ACL statement is
5670 processed. Previously this caused the whole ACL to be aborted.
5672 13. Added the iplsearch lookup type.
5674 14. Added ident_timeout as a log selector.
5676 15. Added tls_certificate_verified as a log selector.
5678 16. Added a global option tls_require_ciphers (compare the smtp transport
5679 option of the same name). This controls incoming TLS connections.
5681 17. I finally figured out how to make tls_require_ciphers do a similar thing
5682 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5683 before starting the TLS session.
5685 18. Tabs are now shown as \t in -bP output.
5687 19. If the log selector return_path_on_delivery was set, Exim crashed when
5688 bouncing a message because it had too many Received: header lines.
5690 20. If two routers both had headers_remove settings, and the first one included
5691 a superfluous trailing colon, the final name in the first list and the
5692 first name in the second list were incorrectly joined into one item (with a
5693 colon in the middle).
5699 1. Added -C and -D options to the exinext utility, mainly to make it easier
5700 to include in the automated testing, but these could be helpful when
5701 multiple configurations are in use.
5703 2. The exinext utility was not formatting the output nicely when there was
5704 an alternate port involved in the retry record key, nor when there was a
5705 message id as well (for retries that were specific to a specific message
5706 and a specific host). It was also confused by IPv6 addresses, because of
5707 the additional colons they contain. I have fixed the IPv4 problem, and
5708 patched it up to do a reasonable job for IPv6.
5710 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5711 delivery, the log line now contains "pipelined" if PIPELINING was used.
5713 4. An SMTP transport process used to panic and die if the bind() call to set
5714 an explicit outgoing interface failed. This has been changed; it is now
5715 treated in the same way as a connect() failure.
5717 5. A reference to $sender_host_name in the part of a conditional expansion
5718 that was being skipped was still causing a DNS lookup. This no longer
5721 6. The def: expansion condition was not recognizing references to header lines
5722 that used bh_ and bheader_.
5724 7. Added the _cache feature to named lists.
5726 8. The code for checking quota_filecount in the appendfile transport was
5727 allowing one more file than it should have been.
5729 9. For compatibility with Sendmail, the command line option
5737 and sets the incoming protocol and host name (for trusted callers). The
5738 host name and its colon can be omitted when only the protocol is to be set.
5739 Note the Exim already has two private options, -pd and -ps, that refer to
5740 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5741 "s", but I don't think that's a major issue.
5743 10. A number of refactoring changes to the code, none of which should affect
5746 (a) The number of logging options was getting close to filling up the
5747 32-bit word that was used as a bit map. I have split them into two classes:
5748 those that are passed in the argument to log_write(), and those that are
5749 only ever tested independently outside of that function. These are now in
5750 separate 32-bit words, so there is plenty of room for expansion again.
5751 There is no change in the user interface or the logging behaviour.
5753 (b) When building, for example, log lines, the code previously used a
5754 macro that called string_cat() twice, in order to add two strings. This is
5755 not really sufficiently general. Furthermore, there was one instance where
5756 it was actually wrong because one of the argument was used twice, and in
5757 one call a function was used. (As it happened, calling the function twice
5758 did not affect the overall behaviour.) The macro has been replaced by a
5759 function that can join an arbitrary number of extra strings onto a growing
5762 (c) The code for expansion conditions now uses a table and a binary chop
5763 instead of a serial search (which was left over from when there were very
5764 few conditions). Also, it now recognizes conditions like "pam" even when
5765 the relevant support is not compiled in: a suitably worded error message is
5766 given if an attempt is made to use such a condition.
5768 11. Added ${time_interval:xxxxx}.
5770 12. A bug was causing one of the ddress fields not to be passed back correctly
5771 from remote delivery subprocesses. The field in question was not being
5772 subsequently used, so this caused to problems in practice.
5774 13. Added new log selectors queue_time and deliver_time.
5776 14. Might have fixed a bug in maildirsizefile handling that threw up
5777 "unexpected character" debug warnings, and recalculated the data
5778 unnecessarily. In any case, I expanded the warning message to give more
5781 15. Added the message "Restricted characters in address" to the statements in
5782 the default ACL that block characters like @ and % in local parts.
5784 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5785 Three changes have been made:
5787 (a) There was a serious bug; a negative response to MAIL caused the whole
5788 recipient domain to be cached as invalid, thereby blocking all messages
5789 to all local parts at the same domain, from all senders. This bug has
5790 been fixed. The domain is no longer cached after a negative response to
5791 MAIL if the sender used is not empty.
5793 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5796 (c) A new callout option, "use_sender" has been added for people who want
5797 the modified behaviour.
5803 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5806 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5807 indeed breaks things for older releases.
5809 3. Added additional logging to the case where there is a problem reading data
5810 from a filter that is running in a subprocess using a pipe, in order to
5811 try to track down a specific problem.
5813 4. Testing facility fudge: when running in the test harness and attempting
5814 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5815 getting "No route to host". Convert this to a timeout.
5817 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5820 6. Some OS don't have socklen_t but use size_t instead. This affects the
5821 fifth argument of getsockopt() amongst other things. This is now
5822 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5823 can be set for individual OS. I have set it for SunOS5, OSF1, and
5824 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5825 some earlier ones do not.
5827 7. Change 4.30/15 was not doing the test caselessly.
5829 8. The standard form for an IPv6 address literal was being rejected by address
5830 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5831 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5832 this, as well as the form without the "IPv6" on the front (but only when
5833 address literals are enabled, of course).
5835 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5837 10. Exim crashed if a message with an empty sender address specified by -f
5838 encountered a router with an errors_to setting. This could be provoked only
5839 by a command such as
5843 where an empty string was supplied; "<>" did not hit this bug.
5845 11. Installed PCRE release 4.5.
5847 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5848 remained set. It is now erased.
5850 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5851 times from message ids (which are base 36 rather than the normal 62).
5853 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5854 were being counted as actual protocol errors, and logged if the log
5855 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5856 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5857 and DATA following a set of rejected RCPTs do not count as protocol errors.
5858 In other words, Exim assumes they were pipelined, though this may not
5859 actually be the case. Of course, in all cases the client gets an
5860 appropriate error code.
5862 15. If a lookup fails in an ACL condition, a message about the failure may
5863 be available; it is used if testing the ACL cannot continue, because most
5864 such messages specify what the cause of the deferral is. However, some
5865 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5866 that caused an old message to be retained and used if a later statement
5867 caused a defer, replacing the real cause of the deferral.
5869 16. If an IP address had so many PTR records that the DNS lookup buffer
5870 was not large enough to hold them, Exim could crash while trying to process
5871 the truncated data. It now detects and logs this case.
5873 17. Further to 4.21/58, another change has been made: if (and only if) the
5874 first line of a message (the first header line) ends with CRLF, a bare LF
5875 in a subsequent header line has a space inserted after it, so as not to
5876 terminate the header.
5878 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5879 unnecessarily, used atoi() instead of strtol(), and didn't check the
5880 termination when getting file sizes from file names by regex.
5882 19. Completely re-implemented the support for maildirsize files, in the light
5883 of a number of problems with the previous contributed implementation
5884 (4.30/29). In particular:
5886 . If the quota is zero, the maildirsize file is maintained, but no quota is
5889 . If the maildir directory does not exist, it is created before any attempt
5890 to write a maildirsize file.
5892 . The quota value in the file is just a cache; if the quota is changed in
5893 the transport, the new value overrides.
5895 . A regular expression is available for excluding directories from the
5898 20. The autoreply transport checks the characters in options that define the
5899 message's headers; it allows continued headers, but it was checking with
5900 isspace() after an embedded newline instead of explicitly looking for a
5903 21. If all the "regular" hosts to which an address was routed had passed their
5904 expiry times, and had not reached their retry times, the address was
5905 bounced, even if fallback hosts were defined. Now Exim should go on to try
5908 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5909 equivalent code in the SMTP transport. Some hosts send humungous responses
5910 to HELO/EHLO, more than 1024 it seems.
5912 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5913 gives compiler warnings in some environments. I've now done it "properly",
5916 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5917 (because of problems with the built-in one) was declared to return uschar *
5918 instead of char *, causing compiler failure.
5920 25. Fixed a file descriptor leak when processing alias/forward files.
5922 26. Fixed a minor format string issue in dbfn.c.
5924 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5926 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5927 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5930 29. When a sender address is verified, it is cached, to save repeating the test
5931 when there is more than one recipient in a message. However, when the
5932 verification involves a callout, it is possible for different callout
5933 options to be set for different recipients. It is too complicated to keep
5934 track of this in the cache, so now Exim always runs a verification when a
5935 callout is required, relying on the callout cache for the optimization.
5936 The overhead is duplication of the address routing, but this should not be
5939 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5940 to be verified with callout=postmaster, and the main callout worked but the
5941 postmaster check failed, the verification correctly failed. However, if a
5942 subsequent RCPT command asked for sender verification *without* the
5943 postmaster check, incorrect caching caused this verification also to fail,
5946 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5947 it was not caching the DNS options (qualify_single, search_parents) that
5948 were used when the lookup failed. A subsequent lookup with different
5949 options therefore always gave the same answer, though there were cases
5950 where it should not have. (Example: a "domains = !$mx_any" option on a
5951 dnslookup router: the "domains" option is always processed without any
5952 widening, but the router might have qualify_single set.) Now Exim uses the
5953 cached value only when the same options are set.
5955 32. Added John Jetmore's "exipick" utility to the distribution.
5957 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5958 than a timeout (e.g. a certificate is required, and is not provided), an
5959 Exim server now closes the connection immediately. Previously it waited for
5960 the client to close - but if the client is SSL, it seems that they each
5961 wait for each other, leading to a delay before one of them times out.
5963 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5964 maintained 0.8.x compatibility because I don't think many are using it, and
5965 it is clearly obsolete.
5967 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5970 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5971 client certificate was expired. A simple patch fixes this, though I don't
5972 understand the full logic of why the verify callback is called multiple
5975 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5976 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5977 which causes problems with some clients (such as the Certicom SSL Plus
5978 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5979 disables the coutermeasure allowing Eudora to connect."
5981 38. Exim was not checking that a write() to a log file succeeded. This could
5982 lead to Bad Things if a log got too big, in particular if it hit a file
5983 size limit. Exim now panics and dies if it cannot write to a log file, just
5984 as it does if it cannot open a log file.
5986 39. Modified OS/Makefile-Linux so that it now contains
5988 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5990 The two -D definitions ensure that Exim is compiled with large file
5991 support, which makes it possible to handle log files that are bigger than
5994 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5995 instance) a domain was checked against a named list that involved a lookup,
5996 causing $domain_data to be set, then another domain was checked against the
5997 same list, then the first domain was re-checked, the value of $domain_data
5998 after the final check could be wrong. In particular, if the second check
5999 failed, it could be set empty. This bug probably also applied to
6002 41. The strip_trailing_dot option was not being applied to the address given
6003 with the -f command-line option.
6005 42. The code for reading a message's header from the spool was incrementing
6006 $received_count, but never initializing it. This meant that the value was
6007 incorrect (doubled) while delivering a message in the same process in which
6008 it was received. In the most common configuration of Exim, this never
6009 happens - a fresh exec is done - but it can happen when
6010 deliver_drop_privilege is set.
6012 43. When Exim logs an SMTP synchronization error - client data sent too soon -
6013 it now includes up to 150 characters of the unexpected data in the log
6016 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6017 and building data strings. The size of both of these buffers was 10 000
6018 bytes - far larger than anybody would *ever* want, thought I. Needless to
6019 say, somebody hit the limit. I have increased the maximum line length to
6020 20 000 and the maximum data length of concatenated lines to 100 000. I have
6021 also fixed two bugs, because there was no checking on these buffers. Tsk,
6022 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6023 buffer is too small.
6025 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6026 lsearch lookups. Now it does.
6028 46. When parsing a route_list item in a manualroute router, a fixed-length
6029 buffer was used for the list of hosts. I made this 1024 bytes long,
6030 thinking that nobody would ever have a list of hosts that long. Wrong.
6031 Somebody had a whole pile of complicated expansion conditions, and the
6032 string was silently truncated, leading to an expansion error. It turns out
6033 that it is easier to change to an unlimited length (owing to other changes
6034 that have happened since this code was originally written) than to build
6035 structure for giving a limitation error. The length of the item that
6036 expands into the list of hosts is now unlimited.
6038 47. The lsearch lookup could not handle data where the length of text line was
6039 more than 4095 characters. Such lines were truncated, leading to shortened
6040 data being returned. It should now handle lines of any length.
6042 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6043 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6046 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6047 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6048 no longer happens in comment lines. A list of replacements is now placed
6049 at the head of all of the source files, except those whose only change is
6050 to replace PERL_COMMAND in the very first #! line.
6052 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6053 messages on the queue, with a bottom-up merge sort, using code contributed
6054 by Michael Haardt. This should make operations like -bp somewhat faster on
6055 large queues. It won't affect queue runners, except when queue_run_in_order
6058 51. Installed eximstats 1.31 in the distribution.
6060 52. Added support for SRV lookups to the dnslookup router.
6062 53. If an ACL referred to $message_body or $message_body_end, the value was not
6063 reset for any messages that followed in the same SMTP session.
6065 54. The store-handling optimization for building very long strings was not
6066 differentiating between the different store pools. I don't think this
6067 actually made any difference in practice, but I've tidied it.
6069 55. While running the routers to verify a sender address, $sender_address
6070 was still set to the sender address. This is wrong, because when routing to
6071 send a bounce to the sender, it would be empty. Therefore, I have changed
6072 it so that, while verifying a sender address, $sender_address is set to <>.
6073 (There is no change to what happens when verifying a recipient address.)
6075 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6076 target A or AAAA records (if not already returned) without resetting the
6077 qualify_single or search_parents options of the DNS resolver. These are
6078 inappropriate in this case because the targets of MX and SRV records must
6079 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6080 target that, when qualified, matched something in the local domain. These
6081 two options are now turned off when doing these lookups.
6083 57. It seems that at least some releases of Reiserfs (which does not have the
6084 concept of a fixed number of inodes) returns zero and not -1 for the
6085 number of available inodes. This interacted badly with check_spool_inodes,
6086 which assumed that -1 was the "no such thing" setting. What I have done is
6087 to check that the total number of inodes is greater than zero before doing
6088 the test of how many are available.
6090 58. When a "warn" ACL statement has a log_message modifier, the message is
6091 remembered, and not repeated. This is to avoid a lot of repetition when a
6092 message has many recipients that cause the same warning to be written.
6093 However, Exim was preserving the list of already written lines for an
6094 entire SMTP session, which doesn't seem right. The memory is now reset if a
6095 new message is started.
6097 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6098 debugging output unless log_rewrite was also set.
6100 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6101 of (int)(handle) when we know that handle contains (void *)(-1).
6103 61. The Exim daemon panic-logs an error return when it closes the incoming
6104 connection. However "connection reset by peer" seems to be common, and
6105 isn't really an error worthy of noting specially, so that particular error
6108 62. When Exim is trying to find all the local interfaces, it used to panic and
6109 die if the ioctl to get the interface flags failed. However, it seems that
6110 on at least one OS (Solaris 9) it is possible to have an interface that is
6111 included in the list of interfaces, but for which you get a failure error
6112 for this call. This happens when the interface is not "plumbed" into a
6113 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6114 failure of the "get flags" call assumes that the interface is down.
6116 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6117 makes life easier for people who are doing arithmetic on fields extracted
6118 from dates, where you often get leading zeros that should not be
6119 interpreted as octal.
6121 64. Added qualify_domain to the redirect router, to override the global
6124 65. If a pathologically long header line contained very many addresses (the
6125 report of this problem mentioned 10 000) and each of them was rewritten,
6126 Exim could use up a very large amount of memory. (It kept on making new
6127 copies of the header line as it rewrote, and never released the old ones.)
6128 At the expense of a bit more processing, the header rewriting function has
6129 been changed so that it no longer eats memory in this way.
6131 66. The generation of the Received: header has been moved from the time that a
6132 message starts to be received, to the time that it finishes. The timestamp
6133 in the Received: header should now be very close to that of the <= log
6134 line. There are two side-effects of this change:
6136 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6137 logged header lines no longer include the local Received: line, because
6138 it has not yet been created. The same applies to a copy of the message
6139 that is returned to a non-SMTP sender when a message is rejected.
6141 (b) When a filter file is tested using -bf, no additional Received: header
6142 is added to the test message. After some thought, I decided that this
6145 This change does not affect the value of $received_for. It is still set
6146 after address rewriting, but before local_scan() is called.
6148 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6150 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6151 gave an unhelpful panic error message, and a defer error. I have managed to
6152 change this behaviour so that it now rejects any supplied certificate,
6153 which seems right, as the list of acceptable certificates is empty.
6155 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6156 gave an unhelpful defer error. I have not managed to make this reject any
6157 supplied certificates, but the error message it gives is "no certificate
6158 supplied", which is not helpful.
6160 70. exigrep's output now also includes lines that are not associated with any
6161 message, but which match the given pattern. Implemented by a patch from
6162 Martin Sluka, which also tidied up the Perl a bit.
6164 71. Recipient callout verification, like sender verification, was using <> in
6165 the MAIL FROM command. This isn't really the right thing, since the actual
6166 sender may affect whether the remote host accepts the recipient or not. I
6167 have changed it to use the actual sender in the callout; this means that
6168 the cache record is now keyed on a recipient/sender pair, not just the
6169 recipient address. There doesn't seem to be a real danger of callout loops,
6170 since a callout by the remote host to check the sender would use <>.
6171 [SEE ABOVE: changed after hitting problems.]
6173 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6174 temporary errors. However, in the case of such a code being given after
6175 the end of a data transmission (i.e. after ".") Exim was failing to write
6176 a retry record for the message. (Yes, there was some broken host that was
6177 actually sending 8xx at this point.)
6179 73. An unknown lookup type in a host list could cause Exim to panic-die when
6180 the list was checked. (An example that provoked this was putting <; in the
6181 middle of a list instead of at the start.) If this happened during a DATA
6182 ACL check, a -D file could be left lying around. This kind of configuration
6183 error no longer causes Exim to die; instead it causes a defer error. The
6184 incident is still logged to the main and panic logs.
6186 74. Buglet left over from Exim 3 conversion. The message "too many messages
6187 in one connection" was written to the rejectlog but not the mainlog, except
6188 when address rewriting (yes!) was being logged.
6190 75. Added write_rejectlog option.
6192 76. When a system filter was run not as root (that is, when system_filter_user
6193 was set), the values of the $n variables were not being returned to the
6194 main process; thus, they were not subsequently available in the $sn
6197 77. Added +return_path_on_delivery log selector.
6199 78. A connection timeout was being treated differently from recipients deferred
6200 when testing hosts_max_try with a message that was older than the host's
6201 retry timeout. (The host should not be counted, thus allowing all hosts to
6202 be tried at least once before bouncing.) This may have been the cause of an
6203 occasionally reported bug whereby a message would remain on the queue
6204 longer than the retry timeout, but would be bounced if a delivery was
6205 forced. I say "may" because I never totally pinned down the problem;
6206 setting up timeout/retry tests is difficult. See also the next item.
6208 79. The ultimate address timeout was not being applied to errors that involved
6209 a combination of host plus message (for example, a timeout on a MAIL
6210 command). When an address resolved to a number of possible hosts, and they
6211 were not all tried for each delivery (e.g. because of hosts_max_try), a
6212 message could remain on the queue longer than the retry timeout.
6214 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6217 81. Fixed an obscure SMTP outgoing bug which required at least the following
6218 conditions: (a) there was another message waiting for the same server;
6219 (b) the server returned 5xx to all RCPT commands in the first message so
6220 that the message was not completed; (c) the server dropped the connection
6221 or gave a negative response to the RSET that Exim sends to abort the
6222 transaction. The observed case was a dropped connection after DATA that had
6223 been sent in pipelining mode. That is, the server had advertised PIPELINING
6224 but was not implementing it correctly. The effect of the bug was incorrect
6225 behaviour, such as trying another host, and this could lead to a crash.
6231 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6232 and daemon.c were passed as pointers to ints; they should have been
6233 pointers to socklen_t variables (which are typically unsigned ints).
6235 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6238 3. Fixed a really odd bug that affected only the testing scheme; patching a
6239 certain fixed string in the binary changed the value of another string that
6240 happened to be identical to the end of the original first string.
6242 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6243 name", it returns that address as the IP address. On some operating
6244 systems (e.g. Solaris), it also passes back the IP address string as the
6245 "host name". However, on others (e.g. Linux), it passes back an empty
6246 string. Exim wasn't checking for this, and was changing the host name to an
6247 empty string, assuming it had been canonicalized.
6249 5. Although rare, it is permitted to have more than one PTR record for a given
6250 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6251 all the names associated with an address, because they do in Solaris.
6252 However, it seems that they do not in Linux for data that comes from the
6253 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6254 I found this out when I moved to a new Linux workstation and tried to run
6255 the Exim test suite.
6257 To get round this problem I have changed the code so that it now does its
6258 own call to the DNS to look up PTR records when searching for a host name.
6259 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6260 addresses that are only in /etc/hosts are still found.
6262 This behaviour is, however, controlled by an option called host_lookup_
6263 order, which defaults to "bydns:byaddr". If people want to use the other
6264 order, or indeed, just use one or the other means of lookup, they can
6265 specify it in this variable.
6267 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6268 some operating systems, this comes back from gethostbyaddr() as an empty
6269 string, and this is what Exim used to test for. However, it seems that in
6270 other systems, "." is yielded. Exim now tests for this case too.
6272 7. The values of check_spool_space and check_log_space are now held internally
6273 as a number of kilobytes instead of an absolute number of bytes. If a
6274 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6275 kilobyte. This means that much larger values can be stored.
6277 8. Exim monitor: an attempt to get the action menu when not actually pointing
6278 at a message produces an empty menu entitled "No message selected". This
6279 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6280 no entries in it ("Shell widget menu has zero width and/or height"). So I
6281 have added a single, blank menu entry in this case.
6283 9. Added ${quote_local_part.
6285 10. MIME decoding is now applied to the contents of Subject: header lines when
6288 11. Now that a reference to $sender_host_address automatically causes a reverse
6289 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6290 host lookup before query-style lookups in lists that might use this
6291 variable. This has therefore been abolished, and the "net-" prefix is no
6292 longer necessary for query-style lookups.
6294 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6295 have been a typo for LFLAGS, so it has been changed.
6297 13. The install script calls Exim with "-C /dev/null" in order to find the
6298 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6299 to be output. However, since Exim outputs its version number before the
6300 error, it didn't break the script. It just looked ugly. I fixed this by
6301 always allowing "-C /dev/null" if the caller is root.
6303 14. Ignore overlarge ACL variable number when reading spool file - insurance
6304 against a later release with more variables having written the file.
6306 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6307 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6308 this, as well as the form without the "IPv6" on the front.
6310 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6311 OS/Makefile-Darwin file.
6313 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6314 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6316 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6317 variable. However, some OS use ints for inodes. I've added cast to long int
6318 to get rid of the compiler warning.
6320 19. I had forgotten to lock out "/../" in configuration file names when
6321 ALT_CONFIG_PREFIX was set.
6323 20. Routers used for verification do not need to specify transports. However,
6324 if such a router generated a host list, and callout was configured, Exim
6325 crashed, because it could not find a port number from the (non-existent)
6326 transport. It now assumes port 25 in this circumstance.
6328 21. Added the -t option to exigrep.
6330 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6331 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6332 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6333 them is set, LOOKUP_LSEARCH is forced.
6335 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6337 24. Added sender and host information to the "rejected by local_scan()" log
6338 line; previously there was no indication of these.
6340 25. Added .include_if_exists.
6342 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6343 Linux. It turns out that not all file systems support this. Apparently some
6344 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6345 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6346 not supported on the file descriptor, is now ignored when Exim is trying to
6347 sync a directory. This applies only to Linux.
6349 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6351 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6352 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6355 29. Added support for maildirsize files from supplied patch (modified a bit).
6357 30. The use of :fail: followed by an empty string could lead Exim to respond to
6358 sender verification failures with (e.g.):
6360 550 Verification failed for <xxx>
6361 550 Sender verify failed
6363 where the first response line was missing the '-' that indicates it is not
6364 the final line of the response.
6366 31. The loop for finding the name of the user that called Exim had a hardwired
6367 limit of 10; it now uses the value of finduser_retries, which is used for
6368 all other user lookups.
6370 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6373 33. Exim was neglecting to zero errno before one call of strtol() when
6374 expanding a string and expecting an integer value. On some systems this
6375 resulted in spurious "integer overflow" errors. Also, it was casting the
6376 result into an int without checking.
6378 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6379 did not work. The code looks as if it has *never* worked, though it appears
6380 to have been documented since at least release 1.62. I have made it work.
6382 35. The "timeout_DNS" error in retry rules, also documented since at least
6383 1.62, also never worked. As it isn't clear exactly what this means, and
6384 clearly it isn't a major issue, I have abolished the feature by treating it
6385 as "timeout", and writing a warning to the main and panic logs.
6387 36. The display of retry rules for -brt wasn't always showing the error code
6390 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6391 timeout_connect_A, timeout_connect_MX.
6393 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6394 to the empty sender.
6396 39. The daemon was not analysing the content of -oX till after it had closed
6397 stderr and disconnected from the controlling terminal. This meant that any
6398 syntax errors were only noted on the panic log, and the return code from
6399 the command was 0. By re-arranging the code a little, I've made the
6400 decoding happen first, so such errors now appear on stderr, and the return
6401 code is 1. However, the actual setting up of the sockets still happens in
6402 the disconnected process, so errors there are still only recorded on the
6405 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6406 connections (as happens on some IP stacks) was logged at start up time as
6407 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6408 it from "IPv6 and IPv4", which means that two separate sockets are being
6411 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6412 says whether AF_INET or AF_INET6 was passed as an argument.
6414 42. Exiwhat output was messed up when time zones were included in log
6417 43. Exiwhat now gives more information about the daemon's listening ports,
6418 and whether -tls-on-connect was used.
6420 44. The "port" option of the smtp transport is now expanded.
6422 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6423 silently ignored. Now an error message is written to the main and panic
6426 46. There's a new ACL modifier called "logwrite" which writes to a log file
6427 as soon as it is encountered.
6429 47. Added $local_user_uid and $local_user_gid at routing time.
6431 48. Exim crashed when trying to verify a sender address that was being
6434 49. Exim was recognizing only a space character after ".include". It now also
6435 recognizes a tab character.
6437 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6438 extracting the relevant information from the specification. The man page no
6439 longer contains scrambled data for the -d option, and I've added a section
6440 at the front about calling Exim under different names.
6442 51. Added "extra_headers" argument to the "mail" command in filter files.
6444 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6447 53. Installed eximstats 1.29.
6449 54. Added transport_filter_timeout as a generic transport option.
6451 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6452 Cc: header lines. This was required by RFC 822, but it not required by RFC
6455 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6456 incoming messages that did not have them. Now it does so only if the
6457 message originates locally, that is, if there is no associated remote host
6458 address. When Resent- header lines are present, this applies to the Resent-
6459 lines rather than the non-Resent- lines.
6461 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6462 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6464 58. Messages for configuration errors now include the name of the main
6465 configuration file - useful now that there may be more than one file in a
6466 list (.included file names were always shown).
6468 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6469 for those rare installations that do not start the daemon as root or run it
6470 setuid root. I've cut out the call to initgroups() if the daemon is not
6473 60. The Exim user and group can now be bound into the binary as text strings
6474 that are looked up at the start of Exim's processing.
6476 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6478 62. Added $mailstore_basename variable.
6480 63. Installed patch to sieve.c from Michael Haardt.
6482 64. When Exim failed to open the panic log after failing to open the main log,
6483 the original message it was trying to log was written to stderr and debug
6484 output, but if they were not available (the usual case in production), it
6485 was lost. Now it is written to syslog before the two lines that record the
6486 failures to open the logs.
6488 65. Users' Exim filters run in subprocesses under the user's uid. It is
6489 possible for a "deliver" command or an alias in a "personal" command to
6490 provoke an address rewrite. If logging of address rewriting is configured,
6491 this fails because the process is not running as root or exim. There may be
6492 a better way of dealing with this, but for the moment (because 4.30 needs
6493 to be released), I have disabled address rewrite logging when running a
6494 filter in a non-root, non-exim process.
6500 1. The buildconfig auxiliary program wasn't quoting the value set for
6501 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6502 not defined. This bug was masked in 4.22 by the effect that was fixed in
6505 2. Some messages that were rejected after a message id was allocated were
6506 shown as "incomplete" by exigrep. It no longer does this for messages that
6507 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6509 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6510 have allow_domain_literals set, the ID did not get logged in the <= line.
6511 Domain literals are now always recognized in Message-ID: header lines.
6513 4. The first argument for a ${extract expansion item is the key name or field
6514 number. Leading and trailing spaces in this item were not being ignored,
6515 causing some misleading effects.
6517 5. When deliver_drop_privilege was set, single queue runner processes started
6518 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6519 same command in the process it spins off) were not dropping privilege.
6521 6. When the daemon running as "exim" started a queue runner, it always
6522 re-executed Exim in the spun-off process. This is a waste of effort when
6523 deliver_drop_privilege is set. The new process now just calls the
6524 queue-runner function directly.
6530 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6533 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6534 ignored. Though the use of -oP was forcing the writing of a pid file, it
6535 was always written to the default place.
6537 3. If the message "no IP address found for host xxxx" is generated during
6538 incoming verification, it is now followed by identification of the incoming
6539 connection (so you can more easily find what provoked it).
6541 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6543 5. Added some features to "harden" Exim a bit more against certain attacks:
6545 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6546 be put in Local/Makefile. This is like the never_users runtime option,
6547 but it cannot be overridden. The default setting is "root".
6549 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6550 prefix string with which any file named in a -C command line option
6553 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6554 is retained for -C and -D only if the caller of Exim is root. Without
6555 it, the exim user may also use -C and -D and retain privilege.
6557 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6558 command line option is disabled.
6560 6. Macro names set by the -D option must start with an upper case letter, just
6561 like macro names defined in the configuration file.
6563 7. Added "dereference=" facility to LDAP.
6565 8. Two instances of the typo "uknown" in the source files are fixed.
6567 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6568 the Configure-Makefile script screwed up while processing it.
6570 10. Incorporated PCRE 4.4.
6572 11. The SMTP synchronization check was not operating right at the start of an
6573 SMTP session. For example, it could not catch a HELO sent before the client
6574 waited for the greeting. There is now a check for outstanding input at the
6575 point when the greeting is written. Because of the duplex, asynchronous
6576 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6577 way, but not yet received, when the check is performed.
6579 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6580 on TCP/IP sockets, because this apparently causes some broken clients to
6583 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6584 unchanged) from the Cygwin maintainer.
6586 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6587 was supported instead of testing for mbx. Effectively a typo.
6589 15. The spa authenticator server code was not checking that the input it
6590 received was valid base64.
6592 16. The debug output line for the "set" modifier in ACLs was not showing the
6593 name of the variable that was being set.
6595 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6597 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6598 lookup on its own. Something else was needed to trigger the lookup. For
6599 example, a match in host_lookup or the need for a host name in a host list.
6600 Now, if $sender_host_name is referenced and the host name has not yet been
6601 looked up, a lookup is performed. If the lookup fails, the variable remains
6602 empty, and $host_lookup_failed is set to "1".
6604 19. Added "eqi" as a case-independent comparison operator.
6606 20. The saslauthd authentication condition could segfault if neither service
6607 nor realm was specified.
6609 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6610 error message that was logged was misleading, and incoming SMTP
6611 connections were dropped. The message is now more accurate, and temporary
6612 errors are given to SMTP connections.
6614 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6615 (except RSET) with a 421 error, until QUIT is received. However, it was
6616 failing to send a response to QUIT.
6618 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6619 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6620 ACL is now run after the helo_try_verify_hosts code.
6622 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6623 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6624 case-independent, so other case variants are also recognized). Apparently
6625 some systems use these upper case variants.
6627 25. If more than two messages were waiting for the same host, and a transport
6628 filter was specified for the transport, Exim sent two messages over the
6629 same TCP/IP connection, and then failed with "socket operation on non-
6630 socket" when it tried to send the third.
6632 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6634 27. The extern definition of crypt16() in expand.c was not being excluded when
6635 the OS had its own crypt16() function.
6637 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6638 as a preferred synonym for return_size_limit, both as an option and as an
6641 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6643 30. Changed the default configuration ACL to relax the local part checking rule
6644 for addresses that are not in any local domains. For these addresses,
6645 slashes and pipe symbols are allowed within local parts, but the sequence
6646 /../ is explicitly forbidden.
6648 31. SPA server authentication was not clearing the challenge buffer before
6651 32. log_message in a "warn" ACL statement was writing to the reject log as
6652 well as to the main log, which contradicts the documentation and doesn't
6653 seem right (because no rejection is happening). So I have stopped it.
6655 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6656 However, I am unable to do any testing of this.
6658 34. Fixed an infelicity in the appendfile transport. When checking directories
6659 for a mailbox, to see if any needed to be created, it was accidentally
6660 using path names with one or more superfluous leading slashes; tracing
6661 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6663 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6664 contents are added to the log line that is written for every discarded
6665 recipient. (Previously a log_message setting was ignored.)
6667 36. The ${quote: operator now quotes the string if it is empty.
6669 37. The install script runs exim in order to find its version number. If for
6670 some reason other than non-existence or emptiness, which it checks, it
6671 could not run './exim', it was installing it with an empty version number,
6672 i.e. as "exim-". This error state is now caught, and the installation is
6675 38. An argument was missing from the function that creates an error message
6676 when Exim fails to connect to the socket for saslauthd authentication.
6677 This could cause Exim to crash, or give a corrupted message.
6679 39. Added isip, isip4, and isip6 to ${if conditions.
6681 40. The ACL variables $acl_xx are now saved with the message, and can be
6682 accessed later in routers, transports, and filters.
6684 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6685 strings in the file are not string-expanded.
6687 42. If a MAIL command specified a SIZE value that was too large to fit into an
6688 int variable, the check against message_size_limit failed. Such values are
6689 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6690 day this will have to be increased, but I don't think I want to be around
6691 when emails are that large.
6698 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6699 iconv() is not standard in FreeBSD.
6701 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6702 IPv6 enabled. The observed symptom was a segmentation fault on return from
6703 the function os_common_find_running_interfaces() in src/os.c.
6705 3. In the check_special_case() function in daemon.c I had used "errno" as an
6706 argument name, which causes warnings on some systems. This was basically a
6707 typo, since it was named "eno" in the comments!
6709 4. The code that waits for the clock to tick (at a resolution of some fraction
6710 of a second) so as to ensure message-id uniqueness was always waiting for
6711 at least one whole tick, when it could have waited for less. [This is
6712 almost certainly not relevant at current processor speeds, where it is
6713 unlikely to ever wait at all. But we try to future-proof.]
6715 5. The function that sleeps for a time interval that includes fractions of a
6716 second contained a race. It did not block SIGALRM between setting the
6717 timer, and suspending (a couple of lines later). If the interval was short
6718 and the sigsuspend() was delayed until after it had expired, the suspension
6719 never ended. On busy systems this could lead to processes getting stuck for
6722 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6723 process, before it forks any delivery processes. The open lookup caching
6724 mechanism meant that the open file or database connection was passed into
6725 the delivery process. The problem was that delivery processes always tidy
6726 up cached lookup data. This could cause a problem for the next delivery
6727 process started by the queue runner, because the external queue runner
6728 process does not know about the closure. So the next delivery process
6729 still has data in the lookup cache. In the case of a file lookup, there was
6730 no problem because closing a file descriptor in a subprocess doesn't affect
6731 the parent. However, if the lookup was caching a connection to a database,
6732 the connection was closed, and the second delivery process was likely to
6733 see errors such as "PGSQL: query failed: server closed the connection
6734 unexpectedly". The problem has been fixed by closing all cached lookups
6735 in a queue runner before running a delivery process.
6737 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6738 seem to have the "const" qualifier which it has on other OS. I've
6741 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6742 *of the same type* (client or server) with the same public name that an
6743 error should be diagnosed.
6745 9. When Exim looked up a host name for an IP address, but failed to find the
6746 original IP address when looking up the host name (a safety check), it
6747 output the message "<ip address> does not match any IP for NULL", which was
6748 confusing, to say the least. The bug was that the host name should have
6749 appeared instead of "NULL".
6751 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6752 user that is built into the binary, and the -C or -D options is used, root
6753 privilege is dropped before the configuration file is read. In addition,
6754 logging is switched to stderr instead of the normal log files. If the
6755 configuration then re-defines the Exim user, the unprivileged environment
6756 is probably not what is expected, so Exim logs a panic warning message (but
6759 However, if deliver_drop_privilege is set, the unprivileged state may well
6760 be exactly what is intended, so the warning has been cut out in that case,
6761 and Exim is allowed to try to write to its normal log files.
6767 1. smtp_return_error_details was not giving details for temporary sender
6768 or receiver verification errors.
6770 2. Diagnose a configuration error if two authenticators have the same public
6773 3. Exim used not to create the message log file for a message until the first
6774 delivery attempt. This could be confusing when incoming messages were held
6775 for policy or load reasons. The message log file is now created at the time
6776 the message is received, and an initial "Received" line is written to it.
6778 4. The automatically generated man page for command line options had a minor
6779 bug that caused no ill effects; however, a more serious problem was that
6780 the procedure for building the man page automatically didn't always
6781 operate. Consequently, release 4.20 contains an out-of-date version. This
6782 shouldn't happen again.
6784 5. When building Exim with embedded Perl support, the script that builds the
6785 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6786 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6788 6. The freeze_tell option was not being used for messages that were frozen on
6789 arrival, either by an ACL or by local_scan().
6791 7. Added the smtp_incomplete_transaction log selector.
6793 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6794 was accepting AUTH without a new EHLO.
6796 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6797 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6800 10. Logging of TCP/IP connections (when configured) now happens in the main
6801 daemon process instead of the child process, so that the TCP/IP connection
6802 count is more accurate (but it can never be perfect).
6804 11. The use of "drop" in a nested ACL was not being handled correctly in the
6805 outer ACL. Now, if condition failure induced by the nested "drop" causes
6806 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6807 or "require"), the connection is dropped.
6809 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6810 that yield "discard" can now be used with an "accept" or a "discard" verb,
6811 but an error is generated for any others (because I can't see a useful way
6812 to define what should happen).
6814 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6815 are now processed in the same way as lines in the Exim configuration file.
6816 In particular, continuation lines are supported.
6818 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6820 15. Added -ti meaning -t -i.
6822 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6823 domains, and warn by logging if others are found.
6825 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6826 returned from GIFCONF when Exim is trying to find the list of interfaces on
6827 a host. The code in os.c has been modified to copy each ifreq to an aligned
6828 structure in all cases.
6830 Also, in some cases, the returned ifreq's were being copied to a 'struct
6831 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6832 means the last couple of bytes of an IPv6 address could be chopped if the
6833 ifreq contained only a normal sockaddr (14 bytes storage).
6835 18. Named domain lists were not supported in the hosts_treat_as_local option.
6836 An entry such as +xxxx was not recognized, and was treated as a literal
6839 19. Ensure that header lines added by a DATA ACL are included in the reject log
6840 if the ACL subsequently rejects the message.
6842 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6843 MD5 (which is deprecated).
6845 21. When testing a filter file using -bf, Exim was writing a message when it
6846 took the sender from a "From " line in the message, but it was not doing so
6847 when it took $return_path from a Return-Path: header line. It now does.
6849 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6850 with a valid header line field name (a series of printing characters
6851 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6853 23. Changed "disc" in the source to "disk" to conform to the documentation and
6854 the book and for uniformity.
6856 24. Ignore Sendmail's -Ooption=value command line item.
6858 25. When execve() failed while trying to run a command in a pipe transport,
6859 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6860 could be confused with a return value of 69 from the command itself. This
6861 has been changed to 127, the value the shell returns if it is asked to run
6862 a non-existent command. The wording for the related log line suggests a
6863 non-existent command as the problem.
6865 26. If received_header_text expands to an empty string, do not add a Received:
6866 header line to the message. (Well, it adds a token one on the spool, but
6867 marks it "old" so that it doesn't get used or transmitted.)
6869 27. Installed eximstats 1.28 (addition of -nt option).
6871 28. There was no check for failure on the call to getsockname() in the daemon
6872 code. This can fail if there is a shortage of resources on the system, with
6873 ENOMEM, for example. A temporary error is now given on failure.
6875 29. Contrary to the C standard, it seems that in some environments, the
6876 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6877 program. Exim now does this explicitly; it affects the formatting of
6878 timestamps using strftime().
6880 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6881 complaints. I've now initialized all the variables, to avoid this.
6883 32. Header lines added by a system filter were not being "seen" during
6884 transport-time rewrites.
6886 33. The info_callback() function passed to OpenSSL is set up with type void
6887 (*)(SSL *, int, int), as described somewhere. However, when calling the
6888 function (actually a macro) that sets it up, the type void(*)() is
6889 expected. I've put in a cast to prevent warnings from picky compilers.
6891 34. If a DNS black list lookup found a CNAME record, but there were no A
6892 records associated with the domain it pointed at, Exim crashed.
6894 35. If a DNS black list lookup returned more than one A record, Exim ignored
6895 all but the first. It now scans all returned addresses if a particular IP
6896 value is being sought. In this situation, the contents of the
6897 $dnslist_value variable are a list of all the addresses, separated by a
6900 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6901 to accept a host name and all its aliases if the forward lookup for any of
6902 them yielded the IP address of the incoming connection. Now it accepts only
6903 those names whose forward lookup yields the correct IP address. Any other
6904 names are discarded. This closes a loophole whereby a rogue DNS
6905 administrator could create reverse DNS records to break through a
6906 wildcarded host restriction in an ACL.
6908 37. If a user filter or a system filter that ran in a subprocess used any of
6909 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6910 the wrong values were passed to the pipe command ($thisaddress had the
6911 value of $0, $0 had the value of $1, etc). This bug was introduced by
6912 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6914 38. Improved the line breaking for long SMTP error messages from ACLs.
6915 Previously, if there was no break point between 40 and 75 characters, Exim
6916 left the rest of the message alone. Two changes have been made: (a) I've
6917 reduced the minimum length to 35 characters; (b) if it can't find a break
6918 point between 35 and 75 characters, it looks ahead and uses the first one
6919 that it finds. This may give the occasional overlong line, but at least the
6920 remaining text gets split now.
6922 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6923 file descriptors might be low, and that setting 1000 would always raise it.
6924 It turns out that in some environments, the limit is already over 1000 and
6925 that lowering it causes trouble. So now Exim takes care not to decrease it.
6927 40. When delivering a message, the value of $return_path is set to $sender_
6928 address at the start of routing (routers may change the value). By an
6929 oversight, this default was not being set up when an address was tested by
6930 -bt or -bv, which affected the outcome if any router or filter referred to
6933 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6934 log only when "message" or "log_message" are set. However, if one of the
6935 conditions was an address verification, or a call to a nested ACL, the
6936 messages generated by the underlying test were being passed through. This
6937 no longer happens. The underlying message is available in $acl_verify_
6938 message for both "message" and "log_message" expansions, so it can be
6939 passed through if needed.
6941 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6942 new expansion $bh_ to give the encoded byte string without charset
6943 translation. Translation happens only if iconv() is available; HAVE_ICONV
6944 indicates this at build time. HEADERS_CHARSET gives the charset to
6945 translate to; headers_charset can change it in the configuration, and
6946 "headers charset" can change it in an individual filter file.
6948 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6949 that creates RFC 2047 encoded "words" labels them as that charset instead
6950 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6951 expansion operator; (ii) when Exim creates a From: line for a local
6952 message; (iii) when a header line is rewritten to include a "phrase" part.
6954 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6955 buggy, causing it to skip the first lines of messages whose message ID
6956 ended in 'D'. This would not have bitten before Exim release 4.14, because
6957 message IDs were unlikely to end in 'D' before then. The effect was to have
6958 incorrect size information for certain domains.
6960 45. #include "config.h" was missing at the start of the crypt16.c module. This
6961 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6964 46. If there was a timeout during a "random" callout check, Exim treated it as
6965 a failure of the random address, and carried on sending RSET and the real
6966 address. If the delay was just some slowness somewhere, the response to the
6967 original RCPT would be taken as a response to RSET and so on, causing
6968 mayhem of various kinds.
6970 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6971 when I implemented it. It didn't allow for the fact that some option values
6972 may legitimately be negative (e.g. size_addition), and it didn't even do
6973 the right test for positive values.
6975 48. Domain names in DNS records are case-independent. Exim always looks them up
6976 in lower case. Some resolvers return domain names in exactly the case they
6977 appear in the zone file, that is, they may contain uppercase letters. Not
6978 all resolvers do this - some return always lower case. Exim was treating a
6979 change of case by a resolver as a change of domain, similar to a widening
6980 of a domain abbreviation. This triggered its re-routing code and so it was
6981 trying to route what was effectively the same domain again. This normally
6982 caused routing to fail (because the router wouldn't handle the domain
6983 twice). Now Exim checks for this case specially, and just changes the
6984 casing of the domain that it ultimately uses when it transmits the message
6987 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6990 50. If a filter generated a file delivery with a non-absolute name (possible if
6991 no home directory exists for the router), the forbid_file option was not
6994 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6995 the existing equality matching.
6997 52. Exim was using ints instead of ino_t variables in some places where it was
6998 dealing with inode numbers.
7000 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
7001 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
7002 called TMPDIR, and if it finds it is different, it changes its value.
7004 54. The smtp_printf() function is now made available to local_scan() so
7005 additional output lines can be written before returning. There is also an
7006 smtp_fflush() function to enable the detection of a dropped connection.
7007 The variables smtp_input and smtp_batched_input are exported to
7010 55. Changed the default runtime configuration: the message "Unknown user"
7011 has been removed from the ACL, and instead placed on the localuser router,
7012 using the cannot_route_message feature. This means that any verification
7013 failures that generate their own messages won't get overridden. Similarly,
7014 the "Unrouteable address" message that was in the ACL for unverifiable
7015 relay addresses has also been removed.
7017 56. Added hosts_avoid_esmtp to the smtp transport.
7019 57. The exicyclog script was not checking for the esoteric option
7020 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7021 will work only if exicyclog is run under the appropriate euid.
7023 58. Following a discussion on the list, the rules by which Exim recognises line
7024 endings on incoming messages have been changed. The -dropcr and drop_cr
7025 options are now no-ops, retained only for backwards compatibility. The
7026 following line terminators are recognized: LF CRLF CR. However, special
7027 processing applies to CR:
7029 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7030 nor a local message in the state where . is a terminator.
7032 (ii) If a bare CR is encountered in a header line, an extra space is added
7033 after the line terminator so as not to end the header. The reasoning
7034 behind this is that bare CRs in header lines are most likely either
7035 to be mistakes, or people trying to play silly games.
7037 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7038 was being incorrectly given as 18 bytes larger than it should have been.
7039 This is a VOB (very old bug).
7041 60. This may never have affected anything current, but just in case it has:
7042 When the local host is found other than at the start of a list of hosts,
7043 the local host, those with the same MX, and any that follow, are discarded.
7044 When the list in question was part of a longer list of hosts, the following
7045 hosts (not currently being processed) were also being discarded. This no
7046 longer happens. I'm not sure if this situation could ever has previously
7049 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7052 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7053 header. This is something that is recommended in a new Internet Draft, and
7054 is something that is documented as being done by Sendmail. There are two
7055 possible values. For messages generated by the autoreply transport, Exim
7058 Auto-Submitted: auto-replied
7060 whereas for all other generated messages (e.g. bounces) it adds
7062 Auto-Submitted: auto-generated
7064 63. The "personal" condition in filters now includes a test for the
7065 Auto-Submitted: header. If it contains the string "auto-" the message it
7066 not considered personal.
7068 64. Added rcpt_include_affixes as a generic transport option.
7070 65. Added queue_only_override (default true).
7072 66. Added the syslog_duplication option.
7074 67. If what should have been the first header line of a message consisted of
7075 a space followed by a colon, Exim was mis-interpreting it as a header line.
7076 It isn't of course - it is syntactically invalid and should therefore be
7077 treated as the start of the message body. The misbehaviour could have
7078 caused a number of strange effects, including loss of data in subsequent
7079 header lines, and spool format errors.
7081 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7082 client host had authenticated. This control can now be exercised by an ACL
7083 for more flexibility.
7085 69. By default, callouts do not happen when testing with -bh. There is now a
7086 variant, -bhc, which does actually run the callout code, including
7087 consulting and updating the callout cache.
7089 70. Added support for saslauthd authentication, courtesy of Alexander
7092 71. If statvfs() failed on the spool or log directories while checking their
7093 size for availability, Exim confusingly gave the error "space shortage".
7094 Furthermore, in debugging mode it crashed with a floating point exception.
7095 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7096 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7097 serious problem, Exim now writes to the main and panic logs when this
7098 happens, with details of the failure. It then refuses to accept the
7099 incoming message, giving the message "spool directory problem" or "log
7100 directory problem" with a 421 code for SMTP messages.
7102 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7103 0, 1, and 2 exist, because some OS complain for execs without them (see
7104 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7105 descriptors. However, the code omitted to check that the open succeeded,
7106 causing mysterious errors if for some reason the permissions on /dev/null
7107 got screwed. Now Exim writes a message to the main and panic logs, and
7108 bombs out if it can't open /dev/null.
7110 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7111 interact so that it is all more flexible. It is supposed to remain
7112 backwards compatible. Also added extra_local_interfaces.
7114 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7115 to bomb out with an assertion failure - to the client this appears as a
7116 connection drop. This problem occurs in the part of the code that was taken
7117 from the Samba project. Fortunately, the assertion is in a very simple
7118 function, so I have fixed this by reproducing the function inline in the
7119 one place where it is called, and arranging for authentication to fail
7120 instead of killing the process with assert().
7122 75. The SPA client code was not working when the server requested OEM rather
7123 than Unicode encoding.
7125 76. Added code to make require_files with a specific uid setting more usable in
7126 the case where statting the file as root fails - usually a non-root-mounted
7127 NFS file system. When this happens and the failure is EACCES, Exim now
7128 forks a subprocess and does the per-uid checking as the relevant uid.
7130 77. Added process_log_path.
7132 78. If log_file_path was not explicitly set, a setting of check_log_space or
7133 check_log_inodes was ignored.
7135 79. If a space check for the spool or log partitions fails, the incident is now
7136 logged. Of course, in the latter case the data may get lost...
7138 80. Added the %p formatting code to string_format() so that it can be used to
7139 print addresses in debug_print(). Adjusted all the address printing in the
7140 debugging in store.c to use %p rather than %d.
7142 81. There was a concern that a line of code in smtp_in.c could overflow a
7143 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7144 initially expressed, the concern was not well-founded, because trailing
7145 spaces are removed early. However, if the trailing spaces were followed by
7146 a NULL, they did not get removed, so the overflow was possible. Two fixes
7149 (a) I re-wrote the offending code in a cleaner fashion.
7150 (b) If an incoming SMTP command contains a NULL character, it is rejected
7153 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7154 runs initgroups(), so that if the Exim user is in any additional groups,
7155 they will be used during message reception.
7161 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob