1 Change log file for Exim from version 4.21
2 ------------------------------------------
3 This document describes *changes* to previous versions, that might
4 affect Exim's operation, with an unchanged configuration file. For new
5 options, and new features, see the NewStuff file next to this ChangeLog.
8 Since Exim version 4.90
9 -----------------------
11 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
12 When redis_servers is set to a list of > 1 element, and the Redis servers
13 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
14 case of MOVED into a DEFER case instead, thus moving the query onto the
15 next server in the list. For a cluster of N elements, all N servers must
16 be defined in redis_servers.
18 JH/01 Replace the store_release() internal interface with store_newblock(),
19 which internalises the check required to safely use the old one, plus
20 the allocate and data copy operations duplicated in both (!) of the
23 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
24 modifier. This matches the restriction on the commandline.
26 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
27 Previously only the last row was returned.
29 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
30 we assumed that tags in the header were well-formed, and parsed the
31 element content after inspecting only the first char of the tag.
32 Assumptions at that stage could crash the receive process on malformed
35 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
36 While running the DKIM ACL we operate on the Permanent memory pool so that
37 variables created with "set" persist to the DATA ACL. Also (at any time)
38 DNS lookups that fail create cache records using the Permanent pool. But
39 expansions release any allocations made on the current pool - so a dnsdb
40 lookup expansion done in the DKIM ACL releases the memory used for the
41 DNS negative-cache, and bad things result. Solution is to switch to the
42 Main pool for expansions.
43 While we're in that code, add checks on the DNS cache during store_reset,
44 active in the testsuite.
45 Problem spotted, and debugging aided, by Wolfgang Breyha.
47 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
48 When none of the hosts presented to a transport match an already-open
49 connection, close it and proceed with the list. Previously we would
50 queue the message. Spotted by Lena with Yahoo, probably involving
53 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
54 Previously a spurious "250 OK id=" response was appended to the proper
57 JH/08 The "support for" informational output now, which built with Content
58 Scanning support, has a line for the malware scanner interfaces compiled
59 in. Interface can be individually included or not at build time.
61 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
62 by the template makefile "src/EDITME". The "STREAM" support for an older
63 ClamAV interface method is removed.
65 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
66 rows affected is given instead).
68 JH/11 The runtime Berkeley DB library version is now additionally output by
69 "exim -d -bV". Previously only the compile-time version was shown.
71 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
72 SMTP connection. Previously, when one had more receipients than the
73 first, an abortive onward connection was made. Move to full support for
74 multiple onward connections in sequence, handling cutthrough connection
75 for all multi-message initiating connections.
77 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
78 routers. Previously, a multi-recipient message would fail to match the
79 onward-connection opened for the first recipient, and cause its closure.
81 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
82 a timeout on read on a GnuTLS initiating connection, resulting in the
83 initiating connection being dropped. This mattered most when the callout
84 was marked defer_ok. Fix to keep the two timeout-detection methods
87 JH/15 Relax results from ACL control request to enable cutthrough, in
88 unsupported situations, from error to silently (except under debug)
89 ignoring. This covers use with PRDR, frozen messages, queue-only and
92 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
94 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
95 metadata, resulting in a crash in free().
97 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
98 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
99 Broken also in d185889f4, with init system revamp.
101 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
102 Previously we abruptly closed the connection after reading a malware-
103 found indication; now we go on to read the "scan ok" response line,
106 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
107 ACL. Previously, a crash would result.
109 JH/19 Speed up macro lookups during configuration file read, by skipping non-
110 macro text after a replacement (previously it was only once per line) and
111 by skipping builtin macros when searching for an uppercase lead character.
113 JH/20 DANE support moved from Experimental to mainline. The Makefile control
114 for the build is renamed.
116 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
117 was allocated for every new TLS startup, meaning one per message. Fix
118 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
120 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
121 reported the original. Fix to report (as far as possible) the ACL
122 result replacing the original.
124 JH/23 Fix memory leak during multi-message connections using STARTTLS under
125 OpenSSL. Certificate information is loaded for every new TLS startup,
126 and the resources needed to be freed.
128 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
130 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
133 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
134 DATA response info to the (existing) per-recipient response info for
135 the "C=" log element. It can have useful tracking info from the
136 destination system. Patch from Simon Arlott.
142 JH/01 Rework error string handling in TLS interface so that the caller in
143 more cases is responsible for logging. This permits library-sourced
144 string to be attached to addresses during delivery, and collapses
145 pairs of long lines into single ones.
147 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
148 during configuration. Wildcards are allowed and expanded.
150 JH/02 Rework error string handling in DKIM to pass more info back to callers.
151 This permits better logging.
153 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
154 do not close it down and have the child transport start it up again on
155 the passed-on TCP connection. Instead, proxy the child (and any
156 subsequent ones) for TLS via a unix-domain socket channel. Logging is
157 affected: the continued delivery log lines do not have any DNSSEC, TLS
158 Certificate or OCSP information. TLS cipher information is still logged.
160 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
161 identical IP addresses on different listening ports. Will also affect
164 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
165 add noisy ifdef guards to special-case this sillyness.
166 Patch from Bernd Kuhls.
168 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
169 than 255 are no longer allowed.
171 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
172 Disable the session-cache too, which might reduce our load. Since we
173 currrectly use a new context for every connection, both as server and
174 client, there is no benefit for these.
175 GnuTLS appears to not support tickets server-side by default (we don't
176 call gnutls_session_ticket_enable_server()) but client side is enabled
177 by default on recent versions (3.1.3 +) unless the PFS priority string
180 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
181 <https://reproducible-builds.org/specs/source-date-epoch/>.
183 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
184 the check for any unsuccessful recipients did not notice the limit, and
185 erroneously found still-pending ones.
187 JH/08 Pipeline CHUNKING command and data together, on kernels that support
188 MSG_MORE. Only in-clear (not on TLS connections).
190 JH/09 Avoid using a temporary file during transport using dkim. Unless a
191 transport-filter is involved we can buffer the headers in memory for
192 creating the signature, and read the spool data file once for the
193 signature and again for transmission.
195 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
196 4.77 as the kernel support then wasn't solid, having issues in 64bit
197 mode. Now, it's been long enough. Add support for FreeBSD also.
199 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
200 case where the routing stage had gathered several addresses to send to
201 a host before calling the transport for the first, we previously failed
202 to close down TLS in the old transport process before passing the TCP
203 connection to the new process. The new one sent a STARTTLS command
204 which naturally failed, giving a failed delivery and bloating the retry
205 database. Investigation and fix prototype from Wolfgang Breyha.
207 JH/12 Fix check on SMTP command input synchronisation. Previously there were
208 false-negatives in the check that the sender had not preempted a response
209 or prompt from Exim (running as a server), due to that code's lack of
210 awareness of the SMTP input buffering.
212 PP/04 Add commandline_checks_require_admin option.
213 Exim drops privileges sanely, various checks such as -be aren't a
214 security problem, as long as you trust local users with access to their
215 own account. When invoked by services which pass untrusted data to
216 Exim, this might be an issue. Set this option in main configuration
217 AND make fixes to the calling application, such as using `--` to stop
220 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
221 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
222 responses to those, into a single TLS record each way (this usually means
223 a single packet). As a side issue, smtp_enforce_sync now works on TLS
226 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
227 affects you only if you're dancing at the edge of the param size limits.
228 If you are, and this message makes sense to you, then: raise the
229 configured limit or use OpenSSL 1.1. Nothing we can do for older
232 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
233 cmdline element to get the documented default one. Previously it was
236 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
237 are unsuable for non-message-specific reasons - eg. connection timeout,
238 banner-time rejection.
240 JH/16 Fix logging of delivery remote port, when specified by router, under
243 PP/06 Repair manualroute's ability to take options in any order, even if one
244 is the name of a transport.
247 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
249 JH/17 Change the list-building routines interface to use the expanding-string
250 triplet model, for better allocation and copying behaviour.
252 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
253 Previously it was constructed the first time a possibly-matching string
254 was met in the configuration file input during startup; now it is done
257 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
258 compatible one, to avoid the (poorly documented) possibility of a config
259 file in the working directory redirecting the DB files, possibly correpting
260 some existing file. CVE-2017-10140 assigned for BDB.
262 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
263 cache-hot. Previously, although the result was properly cached, the
264 initial verify call returned a defer.
266 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
267 the main verify for receipient in uncached-mode.
269 JH/22 Retire historical build files to an "unsupported" subdir. These are
270 defined as "ones for which we have no current evidence of testing".
272 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
273 if present. Previously it was ignored.
275 JH/24 Start using specified-initialisers in C structure init coding. This is
276 a C99 feature (it's 2017, so now considered safe).
278 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
279 if was a fixed-sized field and bitmask ops via macros; it is now more
282 PP/07 GitHub PR 56: Apply MariaDB build fix.
283 Patch provided by Jaroslav Å karvada.
285 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
286 during Coverity cleanups [4.87 JH/47]
287 Diagnosis and fix provided by Michael Fischer v. Mollard.
289 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
290 the right size to place the terminating semicolon on its own folded
291 line, the header hash was calculated to an incorrect value thanks to
292 the (relaxed) space the fold became.
294 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
295 and confused the parent.
297 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
298 which could crash as a result. This could lead to undeliverable messages.
300 JH/28 Logging: "next input sent too soon" now shows where input was truncated
303 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
304 matters on fast-turnover and PID-randomising systems, which were getting
305 out-of-order delivery.
307 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
308 a possibly-overlapping copy. The symptom was that "Remote host closed
309 connection in response to HELO" was logged instead of the actual 4xx
312 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
313 Previously only that bufferd was discarded, resulting in SYMTP command
316 JH/32 DKIM: when a message has multiple signatures matching an identity given
317 in dkim_verify_signers, run the dkim acl once for each. Previously only
318 one run was done. Bug 2189.
320 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
321 "panic the current process" to "deliberately defer". The panic log is
322 still written with the problem list name; the mail and reject logs now
323 get a temp-reject line for the message that was being handled, saying
324 something like "domains check lookup or other defer". The SMTP 451
325 message is still "Temporary local problem".
327 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
328 A crafted sequence of BDAT commands could result in in-use memory beeing
329 freed. CVE-2017-16943.
331 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
332 from SMTP input. Previously it was always done; now only done for DATA
333 and not BDAT commands. CVE-2017-16944.
335 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
336 to the message (such as an overlong header line). Previously this was
337 not done and we did not exit BDAT mode. Followon from the previous item
338 though a different problem.
344 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
345 than -2003 did; needs libidn2 in addition to libidn.
347 JH/02 The path option on a pipe transport is now expanded before use.
349 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
350 Patch provided by "Björn", documentation fix added too.
352 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
353 missing a wire-to-host endian conversion.
355 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
356 close after a BDAT command line could be taken as a following command,
357 giving a synch failure. Fix by only checking for synch immediately
358 before acknowledging the chunk.
360 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
361 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
362 macro. Patches provided by Josh Soref.
364 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
365 Previously we did not; the RFC seems ambiguous and VRFY is not listed
366 by IANA as a service extension. However, John Klensin suggests that we
369 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
370 the dkim code may be unix-mode line endings rather than smtp wire-format
371 CRLF, so prepend a CR to any bare LF.
373 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
374 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
376 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
377 the first were themselves being wrongly included in the feed into dkim
378 processing; with most chunk sizes in use this resulted in an incorrect
379 body hash calculated value.
381 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
382 DKIM signature block, for verification. Although advised against by
383 standards it is specifically not ruled illegal.
385 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
387 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
388 missing a body hash (the bh= tag).
390 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
391 It seems that HAProxy sends the Proxy Protocol information in clear and
392 only then does a TLS startup, so do the same.
394 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
395 TCP connections (such as for Spamd) unless the daemon successfully set
396 Fast Open mode on its listening sockets. This fixes breakage seen on
397 too-old kernels or those not configured for Fast Open, at the cost of
398 requiring both directions being enabled for TFO, and TFO never being used
399 by non-daemon-related Exim processes.
401 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
402 endings, at least on the first header line. Try to canonify any that get
403 past that check, despite the cost.
405 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
406 now limited to an arbitrary five deep, while parsing addresses with the
407 strip_excess_angle_brackets option enabled.
409 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
410 instead leave the unprompted TLS handshake in socket buffer for the
411 TLS library to consume.
413 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
415 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
417 JH/16 Drop variables when they go out of scope. Memory management drops a whole
418 region in one operation, for speed, and this leaves assigned pointers
419 dangling. Add checks run only under the testsuite which checks all
420 variables at a store-reset and panics on a dangling pointer; add code
421 explicitly nulling out all the variables discovered. Fixes one known
422 bug: a transport crash, where a dangling pointer for $sending_ip_address
423 originally assigned in a verify callout, is re-used.
425 PP/06 Drop '.' from @INC in various Perl scripts.
427 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
429 PP/08 Reduce a number of compilation warnings under clang; building with
430 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
431 should be warning-free.
433 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
435 HS/01 Fix portability problems introduced by PP/08 for platforms where
436 realloc(NULL) is not equivalent to malloc() [SunOS et al].
438 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
439 chunk. This allows us to accept broken chunked messages. We need a more
440 general solution here.
442 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
443 already-broken messages in the queue.
445 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
447 JH/19 Fix reference counting bug in routing-generated-address tracking.
453 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
454 supports it and a size is available (ie. the sending peer gave us one).
456 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
457 years of being deprecated). The replacements are the ACLs
458 acl_smtp_mime and acl_not_smtp_mime.
460 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
461 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
462 or even an in-clear connection were permitted. Now, if the host lookup
463 was dnssec and dane was requested then the host is only used if the
464 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
465 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
466 if one fails this test.
467 This means that a poorly-configured remote DNS will make it incommunicado;
468 but it protects against a DNS-interception attack on it.
470 JH/04 Bug 1810: make continued-use of an open smtp transport connection
471 non-noisy when a race steals the message being considered.
473 JH/05 If main configuration option tls_certificate is unset, generate a
474 self-signed certificate for inbound TLS connections.
476 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
477 in rewrites and routers.
479 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
480 and logged a warning sing 4.83; now they are a configuration file error.
482 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
483 (lacking @domain). Apply the same qualification processing as RCPT.
485 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
487 JH/10 Support ${sha256:} applied to a string (as well as the previous
490 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
491 a cutthrough deliver is pending, as we always want to make a connection.
492 This also avoids re-routing the message when later placing the cutthrough
493 connection after a verify cache hit.
494 Do not update it with the verify result either.
496 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
497 when routing results in more than one destination address.
499 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
500 signing (which inhibits the cutthrough capability). Previously only
501 the presence of an option was tested; now an expansion evaluating as
502 empty is permissible (obviously it should depend only on data available
503 when the cutthrough connection is made).
505 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
506 the relevant preceding SMTP command did not note the pipelining mode.
508 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
509 Previously they were not counted.
511 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
512 as one having no matching records. Previously we deferred the message
513 that needed the lookup.
515 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
516 distinguished as "(=".
518 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
519 for missing MX records. Previously it only worked for missing A records.
521 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
523 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
524 after the data-go-ahead and data-ack. Patch from Jason Betts.
526 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
527 even for a "none" policy. Patch from Tony Meyer.
529 JH/22 Fix continued use of a connection for further deliveries. If a port was
530 specified by a router, it must also match for the delivery to be
533 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
534 When one of the recipients of a message was unsuitable for the connection
535 (has no matching addresses), we lost track of needing to mark it
536 deferred. As a result mail would be lost.
538 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
540 JH/25 Decoding ACL controls is now done using a binary search; the source code
541 takes up less space and should be simpler to maintain. Merge the ACL
542 condition decode tables also, with similar effect.
544 JH/26 Fix problem with one_time used on a redirect router which returned the
545 parent address unchanged. A retry would see the parent address marked as
546 delivered, so not attempt the (identical) child. As a result mail would
549 JH/27 Fix a possible security hole, wherein a process operating with the Exim
550 UID can gain a root shell. Credit to http://www.halfdog.net/ for
551 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
554 JH/28 Enable {spool,log} filesystem space and inode checks as default.
555 Main config options check_{log,spool}_{inodes,space} are now
556 100 inodes, 10MB unless set otherwise in the configuration.
558 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
559 Previously it only applied to the main-section connection policy
562 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
564 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
565 by me. Added RFC7919 DH primes as an alternative.
567 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
568 are not in the system include path.
570 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
571 GnuTLS, when a session startup failed (eg because the client disconnected)
572 Exim did stdio operations after fclose. This was exposed by a recent
573 change which nulled out the file handle after the fclose.
575 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
576 signed directly by the cert-signing cert, rather than an intermediate
577 OCSP-signing cert. This is the model used by LetsEncrypt.
579 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
581 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
582 an incoming connection.
584 HS/02 Bug 1802: Do not half-close the connection after sending a request
587 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
588 fallback to "prime256v1".
590 JH/34 SECURITY: Use proper copy of DATA command in error message.
591 Could leak key material. Remotely exploitable. CVE-2016-9963.
597 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
598 and 3.4.4 - once the server is enabled to respond to an OCSP request
599 it does even when not requested, resulting in a stapling non-aware
600 client dropping the TLS connection.
602 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
603 support variable-length bit vectors. No functional change.
605 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
606 The I= interface field on outgoing lines is now after the H= remote
607 host field, same as incoming lines. There is a separate
608 outgoing_interface log selector which allows you to disable the
611 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
612 If not running log_selector +smtp_connection the mainlog would be held
613 open indefinitely after a "too many connections" event, including to a
614 deleted file after a log rotate. Leave the per net connection logging
615 leaving it open for efficiency as that will be quickly detected by the
616 check on the next write.
618 HS/01 Bug 1671: Fix post transport crash.
619 Processing the wait-<transport> messages could crash the delivery
620 process if the message IDs didn't exist for some reason. When
621 using 'split_spool_directory=yes' the construction of the spool
622 file name failed already, exposing the same netto behaviour.
624 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
625 mime_regex ACL conditions.
627 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
628 to DSN fail messages (bounces): remote IP, remote greeting, remote response
629 to HELO, local diagnostic string.
631 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
632 log line to debug. Even when configured with a tls authenticator many
633 client connections are expected to not authenticate in this way, so
634 an authenticate fail is not an error.
636 HS/02 Add the Exim version string to the process info. This way exiwhat
637 gives some more detail about the running daemon.
639 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
640 matter for fast-change records such as DNSBLs.
642 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
643 retry record, even if constant. There may be multiple transports with
644 different interface settings and the retry behaviour needs to be kept
647 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
649 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
651 JH/10 Bug 840: fix log_defer_output option of pipe transport
653 JH/11 Bug 830: use same host for all RCPTS of a message, even under
654 hosts_randomize. This matters a lot when combined with mua_wrapper.
656 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
657 ${quote_pgsql:<string>} operator.
659 JH/13 Bug 1708: avoid misaligned access in cached lookup.
661 JH/14 Change header file name for freeradius-client. Relevant if compiling
662 with Radius support; from the Gentoo tree and checked under Fedora.
664 JH/15 Bug 1712: Introduce $prdr_requested flag variable
666 JH/16 Bug 1714: Permit an empty string as expansion result for transport
667 option transport_filter, meaning no filtering.
669 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
671 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
672 defaults to "*" (all hosts). The variable is now available when not built
673 with TLS, default unset, mainly to enable keeping the testsuite sane.
674 If a server certificate is not supplied (via tls_certificate) an error is
675 logged, and clients will find TLS connections fail on startup. Presumably
676 they will retry in-clear.
677 Packagers of Exim are strongly encouraged to create a server certificate
678 at installation time.
680 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
681 with the $config_file variable.
683 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
684 in transport context, after the attempt, and per-recipient. The latter type
685 is per host attempted. The event data is the error message, and the errno
686 information encodes the lookup type (A vs. MX) used for the (first) host,
687 and the trailing two digits of the smtp 4xx response.
689 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
690 to write to mainlog (or rejectlog, paniclog) in the window between file
691 creation and permissions/ownership being changed. Particularly affects
692 installations where exicyclog is run as root, rather than exim user;
693 result is that the running daemon panics and dies.
695 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
697 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
698 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
699 "pri" and "weight". Note that the previous implicit priority given by the
700 list order is no longer honoured.
702 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
705 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
706 by defining SUPPORT_SOCKS.
708 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
709 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
710 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
711 variables are renamed to proxy_{local,external}_{address,port}.
713 JH/27 Move Internationalisation support from Experimental to mainline, enabled
714 for a build by defining SUPPORT_I18N
716 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
717 of the query string, and make ${quote_redis:} do that quoting.
719 JH/29 Move Events support from Experimental to mainline, enabled by default
720 and removable for a build by defining DISABLE_EVENT.
722 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
724 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
725 cached by the daemon.
727 JH/32 Move Redis support from Experimental to mainline, enabled for a build
728 by defining LOOKUP_REDIS. The libhiredis library is required.
730 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
731 keys are given for lookup.
733 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
734 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
735 only supported when built with TLS support. The PolarSSL SHA routines
736 are still used when the TLS library is too old for convenient support.
738 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
739 openssl_options), for security. OpenSSL forces this from version 1.1.0
740 server-side so match that on older versions.
742 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
743 allocation for $value could be released as the expansion processing
744 concluded, but leaving the global pointer active for it.
746 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
747 and to use the domains and local_parts ACL conditions.
749 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
750 incorrectly not doubled on cutthrough transmission, hence seen as a
751 body-termination at the receiving system - resulting in truncated mails.
752 Commonly the sender saw a TCP-level error, and retransmitted the message
753 via the normal store-and-forward channel. This could result in duplicates
754 received - but deduplicating mailstores were liable to retain only the
755 initial truncated version.
757 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
759 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
761 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
762 we're in there, support oversigning also; bug 1309.
764 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
766 HS/04 Add support for keep_environment and add_environment options.
768 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
769 either intentional arithmetic overflow during PRNG, or testing config-
772 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
773 delivery resulted in actual delivery. Cancel cutthrough before DATA
776 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
777 rejects a recipient: pass the reject to the originator.
779 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
780 Many were false-positives and ignorable, but it's worth fixing the
783 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
784 for the new environment-manipulation done at startup. Move the routines
785 from being local to tls.c to being global via the os.c file.
787 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
788 an extract embedded as result-arg for a map, the first arg for extract
789 is unavailable so we cannot tell if this is a numbered or keyed
790 extraction. Accept either.
796 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
799 JH/02 The smtp transport option "multi_domain" is now expanded.
801 JH/03 The smtp transport now requests PRDR by default, if the server offers
804 JH/04 Certificate name checking on server certificates, when exim is a client,
805 is now done by default. The transport option tls_verify_cert_hostnames
806 can be used to disable this per-host. The build option
807 EXPERIMENTAL_CERTNAMES is withdrawn.
809 JH/05 The value of the tls_verify_certificates smtp transport and main options
810 default to the word "system" to access the system default CA bundle.
811 For GnuTLS, only version 3.0.20 or later.
813 JH/06 Verification of the server certificate for a TLS connection is now tried
814 (but not required) by default. The verification status is now logged by
815 default, for both outbound TLS and client-certificate supplying inbound
818 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
821 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
822 Status Notification (bounce) messages are now MIME format per RFC 3464.
823 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
824 under the control of the dsn_advertise_hosts option, and routers may
825 have a dsn_lasthop option.
827 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
828 default, modifiable by a malware= option. The list separator for
829 the options can now be changed in the usual way. Bug 68.
831 JH/10 The smtp_receive_timeout main option is now expanded before use.
833 JH/11 The incoming_interface log option now also enables logging of the
834 local interface on delivery outgoing connections.
836 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
837 if the interface and destination host and port all match.
839 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
842 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
843 Patch from Andrew Lewis.
845 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
846 now supports optional time-restrictions, weighting, and priority
847 modifiers per server. Patch originally by <rommer@active.by>.
849 JH/16 The spamd_address main option now supports a mixed list of local
850 and remote servers. Remote servers can be IPv6 addresses, and
851 specify a port-range.
853 JH/17 Bug 68: The spamd_address main option now supports an optional
854 timeout value per server.
856 JH/18 Bug 1581: Router and transport options headers_add/remove can
857 now have the list separator specified.
859 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
862 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
865 JH/21 Support for the A6 type of dns record is withdrawn.
867 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
868 rather than the verbs used.
870 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
871 from 255 to 1024 chars.
873 JH/24 Verification callouts now attempt to use TLS by default.
875 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
876 are generic router options now. The defaults didn't change.
878 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
879 Original patch from Alexander Shikoff, worked over by JH.
881 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
882 files if ZCAT_COMMAND is not executable.
884 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
886 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
888 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
889 Normally benign, it bites when the pair was led to by a CNAME;
890 modern usage is to not canonicalize the domain to a CNAME target
891 (and we were inconsistent anyway for A-only vs AAAA+A).
893 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
895 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
896 when evaluating $sender_host_dnssec.
898 JH/31 Check the HELO verification lookup for DNSSEC, adding new
899 $sender_helo_dnssec variable.
901 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
903 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
905 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
907 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
908 documented as working, but never had. Support all but $spam_report.
910 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
911 added for tls authenticator.
913 HS/03 Add perl_taintmode main config option
919 TL/01 When running the test suite, the README says that variables such as
920 no_msglog_check are global and can be placed anywhere in a specific
921 test's script, however it was observed that placement needed to be near
922 the beginning for it to behave that way. Changed the runtest perl
923 script to read through the entire script once to detect and set these
924 variables, reset to the beginning of the script, and then run through
925 the script parsing/test process like normal.
927 TL/02 The BSD's have an arc4random API. One of the functions to induce
928 adding randomness was arc4random_stir(), but it has been removed in
929 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
930 function when detected.
932 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
933 cause callback expansion.
935 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
936 syntax errors in an expansion can be treated as a string instead of
937 logging or causing an error, due to the internal use of bool_lax
938 instead of bool when processing it.
940 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
941 server certificates when making smtp deliveries.
943 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
945 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
947 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
949 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
950 Merged patch from Sebastian Wiedenroth.
952 JH/05 Fix results-pipe from transport process. Several recipients, combined
953 with certificate use, exposed issues where response data items split
954 over buffer boundaries were not parsed properly. This eventually
955 resulted in duplicates being sent. This issue only became common enough
956 to notice due to the introduction of connection certificate information,
957 the item size being so much larger. Found and fixed by Wolfgang Breyha.
959 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
960 size buffer was used, resulting in syntax errors when an expansion
963 JH/07 Add support for directories of certificates when compiled with a GnuTLS
964 version 3.3.6 or later.
966 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
967 is EXPERIMENTAL_EVENT, the main-configuration and transport options
968 both become "event_action", the variables become $event_name, $event_data
969 and $event_defer_errno. There is a new variable $verify_mode, usable in
970 routers, transports and related events. The tls:cert event is now also
971 raised for inbound connections, if the main configuration event_action
974 TL/06 In test suite, disable OCSP for old versions of openssl which contained
975 early OCSP support, but no stapling (appears to be less than 1.0.0).
977 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
978 server certificate names available under the smtp transport option
979 "tls_verify_cert_hostname" now do not permit multi-component wildcard
982 JH/10 Time-related extraction expansions from certificates now use the main
983 option "timezone" setting for output formatting, and are consistent
984 between OpenSSL and GnuTLS compilations. Bug 1541.
986 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
987 encoded parameter in the incoming message. Bug 1558.
989 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
990 include certificate info, eximon was claiming there were spoolfile
993 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
995 JH/14 Log delivery-related information more consistently, using the sequence
996 "H=<name> [<ip>]" wherever possible.
998 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
999 are problematic for Debian distribution, omit them from the release
1002 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1004 JH/16 Fix string representation of time values on 64bit time_t architectures.
1007 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1008 output list separator was used.
1013 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1014 checkers that were complaining about end of non-void function with no
1017 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1018 This was a regression introduced in 4.83 by another bugfix.
1020 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1022 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1023 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1029 TF/01 Correctly close the server side of TLS when forking for delivery.
1031 When a message was received over SMTP with TLS, Exim failed to clear up
1032 the incoming connection properly after forking off the child process to
1033 deliver the message. In some situations the subsequent outgoing
1034 delivery connection happened to have the same fd number as the incoming
1035 connection previously had. Exim would try to use TLS and fail, logging
1036 a "Bad file descriptor" error.
1038 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1039 utilities have not been installed.
1041 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1042 temporary space as the ACL may create new global variables.
1044 TL/01 LDAP support uses per connection or global context settings, depending
1045 upon the detected version of the libraries at build time.
1047 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1048 to extract and use the src ip:port in logging and expansions as if it
1049 were a direct connection from the outside internet. PPv2 support was
1050 updated based on HAProxy spec change in May 2014.
1052 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1054 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1055 Properly escape header and check for NULL return.
1057 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1060 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1062 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1063 characters in header names, implemented as a verify condition.
1064 Contributed by Michael Fischer v. Mollard.
1066 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1067 results permerror and temperror. Previous values are deprecated but
1068 still accepted. In a future release, err_perm and err_temp will be
1069 completely removed, which will be a backward incompatibility if the
1070 ACL tests for either of these two old results. Patch contributed by
1071 user bes-internal on the mailing list.
1073 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1075 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1076 selectors, in both main and reject logs.
1078 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1081 JH/07 Add malware type "sock" for talking to simple daemon.
1083 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1085 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1086 routers/transports under cutthrough routing.
1088 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1089 numbers. Touch up "bool" conditional to keep the same definition.
1091 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1093 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1095 JH/12 Expand items in router/transport headers_add or headers_remove lists
1096 individually rather than the list as a whole. Bug 1452.
1098 Required for reasonable handling of multiple headers_ options when
1099 they may be empty; requires that headers_remove items with embedded
1100 colons must have them doubled (or the list-separator changed).
1102 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1103 view the policy declared in the DMARC record. Currently, $dmarc_status
1104 is a combined value of both the record presence and the result of the
1107 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1109 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1110 dnslookup router and the smtp transport (applying to the forward
1113 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1114 of ldap servers used for a specific lookup. Patch provided by Heiko
1117 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1118 New variable $lookup_dnssec_authenticated for observability.
1120 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1121 Patch submitted by Lars Timman.
1123 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1125 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1126 Requires trusted mode and valid format message id, aborts otherwise.
1127 Patch contributed by Heiko Schlichting.
1129 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1130 certextract with support for various fields. Bug 1358.
1132 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1133 is requested by default, modifiable by smtp transport option
1136 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1137 operate on certificate variables to give certificate fingerprints
1138 Also new ${sha256:cert_variable}.
1140 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1142 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1143 Christian Aistleitner.
1145 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1147 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1148 file. Patch from Wolfgang Breyha.
1150 JH/25 Expand the coverage of the delivery $host and $host_address to
1151 client authenticators run in verify callout. Bug 1476.
1153 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1154 align with daemon_smtp_ports. Bug 72.
1156 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1157 support and error reporting did not work properly.
1159 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1160 and is readable. Patch from Andrew Colin Kissa.
1162 TL/14 Enhance documentation of ${run expansion and how it parses the
1163 commandline after expansion, particularly in the case when an
1164 unquoted variable expansion results in an empty value.
1166 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1168 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1169 by renaming away from it. Addresses GH issue 15, reported by
1172 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1174 TL/15 SECURITY: prevent double expansion in math comparison functions
1175 (can expand unsanitized data). Not remotely exploitable.
1182 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1184 PP/02 Make -n do something, by making it not do something.
1185 When combined with -bP, the name of an option is not output.
1187 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1190 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1191 $sender_host_name and config options to manage this, and basic check
1194 PP/05 DSCP support for outbound connections and control modifier for inbound.
1196 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1197 (Only plugin which currently uses this is kerberos4, which nobody should
1198 be using, but we should make it available and other future plugins might
1199 conceivably use it, even though it would break NAT; stuff *should* be
1200 using channel bindings instead).
1202 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1203 name; added for Sendmail compatibility; requires admin caller.
1204 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1205 just ignore it); requires trusted caller.
1206 Also parse but ignore: -Ac -Am -X<logfile>
1209 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1211 TL/02 Add +smtp_confirmation as a default logging option.
1213 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1214 Patch by Magnus Holmgren from 2007-02-20.
1216 TL/04 Bugzilla 1281 - Spec typo.
1217 Bugzilla 1283 - Spec typo.
1218 Bugzilla 1290 - Spec grammar fixes.
1220 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1222 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1224 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1225 mailing list by by Dmitry Isaikin.
1227 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1229 JH/02 Support "G" suffix to numbers in ${if comparisons.
1231 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1233 NM/01 Bugzilla 1197 - Spec typo
1234 Bugzilla 1196 - Spec examples corrections
1236 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1238 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1239 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1242 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1243 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1245 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1246 "acl {{name}{arg}...}", and optional args on acl condition
1249 JH/05 Permit multiple router/transport headers_add/remove lines.
1251 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1253 JH/07 Avoid using a waiting database for a single-message-only transport.
1254 Performance patch from Paul Fisher. Bugzilla 1262.
1256 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1259 JH/09 Add $headers_added variable, with content from use of ACL modifier
1260 add_header (but not yet added to the message). Bugzilla 199.
1262 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1263 Pulled from Bugzilla 817 by Wolfgang Breyha.
1265 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1267 (nb: this is the same fix as in Exim 4.80.1)
1269 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1272 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1275 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1277 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1278 advertise SMTP AUTH mechanism to us, instead of a generic
1279 protocol violation error. Also, make Exim more robust to bad
1280 data from the Dovecot auth socket.
1282 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1284 When a queue runner is handling a message, Exim first routes the
1285 recipient addresses, during which it prunes them based on the retry
1286 hints database. After that it attempts to deliver the message to
1287 any remaining recipients. It then updates the hints database using
1290 So if a recipient address works intermittently, it can get repeatedly
1291 deferred at routing time. The retry hints record remains fresh so the
1292 address never reaches the final cutoff time.
1294 This is a fairly common occurrence when a user is bumping up against
1295 their storage quota. Exim had some logic in its local delivery code
1296 to deal with this. However it did not apply to per-recipient defers
1297 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1299 This change adds a proper retry rule check during routing so that the
1300 final cutoff time is checked against the message's age. We only do
1301 this check if there is an address retry record and there is not a
1302 domain retry record; this implies that previous attempts to handle
1303 the address had the retry_use_local_parts option turned on. We use
1304 this as an approximation for the destination being like a local
1305 delivery, as in LMTP.
1307 I suspect this new check makes the old local delivery cutoff check
1308 redundant, but I have not verified this so I left the code in place.
1310 TF/02 Correct gecos expansion when From: is a prefix of the username.
1312 Test 0254 submits a message to Exim with the header
1316 When I ran the test suite under the user fanf2, Exim expanded
1317 the header to contain my full name, whereas it should have added
1318 a Resent-Sender: header. It erroneously treats any prefix of the
1319 username as equal to the username.
1321 This change corrects that bug.
1323 GF/01 DCC debug and logging tidyup
1324 Error conditions log to paniclog rather than rejectlog.
1325 Debug lines prefixed by "DCC: " to remove any ambiguity.
1327 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1329 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1330 Bug spotted by Jeremy Harris; was flawed since initial commit.
1331 Would have resulted in OCSP responses post-SNI triggering an Exim
1332 NULL dereference and crash.
1334 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1336 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1337 Bug detection, analysis and fix by Samuel Thibault.
1338 Bugzilla 1331, Debian bug #698092.
1340 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1342 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1343 Server implementation by Todd Lyons, client by JH.
1344 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1345 config variable "prdr_enable" controls whether the server
1346 advertises the facility. If the client requests PRDR a new
1347 acl_data_smtp_prdr ACL is called once for each recipient, after
1348 the body content is received and before the acl_smtp_data ACL.
1349 The client is controlled by both of: a hosts_try_prdr option
1350 on the smtp transport, and the server advertisement.
1351 Default client logging of deliveries and rejections involving
1352 PRDR are flagged with the string "PRDR".
1354 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1355 fclose(). Diagnosis by Todd Lyons.
1357 PP/17 Update configure.default to handle IPv6 localhost better.
1358 Patch by Alain Williams (plus minor tweaks).
1361 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1362 This is now consistent with GnuTLS, and is now documented: the
1363 previous undocumented portable approach to treating the option as
1364 unset was to force an expansion failure. That still works, and
1365 an empty string is now equivalent.
1367 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1368 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1369 not performing validation itself.
1371 PP/20 Added force_command boolean option to pipe transport.
1372 Patch from Nick Koston, of cPanel Inc.
1374 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1377 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1379 PP/21 Fix eximon continuous updating with timestamped log-files.
1380 Broken in a format-string cleanup in 4.80, missed when I repaired the
1381 other false fix of the same issue.
1382 Report and fix from Heiko Schlichting.
1385 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1386 Report from Prashanth Katuri.
1388 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1389 It's SecureTransport, so affects any MacOS clients which use the
1390 system-integrated TLS libraries, including email clients.
1392 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1393 using a MIME ACL for non-SMTP local injection.
1394 Report and assistance in diagnosis by Warren Baker.
1396 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1398 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1400 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1401 last id that failed so it may be referenced in subsequent ACL's.
1403 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1406 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1407 ldap library initialization, allowing self-signed CA's to be
1408 used. Also properly sets require_cert option later in code by
1409 using NULL (global ldap config) instead of ldap handle (per
1410 session). Bug diagnosis and testing by alxgomz.
1412 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1413 the src/util/ subdirectory.
1415 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1416 renamed to Transport Post Delivery Action by Jeremy Harris, as
1419 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1420 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1421 redis_servers = needs to be configured which will be used by the redis
1422 lookup. Patch from Warren Baker, of The Packet Hub.
1424 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1426 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1427 hostname or reverse DNS when processing a host list. Used suggestions
1428 from multiple comments on this bug.
1430 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1432 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1433 Missed a few lines, added it to make the runtest require no keyboard
1436 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1437 contains upper case chars. Make router use caseful_local_part.
1439 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1440 support when GnuTLS has been built with p11-kit.
1446 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1448 This, or similar/improved, will also be change PP/11 of 4.82.
1454 PP/01 Handle short writes when writing local log-files.
1455 In practice, only affects FreeBSD (8 onwards).
1456 Bugzilla 1053, with thanks to Dmitry Isaikin.
1458 NM/01 Bugzilla 949 - Documentation tweak
1460 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1463 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1465 PP/02 Implemented gsasl authenticator.
1467 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1469 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1470 `pkg-config foo` for cflags/libs.
1472 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1473 with rest of GSASL and with heimdal_gssapi.
1475 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1476 `pkg-config foo` for cflags/libs for the TLS implementation.
1478 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1479 properties get this fed in as external SSF. A number of robustness
1480 and debugging improvements to the cyrus_sasl authenticator.
1482 PP/08 cyrus_sasl server now expands the server_realm option.
1484 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1485 Patch by Jeremy Harris.
1487 PP/10 Added dbmjz lookup type.
1489 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1491 PP/12 MAIL args handles TAB as well as SP, for better interop with
1492 non-compliant senders.
1493 Analysis and variant patch by Todd Lyons.
1495 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1496 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1497 Patch from Dirk Mueller <dmueller@suse.com>
1499 PP/13 tls_peerdn now print-escaped for spool files.
1500 Observed some $tls_peerdn in wild which contained \n, which resulted
1501 in spool file corruption.
1503 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1504 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1505 or write after TLS renegotiation, which otherwise led to messages
1508 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1509 as a tracking header (ie: a signed header comes before the signature).
1510 Patch from Wolfgang Breyha.
1512 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1513 comma-sep list; embedded commas doubled.
1515 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1517 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1519 Report and patch from Dmitry Banschikov.
1521 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1522 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1523 failures. We appear to now support TLS1.1+ with Exim.
1525 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1526 lets Exim select keys and certificates based upon TLS SNI from client.
1527 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1528 before an outbound SMTP session. New log_selector, +tls_sni.
1530 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1531 NULL dereference. Report and patch from Alun Jones.
1533 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1534 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1535 Not seeing resolver debug output on NetBSD, but suspect this is a
1536 resolver implementation change.
1538 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1539 Left warnings. Added "eximon gdb" invocation mode.
1541 PP/21 Defaulting "accept_8bitmime" to true, not false.
1543 PP/22 Added -bw for inetd wait mode support.
1545 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1546 locate the relevant includes and libraries. Made this the default.
1548 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1549 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1551 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1552 This may cause build issues on older platforms.
1554 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1555 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1556 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1557 Added SNI support via GnuTLS too.
1558 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1560 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1562 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1563 Applied second patch from Janne, implementing suggestion to default
1564 multiple-strings-in-record handling to match SPF spec.
1566 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1568 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1569 read-only, out of scope).
1570 Patch from Wolfgang Breyha, report from Stuart Northfield.
1572 PP/29 Fix three issues highlighted by clang analyser static analysis.
1573 Only crash-plausible issue would require the Cambridge-specific
1574 iplookup router and a misconfiguration.
1575 Report from Marcin Mirosław.
1577 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1579 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1580 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1581 As part of this, removing so much warning spew let me fix some minor
1582 real issues in debug logging.
1584 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1585 assignment on my part. Fixed.
1587 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1588 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1589 Janne Snabb (who went above and beyond: thank you).
1591 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1592 string otherwise requires a connection and a bunch more work and it's
1593 relatively easy to get wrong. Should also expose TLS library linkage
1596 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1597 64-bit ${eval} (JH/03).
1599 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1600 GNU libc to support some of the 64-bit stuff, should not lead to
1601 conflicts. Defined before os.h is pulled in, so if a given platform
1602 needs to override this, it can.
1604 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1605 protection layer was required, which is not implemented.
1606 Bugzilla 1254, patch from Wolfgang Breyha.
1608 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1609 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1610 tls_dhparam take prime identifiers. Also unbreak combination of
1611 OpenSSL+DH_params+TLSSNI.
1613 PP/39 Disable SSLv2 by default in OpenSSL support.
1619 PP/01 Solaris build fix for Oracle's LDAP libraries.
1620 Bugzilla 1109, patch from Stephen Usher.
1622 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1624 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1627 TF/02 Fix a couple more cases where we did not log the error message
1628 when unlink() failed. See also change 4.74-TF/03.
1630 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1631 lock up or crash if it happened to be inside a call to libc when it
1632 got a SIGUSR1 from exiwhat.
1634 The SIGUSR1 handler appends the current process status to the process
1635 log which is later printed by exiwhat. It used to use the general
1636 purpose logging code to do this, but several functions it calls are
1637 not safe for signals.
1639 The new output code in the SIGUSR1 handler is specific to the process
1640 log, and simple enough that it's easy to inspect for signal safety.
1641 Removing some special cases also simplifies the general logging code.
1642 Removing the spurious timestamps from the process log simplifies
1645 TF/04 Improved ratelimit ACL condition.
1647 The /noupdate option has been deprecated in favour of /readonly which
1648 has clearer semantics. The /leaky, /strict, and /readonly update modes
1649 are mutually exclusive. The update mode is no longer included in the
1650 database key; it just determines when the database is updated. (This
1651 means that when you upgrade Exim will forget old rate measurements.)
1653 Exim now checks that the per_* options are used with an update mode that
1654 makes sense for the current ACL. For example, when Exim is processing a
1655 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1656 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1657 must specify per_mail/readonly. If you omit the update mode it defaults to
1658 /leaky where that makes sense (as before) or /readonly where required.
1660 The /noupdate option is now undocumented but still supported for
1661 backwards compatibility. It is equivalent to /readonly except that in
1662 ACLs where /readonly is required you may specify /leaky/noupdate or
1663 /strict/noupdate which are treated the same as /readonly.
1665 A useful new feature is the /count= option. This is a generalization
1666 of the per_byte option, so that you can measure the throughput of other
1667 aggregate values. For example, the per_byte option is now equivalent
1668 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1670 The per_rcpt option has been generalized using the /count= mechanism
1671 (though it's more complicated than the per_byte equivalence). When it is
1672 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1673 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1674 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1675 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1676 Note that using per_rcpt with a non-readonly update mode in more than
1677 one ACL will cause the recipients to be double-counted. (The per_mail
1678 and per_byte options don't have this problem.)
1680 The handling of very low rates has changed slightly. If the computed rate
1681 is less than the event's count (usually one) then this event is the first
1682 after a long gap. In this case the rate is set to the same as this event's
1683 count, so that the first message of a spam run is counted properly.
1685 The major new feature is a mechanism for counting the rate of unique
1686 events. The new per_addr option counts the number of different
1687 recipients that someone has sent messages to in the last time period. It
1688 behaves like per_rcpt if all the recipient addresses are different, but
1689 duplicate recipient addresses do not increase the measured rate. Like
1690 the /count= option this is a general mechanism, so the per_addr option
1691 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1692 example, measure the rate that a client uses different sender addresses
1693 with the options per_mail/unique=$sender_address. There are further
1694 details in the main documentation.
1696 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1698 TF/06 Removed a few PCRE remnants.
1700 TF/07 Automatically extract Exim's version number from tags in the git
1701 repository when doing development or release builds.
1703 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1704 Bugzilla 879. Patch from Paul Fisher.
1706 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1707 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1710 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1712 PP/05 Variable $av_failed, true if the AV scanner deferred.
1713 Bugzilla 1078. Patch from John Horne.
1715 PP/06 Stop make process more reliably on build failure.
1716 Bugzilla 1087. Patch from Heiko Schlittermann.
1718 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1719 Bugzilla 1089. Patch from Heiko Schlittermann.
1721 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1722 Bugzilla 1131. Patch from Holger Weiß.
1724 PP/09 Handle IPv6 addresses with SPF.
1725 Bugzilla 860. Patch from Wolfgang Breyha.
1727 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1729 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1732 PP/11 match_* no longer expand right-hand-side by default.
1733 New compile-time build option, EXPAND_LISTMATCH_RHS.
1734 New expansion conditions, "inlist", "inlisti".
1736 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1738 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1740 PP/14 fix log_write() format string regression from TF/03.
1741 Bugzilla 1152. Patch from Dmitry Isaikin.
1747 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1749 PP/02 Harmonised TLS library version reporting; only show if debugging.
1750 Layout now matches that introduced for other libraries in 4.74 PP/03.
1752 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1754 PP/04 New "dns_use_edns0" global option.
1756 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1759 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1760 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1762 TK/01 Updated PolarSSL code to 0.14.2.
1763 Bugzilla 1097. Patch from Andreas Metzler.
1765 PP/07 Catch divide-by-zero in ${eval:...}.
1766 Fixes bugzilla 1102.
1768 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1771 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1772 format-string attack -- SECURITY: remote arbitrary code execution.
1774 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1775 time unintentionally subject to list matching rules, letting the header
1776 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1777 arbitrary string expansion). This allowed for information disclosure.
1779 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1780 INT_MIN/-1 -- value coerced to INT_MAX.
1786 NM/01 Workaround for PCRE version dependency in version reporting
1789 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1790 This fixes portability to compilers other than gcc, notably
1791 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1793 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1794 makefiles for portability to HP-UX and POSIX correctness.
1796 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1797 Also via indented variable definition in the Makefile.
1798 (Debugging by Oliver Heesakkers).
1800 PP/02 Restore caching of spamd results with expanded spamd_address.
1801 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1803 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1804 Improves build reliability. Fix from: Frank Elsner
1806 NM/02 Fix wide character breakage in the rfc2047 coding
1807 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1809 NM/03 Allow underscore in dnslist lookups
1810 Fixes bug 1026. Patch from Graeme Fowler
1812 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1813 Code patches from Adam Ciarcinski of NetBSD.
1815 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1818 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1819 is logged, to avoid truncation. Patch from John Horne.
1821 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1822 Patch from Jakob Hirsch.
1824 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1825 SQL string expansion failure details.
1826 Patch from Andrey Oktyabrski.
1828 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1829 Patch from Simon Arlott.
1831 PP/09 New lookups functionality failed to compile on old gcc which rejects
1832 extern declarations in function scope.
1833 Patch from Oliver Fleischmann
1835 PP/10 Use sig_atomic_t for flags set from signal handlers.
1836 Check getgroups() return and improve debugging.
1837 Fixed developed for diagnosis in bug 927 (which turned out to be
1840 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1841 Patch from Mark Zealey.
1843 PP/12 Bugzilla 1056: Improved spamd server selection.
1844 Patch from Mark Zealey.
1846 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1847 Based on patch from Heiko Schlittermann.
1849 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1850 Patch from Uwe Doering, sign-off by Michael Haardt.
1852 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1853 variable declaration deep within a block. Bug and patch from
1856 PP/15 lookups-Makefile IRIX compatibility coercion.
1858 PP/16 Make DISABLE_DKIM build knob functional.
1860 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1861 Patch by Simon Arlott
1863 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1864 variable argument macros. Our copy now differs from upstream.
1870 TF/01 Failure to get a lock on a hints database can have serious
1871 consequences so log it to the panic log.
1873 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1874 controlled using the smtp_confirmation log selector.
1876 TF/03 Include the error message when we fail to unlink a spool file.
1878 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1879 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1880 for maintaining out-of-tree patches for some time.
1882 PP/01 Bugzilla 139: Documentation and portability issues.
1883 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1884 Handle per-OS dynamic-module compilation flags.
1886 PP/02 Let /dev/null have normal permissions.
1887 The 4.73 fixes were a little too stringent and complained about the
1888 permissions on /dev/null. Exempt it from some checks.
1889 Reported by Andreas M. Kirchwitz.
1891 PP/03 Report version information for many libraries, including
1892 Exim version information for dynamically loaded libraries. Created
1893 version.h, now support a version extension string for distributors
1894 who patch heavily. Dynamic module ABI change.
1896 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1897 privilege escalation vulnerability whereby the Exim run-time user
1898 can cause root to append content of the attacker's choosing to
1901 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1904 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1905 If dropping privileges for untrusted macros, we disabled normal logging
1906 on the basis that it would fail; for the Exim run-time user, this is not
1907 the case, and it resulted in successful deliveries going unlogged.
1908 Fixed. Reported by Andreas Metzler.
1914 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1915 only prepend for the Resent-* case. Fixes regression introduced in
1916 Exim 4.70 by NM/22 for Bugzilla 607.
1918 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1919 increasing numbers of administrators be bitten by this.
1921 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1923 PP/03 Bugzilla 994: added openssl_options main configuration option.
1925 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1927 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1929 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1931 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1932 without a peer certificate, leading to a segfault because of an
1933 assumption that peers always have certificates. Be a little more
1934 paranoid. Problem reported by Martin Tscholak.
1936 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1937 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1938 NB: ClamAV planning to remove STREAM in "middle of 2010".
1939 CL also introduces -bmalware, various -d+acl logging additions and
1940 more caution in buffer sizes.
1942 PP/09 Implemented reverse_ip expansion operator.
1944 PP/10 Bugzilla 937: provide a "debug" ACL control.
1946 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1948 PP/12 Bugzilla 973: Implement --version.
1950 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1952 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1954 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1956 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1957 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1958 ignore trailing whitespace.
1960 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1962 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1965 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1966 Notification from Dr Andrew Aitchison.
1968 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1969 ExtendedDetectionInfo response format.
1970 Notification from John Horne.
1972 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1975 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1976 XSL and documented dependency on system catalogs, with examples of how
1979 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1982 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1983 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1984 configuration file which is writeable by the Exim user or group.
1986 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1987 of configuration files to cover files specified with the -C option if
1988 they are going to be used with root privileges, not just the default
1991 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1992 option (effectively making it always true).
1994 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1995 files to be used while preserving root privileges.
1997 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1998 that rogue child processes cannot use them.
2000 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2001 run-time user, instead of root.
2003 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2004 Exim run-time user without dropping privileges.
2006 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2007 result string, instead of calling string_vformat() twice with the same
2010 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2011 for other users. Others should always drop root privileges if they use
2012 -C on the command line, even for a whitelisted configure file.
2014 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2016 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2022 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2023 $data_path, and $header_path variables; fixed documentation bugs and
2026 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2027 exipick to access non-standard spools, including the "frozen" queue
2030 NM/01 Bugzilla 965: Support mysql stored procedures.
2031 Patch from Alain Williams
2033 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2035 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2036 Patch from Andreas Metzler
2038 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2039 Patch from Kirill Miazine
2041 NM/05 Bugzilla 671: Added umask to procmail example.
2043 JJ/03 installed exipick 20100323.0, fixing doc bug
2045 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2046 directory. Notification and patch from Dan Rosenberg.
2048 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2050 TK/02 Improve log output when DKIM signing operation fails.
2052 MH/01 Treat the transport option dkim_domain as a colon separated
2053 list, not as a single string, and sign the message with each element,
2054 omitting multiple occurences of the same signer.
2056 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2057 Bugzilla 985, 986. Patch by Simon Arlott
2059 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2060 Patch by Simon Arlott
2062 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2063 MBX locking. Notification from Dan Rosenberg.
2069 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2071 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2073 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2075 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2077 NM/04 Bugzilla 915: Flag broken perl installation during build.
2083 TK/01 Added patch by Johannes Berg that expands the main option
2084 "spamd_address" if it starts with a dollar sign.
2086 TK/02 Write list of recipients to X-Envelope-Sender header when building
2087 the mbox-format spool file for content scanning (suggested by Jakob
2090 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2091 (http://www.dcc-servers.net/) support via dccifd. Activated by
2092 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2094 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2095 by Mark Daniel Reidel <mr@df.eu>.
2097 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2098 When building exim an external PCRE library is now needed -
2099 PCRE is a system library on the majority of modern systems.
2100 See entry on PCRE_LIBS in EDITME file.
2102 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2103 conversation. Added nologin parameter to request.
2104 Patch contributed by Kirill Miazine.
2106 TF/01 Do not log submission mode rewrites if they do not change the address.
2108 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2110 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2111 log files in place. Contributed by Roberto Lima.
2113 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2115 TF/03 Bugzilla 615: When checking the local_parts router precondition
2116 after a local_part_suffix or local_part_prefix option, Exim now
2117 does not use the address's named list lookup cache, since this
2118 contains cached lookups for the whole local part.
2120 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2121 Robert Millan. Documentation is in experimental-spec.txt.
2123 TF/04 Bugzilla 668: Fix parallel build (make -j).
2125 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2127 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2128 Patch provided by Jan Srzednicki.
2130 TF/05 Leading white space used to be stripped from $spam_report which
2131 wrecked the formatting. Now it is preserved.
2133 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2134 that they are available at delivery time.
2136 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2138 TF/08 TLS error reporting now respects the incoming_interface and
2139 incoming_port log selectors.
2141 TF/09 Produce a more useful error message if an SMTP transport's hosts
2142 setting expands to an empty string.
2144 NM/06 Bugzilla 744: EXPN did not work under TLS.
2145 Patch provided by Phil Pennock.
2147 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2148 Patch provided by Richard Godbee.
2150 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2151 acl_smtp_notquit, added index entry.
2153 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2154 Patch provided by Eugene Bujak.
2156 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2157 accept(). Patch provided by Maxim Dounin.
2159 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2160 Patch provided by Phil Pennock.
2162 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2164 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2165 Patch provided by Brad "anomie" Jorsch.
2167 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2168 Patch provided by Dean Brooks.
2170 TK/05 Add native DKIM support (does not depend on external libraries).
2172 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2173 Patch provided by Graeme Fowler.
2175 NM/16 Bugzilla 851: Documentation example syntax fix.
2177 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2179 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2182 NM/19 Bugzilla 745: TLS version reporting.
2183 Patch provided by Phil Pennock.
2185 NM/20 Bugzilla 167: bool: condition support.
2186 Patch provided by Phil Pennock.
2188 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2189 clients. Patch provided by Phil Pennock.
2191 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2192 Patch provided by Brad "anomie" Jorsch.
2194 NM/23 Bugzilla 687: Fix misparses in eximstats.
2195 Patch provided by Heiko Schlittermann.
2197 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2198 Patch provided by Heiko Schlittermann.
2200 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2201 plus update to original patch.
2203 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2205 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2206 Patch provided by David Brownlee.
2208 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2210 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2212 NM/30 Bugzilla 888: TLS documentation bugfixes.
2214 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2216 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2217 Unlike the original bugzilla I have changed all shell scripts in src tree.
2219 NM/33 Bugzilla 898: Transport filter timeout fix.
2220 Patch by Todd Rinaldo.
2222 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2223 Patch by Serge Demonchaux.
2225 NM/35 Bugzilla 39: Base64 decode bug fixes.
2226 Patch by Jakob Hirsch.
2228 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2230 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2232 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2234 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2240 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2241 ALT-N's libdkim that I have put here:
2242 http://duncanthrax.net/exim-experimental/
2244 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2245 were called 'true' and it seems that C99 defines that as a reserved
2246 keyword to be used with 'bool' variable types. That means you could
2247 not include C99-style headers which use bools without triggering
2248 build errors in sieve.c.
2250 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2251 as mailq or other aliases. Changed the --help handling significantly
2252 to do whats expected. exim_usage() emits usage/help information.
2254 SC/01 Added the -bylocaldomain option to eximstats.
2256 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2258 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2260 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2266 PH/01 Another patch from the Sieve maintainer.
2268 PH/02 When an IPv6 address is converted to a string for single-key lookup
2269 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2270 dots are used instead of colons so that keys in lsearch files need not
2271 contain colons. This was done some time before quoting was made available
2272 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2273 (notated using the quote facility) so as to distinguish them from IPv4
2274 keys. This meant that lookups for IP addresses in host lists did not work
2275 for iplsearch lookups.
2277 This has been fixed by arranging for IPv6 addresses to be expressed with
2278 colons if the lookup type is iplsearch. This is not incompatible, because
2279 previously such lookups could never work.
2281 The situation is now rather anomalous, since one *can* have colons in
2282 ordinary lsearch keys. However, making the change in all cases is
2283 incompatible and would probably break a number of configurations.
2285 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2288 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2289 conversion specification without a maximum field width, thereby enabling
2290 a rogue spamd server to cause a buffer overflow. While nobody in their
2291 right mind would setup Exim to query an untrusted spamd server, an
2292 attacker that gains access to a server running spamd could potentially
2293 exploit this vulnerability to run arbitrary code as the Exim user.
2295 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2296 $primary_hostname instead of what libspf2 thinks the hosts name is.
2298 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2299 a directory entry by the name of the lookup key. Previously, if a
2300 symlink pointed to a non-existing file or a file in a directory that
2301 Exim lacked permissions to read, a lookup for a key matching that
2302 symlink would fail. Now it is enough that a matching directory entry
2303 exists, symlink or not. (Bugzilla 503.)
2305 PH/03 The body_linecount and body_zerocount variables are now exported in the
2308 PH/04 Added the $dnslist_matched variable.
2310 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2311 This means they are set thereafter only if the connection becomes
2314 PH/06 Added the client_condition to authenticators so that some can be skipped
2315 by clients under certain conditions.
2317 PH/07 The error message for a badly-placed control=no_multiline_responses left
2318 "_responses" off the end of the name.
2320 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2322 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2323 (without spaces) instead of just copying the configuration text.
2325 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2327 PH/11 Added $max_received_linelength.
2329 PH/12 Added +ignore_defer and +include_defer to host lists.
2331 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2332 way in which PCRE > 7.0 is built.
2334 PH/14 Implemented queue_only_load_latch.
2336 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2337 MAIL command. The effect was to mangle the value on 64-bit systems.
2339 PH/16 Another patch from the Sieve maintainer.
2341 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2343 PH/18 If a system quota error occurred while trying to create the file for
2344 a maildir delivery, the message "Mailbox is full" was not appended to the
2345 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2346 only to a quota excession during the actual writing of the file.
2348 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2349 characters?) which causes problems in log lines. The DN values are now
2350 passed through string_printing() before being added to log lines.
2352 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2353 and InterBase are left for another time.)
2355 PH/21 Added message_body_newlines option.
2357 PH/22 Guard against possible overflow in moan_check_errorcopy().
2359 PH/23 POSIX allows open() to be a macro; guard against that.
2361 PH/24 If the recipient of an error message contained an @ in the local part
2362 (suitably quoted, of course), incorrect values were put in $domain and
2363 $local_part during the evaluation of errors_copy.
2369 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2370 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2373 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2374 issue a MAIL command.
2376 PH/02 In an ACL statement such as
2378 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2380 if a client was not listed at all, or was listed with a value other than
2381 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2382 the condition was not true (as it should be), so access was not denied.
2383 The bug was that the ! inversion was incorrectly passed on to the second
2384 item. This has been fixed.
2386 PH/03 Added additional dnslists conditions == and =& which are different from
2387 = and & when the dns lookup returns more than one IP address.
2389 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2390 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2392 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2393 FSYNC, which compiles an option called disable_fsync that allows for
2394 bypassing fsync(). The documentation is heavily laced with warnings.
2396 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2398 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2399 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2400 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2401 including adding "make clean"; (3) Added -fPIC when compiling the test
2402 dynamically loaded module, to get rid of a warning.
2404 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2405 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2406 = 0s. The bug is otherwise harmless.
2408 PH/07 There was a bug in the dovecot authenticator such that the value of
2409 $auth1 could be overwritten, and so not correctly preserved, after a
2410 successful authentication. This usually meant that the value preserved by
2411 the server_setid option was incorrect.
2413 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2415 PH/09 Installed PCRE release 7.0.
2417 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2418 run for batched SMTP input. It is now run at the start of every message
2419 in the batch. While fixing this I discovered that the process information
2420 (output by running exiwhat) was not always getting set for -bs and -bS
2421 input. This is fixed, and it now also says "batched" for BSMTP.
2423 PH/11 Added control=no_pipelining.
2425 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2426 patch, slightly modified), and move the expansion of helo_data till after
2427 the connection is made in the smtp transport (so it can use these
2430 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2432 PH/14 Added log_selector = +pid.
2434 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2436 PH/16 Add ${if forany and ${if forall.
2438 PH/17 Added dsn_from option to vary the From: line in DSNs.
2440 PH/18 Flush SMTP output before performing a callout, unless control =
2441 no_callout_flush is set.
2443 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2444 was true (the default) a successful delivery failed to delete the retry
2445 item, thus causing premature timeout of the address. The bug is now
2448 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2450 PH/21 Long custom messages for fakedefer and fakereject are now split up
2451 into multiline responses in the same way that messages for "deny" and
2452 other ACL rejections are.
2454 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2455 with slight modification.
2457 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2458 draft, changing the syntax and factoring some duplicate code".
2460 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2461 for deliveries of the second and subsequent messages over the same SMTP
2464 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2465 ${reduce, with only minor "tidies".
2467 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2469 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2470 expansion side effects.
2472 PH/27 When a message times out after an over-quota error from an Exim-imposed
2473 quota, the bounce message says "mailbox is full". This message was not
2474 being given when it was a system quota that was exceeded. It now should
2477 MH/03 Made $recipients available in local_scan(). local_scan() already has
2478 better access to the recipient list through recipients_list[], but
2479 $recipients can be useful in postmaster-provided expansion strings.
2481 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2482 in the case of a MAIL command with additional options following the
2483 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2484 were accidentally chopped off.
2486 PH/29 SMTP synchronization checks are implemented when a command is read -
2487 there is a check that no more input is waiting when there shouldn't be
2488 any. However, for some commands, a delay in an ACL can mean that it is
2489 some time before the response is written. In this time, more input might
2490 arrive, invalidly. So now there are extra checks after an ACL has run for
2491 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2492 pipelining has not been advertised.
2494 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2496 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2497 returned in the SMTP response when smtp_return_error_details was false.
2498 This has been fixed.
2500 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2501 instead of the C I/O that was originally supplied, because problems were
2502 reported on Solaris.
2504 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2505 Exim which did not show up earlier: it was assuming that a call to
2506 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2507 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2508 was a macro that became an assignment, so it seemed to work. This has
2509 changed to a proper function call with a void return, hence the compile
2510 error. Exim's code has been fixed.
2512 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2515 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2517 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2519 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2520 because it turns out to be too big or there is a timeout) there is no
2521 "Completed" line in the log. When some messages of this type were
2522 selected by exigrep, they were listed as "not completed". Others were
2523 picked up by some special patterns. I have improved the selection
2524 criteria to be more general.
2526 PH/38 The host_find_failed option in the manualroute router can now be set
2527 to "ignore", to completely ignore a host whose IP address cannot be
2528 found. If all hosts are ignored, the behaviour is controlled by the new
2529 host_all_ignored option.
2531 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2532 homing or because of multiple MX records with /mx) generated more than
2533 one IP address, and the following item turned out to be the local host,
2534 all the secondary addresses of the first item were incorrectly removed
2535 from the list, along with the local host and any following hosts (which
2536 is what is supposed to happen).
2538 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2539 whoever called Exim into the -H file. In the case of the daemon it was
2540 behaving confusingly. When first started, it used values for whoever
2541 started the daemon, but after a SIGHUP it used the Exim user (because it
2542 calls itself on a restart). I have changed the code so that it now always
2545 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2546 message are rejected with the same error (e.g. no authentication or bad
2547 sender address), and a DATA command is nevertheless sent (as can happen
2548 with PIPELINING or a stupid MUA), the error message that was given to the
2549 RCPT commands is included in the rejection of the DATA command. This is
2550 intended to be helpful for MUAs that show only the final error to their
2553 PH/42 Another patch from the Sieve maintainer.
2555 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2556 Eximstats - Fixed some broken HTML links and added missing column headers
2558 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2559 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2561 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2563 PH/43 Yet another patch from the Sieve maintainer.
2565 PH/44 I found a way to check for a TCP/IP connection going away before sending
2566 the response to the final '.' that terminates a message, but only in the
2567 case where the client has not sent further data following the '.'
2568 (unfortunately, this is allowed). However, in many cases there won't be
2569 any further data because there won't be any more messages to send. A call
2570 to select() can be used: if it shows that the input is "ready", there is
2571 either input waiting, or the socket has been closed. An attempt to read
2572 the next input character can distinguish the two cases. Previously, Exim
2573 would have sent an OK response which the client would never have see.
2574 This could lead to message repetition. This fix should cure that, at
2575 least in a lot of common cases.
2577 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2578 advertised in response to EHLO.
2584 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2585 fixed by 4.65/MH/01 (is this a record?) are fixed:
2587 (i) An empty string was always treated as zero by the numeric comparison
2588 operators. This behaviour has been restored.
2590 (ii) It is documented that the numeric comparison operators always treat
2591 their arguments as decimal numbers. This was broken in that numbers
2592 starting with 0 were being interpreted as octal.
2594 While fixing these problems I realized that there was another issue that
2595 hadn't been noticed. Values of message_size_limit (both the global option
2596 and the transport option) were treated as octal if they started with 0.
2597 The documentation was vague. These values are now always treated as
2598 decimal, and I will make that clear in the documentation.
2604 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2605 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2608 MH/01 Don't check that the operands of numeric comparison operators are
2609 integers when their expansion is in "skipping" mode (fixes bug
2610 introduced by 4.64-PH/07).
2612 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2613 child addresses, Exim now panics and dies. Previously, because the count
2614 is held in a short int, deliveries were likely to be lost. As such a
2615 large number of recipients for a single message is ridiculous
2616 (performance will be very, very poor), I have chosen to impose a limit
2617 rather than extend the field.
2623 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2624 leftover -K file (the existence of which was triggered by #402).
2625 While we were at it, introduced process PID as part of the -K
2626 filename. This should rule out race conditions when creating
2629 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2630 processing considerably. Previous code took too long for large mails,
2631 triggering a timeout which in turn triggers #401.
2633 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2634 in the DK code in transports.c. sendfile() is not really portable,
2635 hence the _LINUX specificness.
2637 TF/01 In the add_headers option to the mail command in an Exim filter,
2638 there was a bug that Exim would claim a syntax error in any
2639 header after the first one which had an odd number of characters
2642 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2643 callout verification, Exim cached a "reject" for the entire domain. This
2644 is correct for most verifications, but it is not correct for a recipient
2645 verification with use_sender or use_postmaster set, because in that case
2646 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2647 case of MAIL FROM:<> rejection from other early rejections (e.g.
2648 rejection of HELO). When verifying a recipient using a non-null MAIL
2649 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2650 Whatever the result of the callout, the value of the domain cache is
2651 left unchanged (for any other kind of callout, getting as far as trying
2652 RCPT means that the domain itself is ok).
2654 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2657 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2658 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2659 session, the daemon ignores ECONNECT errors and logs others; it now
2660 ignores EPIPE as well.
2662 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2663 (quoted-printable decoding).
2665 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2666 later the small subsequent patch to fix an introduced bug.
2668 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2670 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2672 PH/08 An error is now given if message_size_limit is specified negative.
2674 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2675 to be given (somewhat) arbitrary names.
2677 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2680 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2681 miscellaneous code fixes
2683 PH/10 Added the log_reject_target ACL modifier to specify where to log
2686 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2687 hostname. This is wrong, because it relates to the incoming message (and
2688 probably the interface on which it is arriving) and not to the outgoing
2689 callout (which could be using a different interface). This has been
2690 changed to use the value of the helo_data option from the smtp transport
2691 instead - this is what is used when a message is actually being sent. If
2692 there is no remote transport (possible with a router that sets up host
2693 addresses), $smtp_active_hostname is used.
2695 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2696 tweaks were necessary in order to get it to work (see also 21 below):
2697 (a) The code assumed that strncpy() returns a negative number on buffer
2698 overflow, which isn't the case. Replaced with Exim's string_format()
2700 (b) There were several signed/unsigned issues. I just did the minimum
2701 hacking in of casts. There is scope for a larger refactoring.
2702 (c) The code used strcasecmp() which is not a standard C function.
2703 Replaced with Exim's strcmpic() function.
2704 (d) The code set only $1; it now sets $auth1 as well.
2705 (e) A simple test gave the error "authentication client didn't specify
2706 service in request". It would seem that Dovecot has changed its
2707 interface. Fortunately there's a specification; I followed it and
2708 changed what the client sends and it appears to be working now.
2710 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2713 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2714 address A is aliased to B and C, where B exists and C does not. Without
2715 -v the output is "A verified" because verification stops after a
2716 successful redirection if more than one address is generated. However,
2717 with -v the child addresses are also verified. Exim was outputting "A
2718 failed to verify" and then showing the successful verification for C,
2719 with its parentage. It now outputs "B failed to verify", showing B's
2720 parentage before showing the successful verification of C.
2722 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2723 look up a TXT record in a specific list after matching in a combined
2726 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2727 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2728 they consult the DNS. I had assumed they would set it the way they
2729 wanted; and indeed my experiments on Linux seem to show that in some
2730 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2731 To be on the safe side, however, I have now made the interface to
2732 host_find_byname() similar to host_find_bydns(), with an argument
2733 containing the DNS resolver options. The host_find_byname() function now
2734 sets these options at its start, just as host_find_bydns() does. The smtp
2735 transport options dns_qualify_single and dns_search_parents are passed to
2736 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2737 of host_find_byname() use the default settings of RES_DEFNAMES
2738 (qualify_single) but not RES_DNSRCH (search_parents).
2740 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2741 spool_read_header() do less string testing, by means of a preliminary
2742 switch on the second character of optional "-foo" lines. (This is
2743 overdue, caused by the large number of possibilities that now exist.
2744 Originally there were few.) While I was there, I also converted the
2745 str(n)cmp tests so they don't re-test the leading "-" and the first
2746 character, in the hope this might squeeze out yet more improvement.
2748 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2749 flag allowing group syntax was set by the header_syntax check but not
2750 turned off, possible causing trouble later; (2) The flag was not being
2751 set at all for the header_verify test, causing "group"-style headers to
2752 be rejected. I have now set it in this case, and also caused header_
2753 verify to ignore an empty address taken from a group. While doing this, I
2754 came across some other cases where the code for allowing group syntax
2755 while scanning a header line wasn't quite right (mostly, not resetting
2756 the flag correctly in the right place). These bugs could have caused
2757 trouble for malformed header lines. I hope it is now all correct.
2759 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2760 with the "reply" argument non-NULL. The code, however (which originally
2761 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2762 but it didn't always do it. This confused somebody who was copying the
2763 code for some other use. I have removed all the tests.
2765 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2766 feature that was used to support insecure browsers during the U.S. crypto
2767 embargo. It requires special client support, and Exim is probably the
2768 only MTA that supported it -- and would never use it because real RSA is
2769 always available. This code has been removed, because it had the bad
2770 effect of slowing Exim down by computing (never used) parameters for the
2771 RSA_EXPORT functionality.
2773 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2774 authenticator to fail if there's a tab character in the incoming data
2775 (there should never be unless someone is messing about, as it's supposed
2776 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2777 if the connection is using TLS or if the remote IP is the same as the
2778 local IP, and the "valid-client-cert option" if a client certificate has
2781 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2782 authenticators. This can be used for authorization after authentication
2783 succeeds. (In the case of plaintext, it servers for both authentication
2786 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2787 if any retry times were supplied.
2789 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2790 connection, where there is no client IP address to check. In this
2791 situation, the verify now always succeeds.
2793 PH/25 Applied John Jetmore's -Mset patch.
2795 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2797 PH/27 In a string expansion for a processed (not raw) header when multiple
2798 headers of the same name were present, leading whitespace was being
2799 removed from all of them, but trailing whitespace was being removed only
2800 from the last one. Now trailing whitespace is removed from each header
2801 before concatenation. Completely empty headers in a concatenation (as
2802 before) are ignored.
2804 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2805 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2807 PH/29 [Removed. This was a change that I later backed out, and forgot to
2808 correct the ChangeLog entry (that I had efficiently created) before
2809 committing the later change.]
2811 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2812 address errors (4xx response to RCPT) over the same connection as other
2813 messages routed to the same hosts. Such deliveries are always "forced",
2814 so retry times are not inspected. This resulted in far too many retries
2815 for the affected addresses. The effect occurred only when there were more
2816 hosts than the hosts_max_try setting in the smtp transport when it had
2817 the 4xx errors. Those hosts that it had tried were not added to the list
2818 of hosts for which the message was waiting, so if all were tried, there
2819 was no problem. Two fixes have been applied:
2821 (i) If there are any address or message errors in an SMTP delivery, none
2822 of the hosts (tried or untried) are now added to the list of hosts
2823 for which the message is waiting, so the message should not be a
2824 candidate for sending over the same connection that was used for a
2825 successful delivery of some other message. This seems entirely
2826 reasonable: after all the message is NOT "waiting for some host".
2827 This is so "obvious" that I'm not sure why it wasn't done
2828 previously. Hope I haven't missed anything, but it can't do any
2829 harm, as the worst effect is to miss an optimization.
2831 (ii) If, despite (i), such a delivery is accidentally attempted, the
2832 routing retry time is respected, so at least it doesn't keep
2833 hammering the server.
2835 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2836 in ${readsocket because some servers need this prod.
2838 PH/32 Added some extra debug output when updating a wait-xxx database.
2840 PH/33 The hint "could be header name not terminated by colon", which has been
2841 given for certain expansion errors for a long time, was not being given
2842 for the ${if def:h_colon_omitted{... case.
2844 PH/34 The spec says: "With one important exception, whenever a domain list is
2845 being scanned, $domain contains the subject domain." There was at least
2846 one case where this was not true.
2848 PH/35 The error "getsockname() failed: connection reset by peer" was being
2849 written to the panic log as well as the main log, but it isn't really
2850 panic-worthy as it just means the connection died rather early on. I have
2851 removed the panic log writing for the ECONNRESET error when getsockname()
2854 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2855 runs only) independently of the message's sender address. This meant
2856 that, if the 4xx error was in fact related to the sender, a different
2857 message to the same recipient with a different sender could confuse
2858 things. In particular, this can happen when sending to a greylisting
2859 server, but other circumstances could also provoke similar problems.
2860 I have changed the default so that the retry time for these errors is now
2861 based a combination of the sender and recipient addresses. This change
2862 can be overridden by setting address_retry_include_sender=false in the
2865 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2866 remote server are returned as part of bounce messages. This was not
2867 happening for LMTP over a pipe (the lmtp transport), but now it is the
2868 same for both kinds of LMTP.
2870 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2871 in header lines that were in fact CNAMEs. This is no longer the case.
2873 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2874 and queue runs started by the daemon processed all messages. This has
2875 been fixed so that -R and -S can now usefully be given with -q<time>.
2877 PH/40 Import PCRE release 6.7 (fixes some bugs).
2879 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2881 PH/42 Give an error if -q is specified more than once.
2883 PH/43 Renamed the variables $interface_address and $interface_port as
2884 $received_ip_address and $received_port, to make it clear that these
2885 values apply to message reception, and not to the outgoing interface when
2886 a message is delivered. (The old names remain recognized, of course.)
2888 PH/44 There was no timeout on the connect() call when using a Unix domain
2889 socket in the ${readsocket expansion. There now is.
2891 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2892 be meaningful with "accept".
2894 SC/01 Eximstats V1.43
2895 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2897 SC/02 Eximstats V1.44
2898 Use a glob alias rather than an array ref in the generated
2899 parser. This improves both readability and performance.
2901 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2902 Collect SpamAssassin and rejection statistics.
2903 Don't display local sender or destination tables unless
2904 there is data to show.
2905 Added average volumes into the top table text output.
2907 SC/04 Eximstats V1.46
2908 Collect data on the number of addresses (recipients)
2909 as well as the number of messages.
2911 SC/05 Eximstats V1.47
2912 Added 'Message too big' to the list of mail rejection
2913 reasons (thanks to Marco Gaiarin).
2915 SC/06 Eximstats V1.48
2916 Mainlog lines which have GMT offsets and are too short to
2917 have a flag are now skipped.
2919 SC/07 Eximstats V1.49 (Alain Williams)
2920 Added the -emptyok flag.
2922 SC/08 Eximstats V1.50
2923 Fixes for obtaining the IP address from reject messages.
2925 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2926 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2927 whitespace changes from 4.64-PH/27
2929 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2932 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2933 are found, allow negative numbers in numeric criteria)
2935 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2937 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2938 to match changes made in 4.64-PH/43
2940 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2942 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2943 "domain" is actually an IP address, and force a failure. This locks out
2944 those revolvers/nameservers that support "A-for-A" lookups, in
2945 contravention of the specifications.
2947 PH/48 When a host name was looked up from an IP address, and the subsequent
2948 forward lookup of the name timed out, the host name was left in
2949 $sender_host_name, contrary to the specification.
2951 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2952 restricted to single-key lookups, Exim was not diagnosing an error if
2953 * or *@ was used with a query-style lookup.
2955 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2957 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2958 long ago, but noone interested enough thought of it. Let's just say that
2959 the "1.1" means that there are some new functions that weren't there at
2960 some point in the past.
2962 PH/51 Error processing for expansion failure of helo_data from an smtp
2963 transport during callout processing was broken.
2965 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2966 tested/used via the -bh/-bhc/-bs options.
2968 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2969 bug, fixed in subsequent PCRE releases).
2971 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2972 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2974 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2980 SC/01 Use a glob alias rather than an array ref in eximstats generated
2981 parser. This improves both readability and performance.
2983 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2984 Don't display local sender or destination tables in eximstats unless
2985 there is data to show.
2986 Added average volumes into the eximstats top table text output.
2988 SC/03 Collect data on the number of addresses (recipients) as well
2989 as the number of messages in eximstats.
2991 TF/01 Correct an error in the documentation for the redirect router. Exim
2992 does (usually) call initgroups() when daemonizing.
2994 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2995 with consistent privilege compared to when running as a daemon.
2997 TF/03 Note in the spec that $authenticated_id is not set for local
2998 submissions from trusted users.
3000 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3001 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3003 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3004 by adding some example configuration directives to the default
3005 configuration file. A little bit of work is required to uncomment the
3006 directives and define how usernames and passwords are checked, but
3007 there is now a framework to start from.
3009 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3010 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3011 without this. I don't know how relevant this is to other LDAP libraries.
3013 PH/02 Add the verb name to the "unknown ACL verb" error.
3015 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3017 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3019 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3020 directory not expanded when it should be if an expanded home directory
3021 was set for the address (which is overridden by the transport).
3023 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3026 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3027 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3028 because it is too late at that time, and has no effect.
3030 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3031 security issue with \' (bugzilla #107). I could not use the
3032 PQescapeStringConn() function, because it needs a PGconn value as one of
3035 PH/08 When testing addresses using -bt, indicate those final addresses that
3036 are duplicates that would not cause an additional delivery. At least one
3037 person was confused, thinking that -bt output corresponded to deliveries.
3038 (Suppressing duplicates isn't a good idea as you lose the information
3039 about possibly different redirections that led to the duplicates.)
3041 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3042 systems where poll() doesn't work, in particular OS X.
3044 PH/10 Added more information to debugging output for retry time not reached.
3046 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3047 operations in malware.c.
3049 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3052 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3053 syslog_duplication set false, log lines that would normally be written
3054 both the the main log and to the reject log were not written to syslog at
3057 PH/14 In the default configuration, change the use of "message" in ACL warn
3058 statements to "add_header".
3060 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3061 not followed by a command (e.g. "seen endif").
3063 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3064 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3067 PH/17 Added extra conditions to the default value of delay_warning_condition
3071 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3072 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3073 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3076 The Auto-Submitted: and various List- headers are standardised, whereas I
3077 don't think Precedence: ever was.
3079 PH/18 Refactored debugging code in route_finduser() to show more information,
3080 in particular, the error code if getpwnam() issues one.
3082 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3083 This is apparently needed in addition to the PH/07 change above to avoid
3084 any possible encoding problems.
3086 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3087 but not after initializing Perl.
3089 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3090 output them only if debugging. By default they are written stderr,
3091 apparently, which is not desirable.
3093 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3096 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3099 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3101 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3102 authenticated or an ident call has been made. Suppress the default
3103 values for $authenticated_id and $authenticated_sender (but permit -oMai
3104 and -oMas) when testing with -bh.
3106 PH/24 Re-jigged the order of the tests in the default configuration so that the
3107 tests for valid domains and recipients precede the DNS black list and CSA
3108 tests, on the grounds that those ones are more expensive.
3110 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3111 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3112 This bug exists in every version of Exim that I still have, right back to
3115 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3116 However, an attempt to turn on case-sensitivity in a regex key by
3117 including (?-i) didn't work because the subject string was already
3118 lowercased, and the effects were non-intuitive. It turns out that a
3119 one-line patch can be used to allow (?-i) to work as expected.
3125 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3126 other effects) broke the use of negated acl sub-conditions.
3128 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3131 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3132 "Deny" causes Exim to reject the incoming connection with a 554 error.
3133 Unfortunately, if there is a major crisis, such as a disk failure,
3134 tcp-wrappers gives "deny", whereas what one would like would be some
3135 kind of temporary error. A kludge has been added to help with this.
3136 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3137 554 error is used if errno is still zero or contains ENOENT (which occurs
3138 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3141 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3143 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3144 errors. Otherwise a message that provokes a temporary error (when other
3145 messages do not) can cause a whole host to time out.
3147 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3148 addresses were routed directly to files or pipes from a redirect router.
3149 File deliveries just didn't batch; pipe deliveries might have suffered
3152 PH/06 A failure to get a lock for a hints database would erroneously always say
3153 "Failed to get write lock", even when it was really a read lock.
3155 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3156 of 0600. This has been changed to use the value of the lockfile_mode
3157 option (which defaults to 0600).
3159 PH/08 Applied small patch from the Sieve maintainer.
3161 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3162 folder from quota calculations, a direct delivery into this folder messed
3163 up the contents of the maildirsize file. This was because the regex was
3164 used only to exclude .Trash (or whatever) when the size of the mailbox
3165 was calculated. There was no check that a delivery was happening into an
3166 excluded directory. This bug has been fixed by ignoring all quota
3167 processing for deliveries into excluded directories.
3169 PH/10 Added the maildirfolder_create_regex option to appendfile.
3175 PH/01 The code for finding all the local interface addresses on a FreeBSD
3176 system running IPv6 was broken. This may well have applied to all BSD
3177 systems, as well as to others that have similar system calls. The broken
3178 code found IPv4 interfaces correctly, but gave incorrect values for the
3179 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3180 that it would not match correctly against @[] and not recognize the IPv6
3183 PH/02 The ipliteral router was not recognizing addresses of the form user@
3184 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3186 PH/03 Added disable_ipv6.
3188 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3189 decoded form, because it is most often used to construct To: headers
3190 lines in autoreplies, and the decoded form may well be syntactically
3191 invalid. However, $reply_address has leading white space removed, and all
3192 newlines turned into spaces so that the autoreply transport does not
3195 PH/05 If group was specified without a user on a router, and no group or user
3196 was specified on a transport, the group from the router was ignored.
3198 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3199 for visible compile-time settings that can be used to change these
3200 numbers, for those that want even more. Backwards compatibility with old
3201 spool files has been maintained. However, going back to a previous Exim
3202 release will lost any variables that are in spool files.
3204 PH/07 Two small changes when running in the test harness: increase delay when
3205 passing a TCP/IP connection to a new process, in case the original
3206 process has to generate a bounce, and remove special handling of
3207 127.0.0.2 (sic), which is no longer necessary.
3209 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3210 be the same on different OS.
3212 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3215 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3216 whether --show-vars was specified or not
3218 JJ/02 exipick: Added support for new ACL variable spool format introduced
3221 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3222 syntactically invalid From: or Reply-to: line, and a filter used this to
3223 generate an autoreply, and therefore failed to obtain an address for the
3224 autoreply, Exim could try to deliver to a non-existent relative file
3225 name, causing unrelated and misleading errors. What now happens is that
3226 it logs this as a hard delivery error, but does not attempt to create a
3229 PH/11 The exinext utility has a -C option for testing purposes, but although
3230 the given file was scanned by exinext itself; it wasn't being passed on
3231 when Exim was called.
3233 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3234 an end-of-file indication when reading a command response.
3236 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3237 compiled. In many other places in Exim, IPv6 addresses are always
3238 recognized, so I have changed this. It also means that IPv4 domain
3239 literals of the form [IPV4:n.n.n.n] are now always recognized.
3241 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3242 used if the router is not running as root, for example, when verifying at
3243 ACL time, or when using -bh. The debugging output from this situation was
3244 non-existent - all you got was a failure to exec. I have made two
3247 (a) Failures to set uid/gid, the current directory, or a process leader
3248 in a subprocess such as that created by queryprogram now generate
3249 suitable debugging output when -d is set.
3251 (b) The queryprogram router detects when it is not running as root,
3252 outputs suitable debugging information if -d is set, and then runs
3253 the subprocess without attempting to change uid/gid.
3255 PH/15 Minor change to Makefile for building test_host (undocumented testing
3258 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3259 additional section of a DNS packet that returns MX or SRV records.
3260 Instead, it always explicitly searches for A/AAAA records. This avoids
3261 major problems that occur when a DNS server includes only records of one
3262 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3263 fixed another bug: if SRV records were looked up and the corresponding
3264 address records were *not* found in the additional section, the port
3265 values from the SRV records were lost.
3267 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3268 using the correct key (the original address) when searching the retry
3269 rules in order to find which one to use for generating the retry hint.
3271 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3272 adding the default one. Similarly, if it contains a Reply-To: header, the
3273 errors_reply_to option, if set, is not used.
3275 PH/19 When calculating a retry time, Exim used to measure the "time since
3276 failure" by looking at the "first failed" field in the retry record. Now
3277 it does not use this if it is later than than the arrival time of the
3278 message. Instead it uses the arrival time. This makes for better
3279 behaviour in cases where some deliveries succeed, thus re-setting the
3280 "first failed" field. An example is a quota failure for a huge message
3281 when small messages continue to be delivered. Without this change, the
3282 "time since failure" will always be short, possible causing more frequent
3283 delivery attempts for the huge message than are intended.
3284 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3286 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3287 $1, $2, $3) because the numerical variables can be reset during some
3288 expansion items (e.g. "match"), thereby losing the authentication data.
3290 PH/21 Make -bV show the size of off_t variables so that the test suite can
3291 decide whether to run tests for quotas > 2G.
3293 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3294 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3295 filecount value is greater than 2G or if a quota value is greater than 2G
3296 on a system where the size of off_t is not greater than 4, a panic error
3299 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3300 never match. The debug and -bh output now contains an explicit error
3301 message indicating a malformed IPv4 address or mask.
3303 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3304 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3305 PH/23 above applies.
3307 PH/25 Do not write to syslog when running in the test harness. The only
3308 occasion when this arises is a failure to open the main or panic logs
3309 (for which there is an explicit test).
3311 PH/26 Added the /no_tell option to "control=freeze".
3313 PH/27 If a host name lookup failed very early in a connection, for example, if
3314 the IP address matched host_lookup and the reverse lookup yielded a name
3315 that did not have a forward lookup, an error message of the form "no IP
3316 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3317 could be logged. Now it outputs the IP address instead of "NULL".
3319 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3320 allows the sender and the authenticated sender to be set when
3321 submitting a message from within Exim. Since child_open_exim() is
3322 documented for local_scan(), the new function should be too.
3324 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3325 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3326 results in an empty string is now treated as unset.
3328 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3330 PH/31 Added sender_verify_fail logging option.
3332 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3333 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3334 tidied the source and removed it altogether.
3336 PH/33 When a queue run was abandoned because the load average was too high, a
3337 log line was always written; now it is written only if the queue_run log
3338 selector is set. In addition, the log line for abandonment now contains
3339 information about the queue run such as the pid. This is always present
3340 in "start" and "stop" lines but was omitted from the "abandon" line.
3342 PH/34 Omit spaces between a header name and the colon in the error message that
3343 is given when verify = headers_syntax fails (if there are lots of them,
3344 the message gets confusing).
3346 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3347 names, as there are now some PTR records that contain slashes. This check
3348 is only to protect against broken name servers that fall over on strange
3349 characters, so the fact that it applies to all lookups doesn't matter.
3351 PH/36 Now that the new test suite is complete, we can remove some of the
3352 special code in Exim that was needed for the old test suite. For example,
3353 sorting DNS records because real resolvers return them in an arbitrary
3354 order. The new test suite's fake resolver always returns records in the
3357 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3358 bounces) except when queue_only is set, to avoid logging races between
3359 the different processes.
3361 PH/38 Panic-die if .include specifies a non-absolute path.
3363 PH/39 A tweak to the "H" retry rule from its user.
3365 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3366 a label. They prevented compilation on older perls.
3368 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3369 a warning to be raised on newish perls.
3371 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3372 on queue. Changes to match documented behaviour of showing count of
3373 messages matching specified criteria.
3375 PH/40 Changed the default ident timeout from 30s to 5s.
3377 PH/41 Added support for the use of login_cap features, on those BSD systems
3378 that have them, for controlling the resources used by pipe deliveries.
3380 PH/42 The content-scanning code uses fopen() to create files in which to put
3381 message data. Previously it was not paying any attention to the mode of
3382 the files. Exim runs with umask(0) because the rest of the code creates
3383 files with open(), and sets the required mode explicitly. Thus, these
3384 files were ending up world-writeable. This was not a big issue, because,
3385 being within the spool directory, they were not world-accessible. I have
3386 created a function called modefopen, which takes an additional mode
3387 argument. It sets umask(777), creates the file, chmods it to the required
3388 mode, then resets the umask. All the relevant calls to fopen() in the
3389 content scanning code have been changed to use this function.
3391 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3392 to 24 hours. This avoids potential overflow problems when processing G
3393 and H retry rules. I suspect nobody ever tinkers with this value.
3395 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3397 PH/45 When the plaintext authenticator is running as a client, the server's
3398 challenges are checked to ensure they are valid base64 strings. By
3399 default, the authentication attempt is cancelled if an invalid string is
3400 received. Setting client_ignore_invalid_base64 true ignores these errors.
3401 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3402 they are received. Thus, the responses can be made to depend on the
3403 challenges. If an invalid string is ignored, an empty string is placed in
3406 PH/46 Messages that are created by the autoreply transport now contains a
3407 References: header, in accordance with RFCs 2822 and 3834.
3409 PH/47 Added authenticated_sender_force to the smtp transport.
3411 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3413 PH/49 Installed latest patch from the Sieve maintainer.
3415 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3416 was also set, the appendfile transport was unnecessarily scanning a
3417 directory of message files (e.g. for maildir delivery) to find the count
3418 of files (along with the size), even though it did not need this
3419 information. It now does the scan only if it needs to find either the
3420 size of the count of files.
3422 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3424 PH/52 Two bugs concerned with error handling when the smtp transport is
3427 (i) Exim was not creating retry information for temporary errors given
3428 for individual recipients after the DATA command when the smtp transport
3429 was used in LMTP mode. This meant that they could be retried too
3430 frequently, and not timed out correctly.
3432 (ii) Exim was setting the flag that allows error details to be returned
3433 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3434 recipients that were returned after the DATA command.
3436 PH/53 This is related to PH/52, but is more general: for any failing address,
3437 when detailed error information was permitted to be returned to the
3438 sender, but the error was temporary, then after the final timeout, only
3439 "retry timeout exceeded" was returned. Now it returns the full error as
3440 well as "retry timeout exceeded".
3442 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3443 do this, and (what is worse) MTAs that accept it.
3445 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3446 will now be deprecated.
3448 PH/56 New os.c-cygwin from the Cygwin maintainer.
3450 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3451 formats (previously only available in exim formats via -bpr, -bpru,
3452 and -bpra. Now also available in native and exiqgrep formats)
3454 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3455 with very large, slow to parse queues
3457 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3459 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3461 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3462 responses to authentication challenges, though it was showing the
3463 challenges; (ii) I've removed the CR characters from the debug output for
3466 PH/58 Allow for the insertion of a newline as well as a space when a string
3467 is turned into more than one encoded-word during RFC 2047 encoding. The
3468 Sieve code now uses this.
3470 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3471 data_4xx, lost_connection, tls_required.
3473 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3474 message was being sent as an SMTP response.
3476 PH/61 Add -l and -k options to exicyclog.
3478 PH/62 When verifying, if an address was redirected to one new address, so that
3479 verification continued, and the new address failed or deferred after
3480 having set something in $address_data, the value of $address_data was not
3481 passed back to the ACL. This was different to the case when no
3482 redirection occurred. The value is now passed back in both cases.
3484 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3485 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3486 use login_cap.h, so on its own it isn't the distinguishing feature. The
3487 new name refers directly to the setclassresources() function.
3489 PH/65 Added configuration files for NetBSD3.
3491 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3493 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3494 is preferred over IPv4.
3496 PH/68 The bounce_return_message and bounce_return_body options were not being
3497 honoured for bounces generated during the reception of non-SMTP messages.
3498 In particular, this applied to messages rejected by the ACL. This bug has
3499 been fixed. However, if bounce_return_message is true and bounce_return_
3500 body is false, the headers that are returned for a non-SMTP message
3501 include only those that have been read before the error was detected.
3502 (In the case of an ACL rejection, they have all been read.)
3504 PH/69 The HTML version of the specification is now built in a directory called
3505 spec_html instead of spec.html, because the latter looks like a path with
3506 a MIME-type, and this confuses some software.
3508 PH/70 Catch two compiler warnings in sieve.c.
3510 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3511 function verify_get_ident() calls ip_connect() to connect a socket, but
3512 if the "connect()" function timed out, ip_connect() used to close the
3513 socket. However, verify_get_ident() also closes the socket later, and in
3514 between Exim writes to the log, which may get opened at this point. When
3515 the socket was closed in ip_connect(), the log could get the same file
3516 descriptor number as the socket. This naturally causes chaos. The fix is
3517 not to close the socket in ip_connect(); the socket should be closed by
3518 the function that creates it. There was only one place in the code where
3519 this was missing, in the iplookup router, which I don't think anybody now
3520 uses, but I've fixed it anyway.
3522 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3523 well as to direct DNS lookups. Otherwise the handling of names in host
3524 lists is inconsistent and therefore confusing.
3530 PH/01 Two changes to the default runtime configuration:
3532 (1) Move the checks for relay_from_hosts and authenticated clients from
3533 after to before the (commented out) DNS black list checks.
3535 (2) Add control=submission to the relay_from_hosts and authenticated
3536 clients checks, on the grounds that messages accepted by these
3537 statements are most likely to be submissions.
3539 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3541 (1) Generate an error if the third argument for the ${prvs expansion is
3544 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3547 (3) Reset the variables that are obtained from the first argument of
3548 ${prvscheck and used in the second argument before leaving the code,
3549 because their memory is reclaimed, so using them afterwards may do
3552 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3553 one (it's much easier than Tom thought :-).
3555 (5) Because of (4), we can now allow for the use of $prvscheck_result
3556 inside the third argument.
3558 PH/03 For some reason, the default setting of PATH when running a command from
3559 a pipe transport was just "/usr/bin". I have changed it to
3562 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3563 anything to be listed in the output from -bV.
3565 PH/05 When a filter generated an autoreply, the entire To: header line was
3566 quoted in the delivery log line, like this:
3568 => >A.N.Other <ano@some.domain> <original@ddress> ...
3570 This has been changed so that it extracts the operative address. There
3571 may be more than one such address. If so, they are comma-separated, like
3574 => >ano@some.domain,ona@other.domain <original@ddress> ...
3576 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3577 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3578 not being looked up in the rDNS to get a host name, Exim was showing the
3579 IP address twice in Received: lines, even though the IP addresses were
3580 identical. For example:
3582 Received: from [1.2.3.4] (helo=[1.2.3.4])
3584 However, if the real host name was known, it was omitting the HELO data
3585 if it matched the actual IP address. This has been tidied up so that it
3586 doesn't show the same IP address twice.
3588 PH/07 When both +timestamp and +memory debugging was on, the value given by
3589 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3590 called by the debug printing, thereby overwriting the timestamp buffer.
3591 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3593 PH/08 When the original message was included in an autoreply transport, it
3594 always said "this is a copy of the message, including all the headers",
3595 even if body_only or headers_only was set. It now gives an appropriate
3598 PH/09 Applied a patch from the Sieve maintainer which:
3600 o fixes some comments
3601 o adds the (disabled) notify extension core
3602 o adds some debug output for the result of if/elsif tests
3603 o points to the current vacation draft in the documentation
3604 and documents the missing references header update
3608 o fixes a bug in processing the envelope test (when testing
3609 multiple envelope elements, the last element determined the
3612 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3613 Electronic Mail") by including:
3615 Auto-submitted: auto-generated
3617 in the messages that it generates (bounce messages and others, such as
3618 warnings). In the case of bounce messages for non-SMTP messages, there was
3619 also a typo: it was using "Auto_submitted" (underscore instead of
3620 hyphen). Since every message generated by Exim is necessarily in response
3621 to another message, thes have all been changed to:
3623 Auto-Submitted: auto-replied
3625 in accordance with these statements in the RFC:
3627 The auto-replied keyword:
3629 - SHOULD be used on messages sent in direct response to another
3630 message by an automatic process,
3632 - MUST NOT be used on manually-generated messages,
3634 - MAY be used on Delivery Status Notifications (DSNs) and Message
3635 Disposition Notifications (MDNs),
3637 - MUST NOT be used on messages generated by automatic or periodic
3638 processes, except for messages which are automatic responses to
3641 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3642 to the default Received: header definition.
3644 PH/12 Added log selector acl_warn_skipped (default on).
3646 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3647 variables because (a) they are in the wrong storage pool and (b) even if
3648 they were copied, it wouldn't work properly because of the caching.
3650 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3651 checking when decoding. Apparently there are clients that generate
3652 overlong encoded strings. Why am I not surprised?
3654 PH/15 If the first argument of "${if match_address" was not empty, but did not
3655 contain an "@" character, Exim crashed. Now it writes a panic log message
3656 and treats the condition as false.
3658 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3660 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3661 extension "envelope-auth". The code is finished and in agreement with
3662 other implementations, but there is no documentation so far and in fact,
3663 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3664 not changing the active code.
3666 Print executed "if" and "elsif" statements when debugging is used. This
3667 helps a great deal to understand what a filter does.
3669 Document more things not specified clearly in RFC3028. I had all this
3670 sorted out, when out of a sudden new issues came to my mind. Oops."
3672 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3675 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3676 canonical form (as documented). However, after a host name lookup from
3677 the IP address, check_host() was doing a simple string comparison with
3678 addresses acquired from the DNS when checking that the found name did
3679 have the original IP as one of its addresses. Since any found IPv6
3680 addresses are likely to be in abbreviated form, the comparison could
3681 fail. Luckily, there already exists a function for doing the comparison
3682 by converting both addresses to binary, so now that is used instead of
3683 the text comparison.
3685 PH/20 There was another similar case to PH/19, when a complete host name was
3686 given in a host list; looking up its IP address could give an abbreviated
3687 form, whereas the current host's name might or might not be abbreviated.
3688 The same fix has been applied.
3694 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3695 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3698 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3699 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3701 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3703 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3704 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3705 Also, for "Auto-Submitted", treat anything other than "no" as
3706 non-personal, in accordance with RFC 3834. (Previously it treated
3707 anything starting "auto-" as non-personal.)
3709 TF/01 The control=submission/name=... option had a problem with syntax
3710 errors if the name included a slash character. The /name= option
3711 now slurps the rest of the string, so it can include any characters
3712 but it must come last in the list of options (after /sender_retain
3715 PH/05 Some modifications to the interface to the fake nameserver for the new
3723 TK/01 Added the "success_on_redirect" address verification option. See
3724 NewStuff for rationale and an example.
3726 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3728 PH/02 Patch to exigrep to allow it to work on syslog lines.
3730 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3731 fread() to skip over the body file's header line, because in Cygwin the
3732 header line is locked and is inaccessible.
3734 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3735 co-exist for some time) to make it clear that it is the Exim ID that is
3736 referenced, not the Message-ID: header line.
3738 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3739 string_format() function, because snprintf() does not exist on all
3742 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3743 ${stat: expansion item.
3745 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3746 protocol synchronization error", to keep the pedants happy.
3748 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3749 well as for IRIX systems, when gcc is being used. See the host.c source
3752 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3754 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3757 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3758 transport and to the smtp transport in LMTP mode.
3760 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3762 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3763 run a filter in a subprocess. This could lead to confusion in subsequent
3764 lookups in the parent process. There should also be a search_tidyup() at
3765 the end of the subprocess.
3767 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3768 only if the host matched helo_try_verify_hosts, which caused the
3769 verification to occur when the EHLO/HELO command was issued. The ACL just
3770 tested the remembered result. Now, if a previous verification attempt has
3771 not happened, "verify = helo" does it there and then.
3773 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3775 TK/03 Fix log output including CR from clamd.
3777 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3778 exist provoked a memory error which could cause a segfault.
3780 PH/15 Installed PCRE 6.2
3782 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3784 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3785 of the problem. Specifically, suggested +O2 rather than +O1 for the
3788 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3790 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3791 record, the port was not correctly logged when the outgoing_port log
3792 selector was set (it logged the transort's default port).
3794 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3795 fallback_hosts, and "hosts" in the smtp transport.
3797 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3798 host errors such as "Connection refused".
3800 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3801 authentication with radiusclient 0.4.9:
3803 - Error returned from rc_read_config was caught wrongly
3804 - Username/password not passed on to radius server due to wrong length.
3806 The presumption is that some radiusclient API changes for 4.51/PH/17
3807 were not taken care of correctly. The code is still untested by me (my
3808 Linux distribution still has 0.3.2 of radiusclient), but it was
3809 contributed by a Radius user.
3811 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3812 expanding the "port" option of the smtp transport.
3814 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3815 while reading a MIME header. Thanks to Tom Hughes for a patch.
3817 PH/24 Include config.h inside local_scan.h so that configuration settings are
3820 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3821 that in an ACL for RCPT (for example), you can examine exactly what was
3824 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3825 commands, but it was not correctly comparing the address with the actual
3826 client host address. Thus, it would show the EHLO address in Received:
3827 header lines when this was not necessary.
3829 PH/27 Added the % operator to ${eval:}.
3831 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3832 it should be ignoring failures (because with -C, for example, it has lost
3833 privilege). It wasn't ignoring creation failures other than "already
3836 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3839 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3840 because an input line was too long, either on its own, or by virtue of
3841 too many continuations, the temporary file was not being removed, and the
3842 return code was incorrect.
3844 PH/31 Missing "BOOL" in function definition in filtertest.c.
3846 PH/32 Applied Sieve patches from the maintainer.
3848 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3850 PH/33 Added "verify = not_blind".
3852 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3853 Local/Makefile (with some defaults set). These are used in built scripts
3854 such as exicyclog, but they have never been used in the exim_install
3855 script (though there are many overriding facilities there). I have
3856 arranged that the exim_install script now takes note of these two
3859 PH/35 Installed configuration files for Dragonfly.
3861 PH/36 When a locally submitted message by a trusted user did not contain a
3862 From: header, and the sender address was obtained from -f or from an SMTP
3863 MAIL command, and the trusted user did not use -F to supply a sender
3864 name, $originator_name was incorrectly used when constructing a From:
3865 header. Furthermore, $originator_name was used for submission mode
3866 messages from external hosts without From: headers in a similar way,
3867 which is clearly wrong.
3869 PH/37 Added control=suppress_local_fixups.
3871 PH/38 When log_selector = +received_sender was set, and the addition of the
3872 sender made the log line's construction buffer exactly full, or one byte
3873 less than full, an overflow happened when the terminating "\n" was
3876 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3877 when the result of a list match is failure because a DNS lookup failed.
3879 PH/40 RM_COMMAND is now used in the building process.
3881 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3882 the "build-* directories that it finds.
3884 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3885 address in a domain literal was a prefix of an interface address.
3887 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3888 when verifying a sender address, unless rewrite_headers is false.
3890 PH/44 Wrote a long comment about why errors_to addresses are verified as
3891 recipients, not senders.
3893 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3894 the ratelimit ACL was added.
3896 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3898 PH/46 Added extra information about PostgreSQL errors to the error string.
3900 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3901 avoiding the need to install special zones in a real server. This is
3902 backwards compatible; if it can't find the fake resolver, it drops back.
3903 Thus, both old and new test suites can be run.
3905 TF/02 Added util/ratelimit.pl
3907 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3908 clock is set back in time.
3910 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3911 Candler <B.Candler@pobox.com>.
3913 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3914 OK for addresses that are the result of redirections.
3916 PH/48 A number of further additions for the benefit of the new test suite,
3917 including a fake gethostbyname() that interfaces to the fake DNS resolver
3920 TF/06 The fix for widen_domains has also been applied to qualify_single and
3921 search_parents which are the other dnslookup options that can cause
3924 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3927 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3929 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3930 service" if it encountered an MX record with an empty target hostname.
3931 The message is now "an MX or SRV record indicated no SMTP service".
3933 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3934 if the DNS of the sending site is misconfigured. This is quite a
3935 common situation. This change restores the behaviour of treating a
3936 helo verification defer as a failure.
3938 PH/51 If self=fail was set on a router, the bounce message did not include the
3939 actual error message.
3945 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3947 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3948 command itself ended in error, the underlying message about the transport
3949 filter timeout was being overwritten with the pipe command error. Now the
3950 underlying error message should be appended to the second error message.
3952 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3954 PH/02 Reduce the amount of output that "make" produces by default. Full output
3955 can still be requested.
3957 PH/03 The warning log line about a condition test deferring for a "warn" verb
3958 was being output only once per connection, rather than after each
3959 occurrence (because it was using the same function as for successful
3960 "warn" verbs). This seems wrong, so I have changed it.
3962 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3963 it should not have, which might have caused a crash in the right
3964 circumstances, but probably never did.
3966 PH/04 Installed a modified version of Tony Finch's patch to make submission
3967 mode fix the return path as well as the Sender: header line, and to
3968 add a /name= option so that you can make the user's friendly name appear
3971 TF/03 Added the control = fakedefer ACL modifier.
3973 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3974 Mark Lowes for thorough testing.
3976 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3978 TK/03 Merged latest SRS patch from Miles Wilton.
3980 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3981 with the definition in sysexits.h (which is #included earlier).
3982 Fortunately, Exim does not actually use EX_OK. The code used to try to
3983 preserve the sysexits.h value, by assuming that macro definitions were
3984 scanned for macro replacements. I have been disabused of this notion,
3985 so now the code just undefines EX_OK before #including unistd.h.
3987 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3988 in the smtp transport. When a block could not be written in a single
3989 write() function, the timeout was being re-applied to each part-write.
3990 This seems wrong - if the receiver was accepting one byte at a time it
3991 would take for ever. The timeout is now adjusted when this happens. It
3992 doesn't have to be particularly precise.
3994 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3995 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3997 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3998 without a domain if the check to <postmaster@domain> fails.
4000 SC/01 Eximstats: added -xls and the ability to specify output files
4001 (patch written by Frank Heydlauf).
4003 SC/02 Eximstats: use FileHandles for outputting results.
4005 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4007 SC/04 Eximstats: fixed display of large numbers with -nvr option
4009 SC/05 Eximstats: fixed merging of reports with empty tables.
4011 SC/06 Eximstats: added the -include_original_destination flag
4013 SC/07 Eximstats: removed tabs and trailing whitespace.
4015 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4017 TK/06 MBOX spool code: Add real "From " MBOX separator line
4018 so the .eml file is really in mbox format (even though
4019 most programs do not really care). Patch from Alex Miller.
4021 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4022 The latter is generated from $received_to and is only set if the
4023 message has one envelope recipient. SA can use these headers,
4024 obviously out-of-the-box. Patch from Alex Miller.
4026 PH/08 The ${def test on a variable was returning false if the variable's
4027 value was "0", contrary to what the specification has always said!
4028 The result should be true unless the variable is empty.
4030 PH/09 The syntax error of a character other than { following "${if
4031 def:variable_name" (after optional whitespace) was not being diagnosed.
4032 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4033 accidental colon was present, for example, could give incorrect results.
4035 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4036 result is used (not including appendfile, where other changes are about
4039 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4040 This involved changing a lot of size variables from int to off_t. It
4041 should work with maildirs and everything.
4043 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4044 spamd dying while we are connected to it.
4046 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4049 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4050 function for BDB 4.3.
4052 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4054 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4055 the sources for PCRE have been re-arranged and more files are now
4058 PH/15 The code I had for printing potentially long long variables in PH/11
4059 above was not the best (it lost precision). The length of off_t variables
4060 is now inspected at build time, and an appropriate printing format (%ld
4061 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4062 to be "long long int" or "long int". This is needed for the internal
4063 formatting function string_vformat().
4065 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4066 the configuration file to be ":syslog", then the script "guesses" where
4067 the logs files are, rather than using the compiled in default. In our
4068 case the guess is not the same as the compiled default, so the script
4069 suddenly stopped working when I started to use syslog. The patch checks
4070 to see if log_file_path is "". If so, it attempts to read it from exim
4071 with no configuration file to get the compiled in version, before it
4072 falls back to the previous guessing code."
4074 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4075 implementing BATV in an Exim configuration. See NewStuff for the gory
4078 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4079 Makefile that are specific to HP-UX.
4081 PH/18 If the "use_postmaster" option was set for a recipient callout together
4082 with the "random" option, the postmaster address was used as the MAIL
4083 FROM address for the random test, but not for the subsequent recipient
4084 test. It is now used for both.
4086 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4087 patch removes a few documentation additions to RFC 3028, because the
4088 latest draft now contains them. It adds the new en;ascii-case comparator
4089 and a new error check for 8bit text in MIME parts. Comparator and
4090 require names are now matched exactly. I enabled the subaddress
4091 extension, but it is not well tested yet (read: it works for me)."
4093 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4094 rework some of the code of TK/09 above to avoid the hardwired use of
4095 "%lld" and "long long". Replaced the call to snprintf() with a call to
4098 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4099 records point to non-existent hosts", "retry timeout exceeded", and
4100 "retry time not reached for any host after a long failure period".
4102 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4103 experimental DomainKeys support:
4105 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4106 (2) On an error such as an illegally used "control", the wrong name for
4107 the control was given.
4109 These problems did NOT occur unless DomainKeys support was compiled.
4111 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4113 PH/24 Added ${if match_ip condition.
4115 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4116 databases so that it will be absolutely obvious if a crash occurs in the
4117 DB library. This is a regular occurrence (often caused by mis-matched
4120 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4121 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4122 picked up on a user's system that detects such things. There doesn't seem
4123 to be a gcc warning option for this - only an attribute that has to be
4124 put on the function's prototype. It seems that in Fedora Core 4 they have
4125 set this on a number of new functions. No doubt there will be more in due
4128 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4129 specify a transport. However, if an address that was verified by such a
4130 router was the subject of a callout, Exim crashed because it tried to
4131 read the rcpt_include_affixes from the non-existent transport. Now it
4132 just assumes that the setting of that option is false. This bug was
4133 introduced by 4.51/PH/31.
4135 PH/28 Changed -d+all to exclude +memory, because that information is very
4136 rarely of interest, but it makes the output a lot bigger. People tend to
4137 do -d+all out of habit.
4139 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4140 code in os-type was giving problems when libc.so lives in lib64, like on
4143 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4144 aren't the modern standard, and it seems that some systems' include files
4145 don't always have them. Exim was already checking for some of the newer
4146 ones like T_AAAA, and defining it itself. I've added checks for all the
4147 record types that Exim uses.
4149 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4150 not automatically generating a new one, as it is supposed to. This
4151 prevented TLS from working. If the file did exist, but contained invalid
4152 data, a new version was generated, as expected. It was only the case of a
4153 non-existent file that was broken.
4155 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4156 with a change in libdomainkeys > 0.64.
4158 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4159 from DNS. If the selector record carries the flag, it now has
4160 precedence over the domain-wide flag.
4162 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4164 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4165 the use of a port name that isn't defined in /etc/services) occurred, the
4166 message was deferred as in a normal delivery, and thus remained on the
4167 spool, instead of being failed because of the mua_wrapper setting. This
4168 is now fixed, and I tidied up some of the mua_wrapper messages at the
4171 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4172 the messages in a hash of arrays rather than using individual hashes.
4173 This is a bit cleaner and results in dramatic memory savings, albeit
4174 at a slight CPU cost.
4176 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4177 as requested by Marc Sherman.
4179 SC/10 Eximstats: added histograms for user specified patterns as requested
4182 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4184 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4185 fopen() in the content-scanning modules that did not already have it.
4191 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4192 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4194 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4196 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4198 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4199 to test Sieve filters that use "vacation".
4201 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4202 that changes the way the GnuTLS parameters are stored in the cache file.
4203 The new format can be generated externally. For backward compatibility,
4204 if the data in the cache doesn't make sense, Exim assumes it has read an
4205 old-format file, and it generates new data and writes a new file. This
4206 means that you can't go back to an older release without removing the
4209 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4210 work if there are any delivery delays because "one_time" forces the
4211 parent to be marked "delivered", so its unseen clone is never tried
4212 again. For this reason, Exim now forbids the simultaneous setting of
4215 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4216 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4217 entry if you want to know the details. The fix, however, neglected to
4218 consider the case when local delivery batching is involved. The test for
4219 "previously delivered" was not happening when checking to see if an
4220 address could be batched with a previous (undelivered) one; under
4221 certain circumstances this could lead to multiple deliveries to the same
4224 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4225 in its include files, and this causes problems building Exim.
4227 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4228 header_syntax) but Exim was just ignoring anything given after a slash.
4229 In particular, this caused confusion with an attempt to use "verify =
4230 reverse_host_lookup/defer_ok". An error is now given when options are
4231 supplied for verify items that do not have them. (Maybe reverse_host_
4232 lookup should have a defer_ok option, but that's a different point.)
4234 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4235 defined by RFC 821) to 2048, because there were problems with some AUTH
4236 commands, and RFC 1869 says the size should be increased for extended
4237 SMTP commands that take arguments.
4239 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4242 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4243 "unknown" error; now it says that the functionality isn't in the binary.
4245 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4246 an address' error message when a string expansion fails (syntax or
4247 whatever). Otherwise the password may appear in the log. Following change
4248 PH/42 below, there is no longer a chance of it appearing in a bounce
4251 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4253 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4254 of its addresses was ever tried. (Bugzilla bug #2.)
4256 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4257 the result incorrectly in the debug output. (It correctly added a newline
4258 to what was transported.)
4260 TF/01 Added $received_time.
4262 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4263 commented out examples of how to interface to a virus scanner and to
4264 SpamAssassin. Also added commented examples of av_scanner and
4265 spamd_address settings.
4267 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4268 and controls are allowed in which ACLs. There were a couple of minor
4269 errors. Some of the entries in the conditions table (which is a table of
4270 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4271 negation of where the condition IS allowed.
4273 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4275 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4276 header file does not have a version number, so I've had to invent a new
4277 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4278 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4279 radiusclient), but it was contributed by a Radius user.
4281 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4282 files or directories, for OpenSSL.
4284 PH/19 When an Exim process that is running as root has to create an Exim log
4285 file, it does so in a subprocess that runs as exim:exim so as to get the
4286 ownership right at creation (otherwise, other Exim processes might see
4287 the file with the wrong ownership). There was no test for failure of this
4288 fork() call, which would lead to the process getting stuck as it waited
4289 for a non-existent subprocess. Forks do occasionally fail when resources
4290 run out. I reviewed all the other calls to fork(); they all seem to check
4293 PH/20 When checking for unexpected SMTP input at connect time (before writing
4294 the banner), Exim was not dealing correctly with a non-positive return
4295 from the read() function. If the client had disconnected by this time,
4296 the result was a log entry for a synchronization error with an empty
4297 string after "input=" when read() returned zero. If read() returned -1
4298 (an event I could not check), uninitialized data bytes were printed.
4299 There were reports of junk text (parts of files, etc) appearing after
4302 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4304 PH/22 Added support for macro redefinition, and (re)definition in between
4305 driver and ACL definitions.
4307 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4308 forgetting to use the resulting value; it was using the unexpanded value.
4310 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4311 hadn't been configured. The fix is from Juergen Kreileder, who
4312 understands it better than I do:
4314 "Here's what I see happening with three configured cyrus_sasl
4315 authenticators configured (plain, login, cram-md5):
4317 On startup auth_cyrus_sasl_init() gets called for each of these.
4318 This means three calls to sasl_listmech() without a specified mech_list.
4319 => SASL tests which mechs of all available mechs actually work
4320 => three warnings about OTP not working
4321 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4323 With the patch, sasl_listmech() also gets called three times. But now
4324 SASL's mech_list option is set to the server_mech specified in the the
4325 authenticator. Or in other words, the answer from sasl_listmech()
4326 gets limited to just the mech you're testing for (which is different
4328 => the return list contains just 'plain' or 'login', 'cram-md5' or
4329 nothing depending on the value of ob->server_mech.
4331 I've just tested the patch: Authentication still works fine,
4332 unavailable mechs specified in the exim configuration are still
4333 caught, and the auth.log warnings about OTP are gone."
4335 PH/25 When debugging is enabled, the contents of the command line are added
4336 to the debugging output, even when log_selector=+arguments is not
4339 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4340 answer is "GNU", and only if the return is "GNU/something" is the answer
4343 PH/27 $acl_verify_message is now set immediately after the failure of a
4344 verification in an ACL, and so is available in subsequent modifiers. In
4345 particular, the message can be preserved by coding like this:
4347 warn !verify = sender
4348 set acl_m0 = $acl_verify_message
4350 Previously, $acl_verify_message was set only while expanding "message"
4351 and "log_message" when a very denied access.
4353 PH/28 Modified OS/os.c-Linux with
4355 -#ifndef OS_LOAD_AVERAGE
4356 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4358 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4359 nomenclature these days.)
4361 PH/29 Installed patch from the Sieve maintainer that adds the options
4362 sieve_useraddress and sieve_subaddress to the redirect router.
4364 PH/30 In these circumstances:
4365 . Two addresses routed to the same list of hosts;
4366 . First host does not offer TLS;
4367 . First host accepts first address;
4368 . First host gives temporary error to second address;
4369 . Second host offers TLS and a TLS session is established;
4370 . Second host accepts second address.
4371 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4372 and peerdn, if requested) that were in fact used only for the second
4375 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4376 attention to any local part prefix or suffix that was matched by the
4377 router that accepted the address. It now behaves in the same way as it
4378 does for delivery: the affixes are removed from the local part unless
4379 rcpt_include_affixes is set on the transport.
4381 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4382 timeout during the DATA phase of an incoming message.
4384 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4385 applied a patch sanctioned by the Sieve maintainer.
4387 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4388 the uid or gid is negative. A case of a negative gid caused this to be
4389 noticed. The fix allows for either to be negative.
4391 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4392 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4395 PH/36 The change PH/12 above was broken. Fixed it.
4397 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4398 the grounds that routing the same address twice would always produce the
4399 same answer. This might have been true once, but it is certainly no
4400 longer true now. Routing a child address may depend on the previous
4401 routing that produced that child. Some complicated redirection strategies
4402 went wrong when messages had multiple recipients, and made Exim's
4403 behaviour dependent on the order in which the addresses were given.
4405 I have moved the duplicate checking until after the routing is complete.
4406 Exim scans the addresses that are assigned to local and remote
4407 transports, and removes any duplicates. This means that more work will be
4408 done, as duplicates will always all be routed, but duplicates are
4409 presumably rare, so I don't expect this is of any significance.
4411 For deliveries to pipes, files, and autoreplies, the duplicate checking
4412 still happens during the routing process, since they are not going to be
4415 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4416 It corrects a timeout issue with spamd. This is Ian's comment: "The
4417 background is that sometimes spamd either never reads data from a
4418 connection it has accepted, or it never writes response data. The exiscan
4419 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4420 blindly assumes that writes won't block so it may never time out."
4422 PH/39 Allow G after quota size as well as K and M.
4424 PH/40 The value set for $authenticated_id in an authenticator may not contain
4425 binary zeroes or newlines because the value is written to log lines and
4426 to spool files. There was no check on this. Now the value is run through
4427 the string_printing() function so that such characters are converted to
4428 printable escape sequences.
4430 PH/41 $message_linecount is a new variable that contains the total number of
4431 lines in the message. Compare $body_linecount, which is the count for the
4434 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4435 bounce and delay warning messages, except in certain special cases, which
4438 (a) An SMTP error message from a remote host;
4439 (b) A message specified in a :fail: redirection;
4440 (c) A message specified in a "fail" command in a system filter;
4441 (d) A message specified in a FAIL return from the queryprogram router;
4442 (e) A message specified by the cannot_route_message router option.
4444 In these cases only, Exim does include the error details in bounce and
4445 warning messages. There are also a few cases where bland messages such
4446 as "unrouteable address" or "local delivery error" are given.
4448 PH/43 $value is now also set for the "else" part of a ${run expansion.
4450 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4451 being worked on, but at least Exim now implements the latest version to
4454 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4455 process to complete was treated as a delivery failure, a timeout while
4456 writing the message to the pipe was logged, but erroneously treated as a
4457 successful delivery. Such timeouts include transport filter timeouts. For
4458 consistency with the overall process timeout, these timeouts are now
4459 treated as errors, giving rise to delivery failures by default. However,
4460 there is now a new Boolean option for the pipe transport called
4461 timeout_defer, which, if set TRUE, converts the failures into defers for
4462 both kinds of timeout. A transport filter timeout is now identified in
4465 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4466 systems where "make" and "gmake" are different, calling "gmake" at top
4467 level broke things. I've arranged for the value of $(MAKE) to be passed
4468 from the Makefile to this script so that it can call the same version of
4472 A note about Exim versions 4.44 and 4.50
4473 ----------------------------------------
4475 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4476 changes of various kinds. As a consequence, a big documentation update was
4477 needed. This delayed the release for rather longer than seemed good, especially
4478 in the light of a couple of (minor) security issues. Therefore, the changes
4479 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4480 release. So 4.44 and 4.50 are in effect two different branches that both start
4483 I have left the 4.50 change log unchanged; it contains all the changes since
4484 4.43. The change log for 4.44 is below; many of its items are identical to
4485 those for 4.50. This seems to be the most sensible way to preserve the
4486 historical information.
4492 1. Minor wording change to the doc/README.SIEVE file.
4494 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4495 computation of the current number of files was incorrect.
4497 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4498 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4501 4. Give more explanation in the error message when the command for a transport
4502 filter fails to execute.
4504 5. There are several places where Exim runs a non-Exim command in a
4505 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4506 was being done only for the command run by the queryprogram router. It is
4507 now done for all such subprocesses. The other cases are: ${run, transport
4508 filters, and the commands run by the lmtp and pipe transports.
4510 6. Added CONFIGURE_GROUP build-time option.
4512 7. Some older OS have a limit of 256 on the maximum number of file
4513 descriptors. Exim was using setrlimit() to set 1000 as a large value
4514 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4515 systems. I've change it so that if it can't get 1000, it tries for 256.
4517 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4518 was an oversight, and furthermore, ever since the addition of extra
4519 controls (e.g. 4.43/32), the checks on when to allow different forms of
4520 "control" were broken. There should now be diagnostics for all cases when a
4521 control that does not make sense is encountered.
4523 9. Added the /retain_sender option to "control=submission".
4525 10. $recipients is now available in the predata ACL (oversight).
4527 11. Tidy the search cache before the fork to do a delivery from a message
4528 received from the command line. Otherwise the child will trigger a lookup
4529 failure and thereby defer the delivery if it tries to use (for example) a
4530 cached ldap connection that the parent has called unbind on.
4532 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4533 of $address_data from the recipient verification was clobbered by the
4534 sender verification.
4536 13. The value of address_data from a sender verification is now available in
4537 $sender_address_data in subsequent conditions in the ACL statement.
4539 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4541 15. Added a new option "connect=<time>" to callout options, to set a different
4544 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4545 was its contents. (It was OK if the option was not defined at all.)
4547 17. A "Completed" log line is now written for messages that are removed from
4548 the spool by the -Mrm option.
4550 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4551 information about exactly what failed.
4553 19. Added -dd to debug only the daemon process.
4555 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4556 handles timeouts, both on the server side and network timeouts. Renamed the
4557 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4559 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4560 It is now set to "smtps".
4562 22. $host_address is now set to the target address during the checking of
4563 ignore_target_hosts.
4565 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4566 being passed; this would have caused $sender_host_name to have been used if
4567 matching the list had actually called for a host name (not very likely,
4568 since this list is usually IP addresses). A host name is now passed as
4571 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4572 code that specifies a non-restarting handler (typically sigaction() in
4573 modern systems) in an attempt to fix a rare and obscure crash bug.
4575 25. Narrowed the window for a race in the daemon that could cause it to ignore
4576 SIGCHLD signals. This is not a major problem, because they are used only to
4577 wake it up if nothing else does.
4579 26. A malformed maildirsize file could cause Exim to calculate negative values
4580 for the mailbox size or file count. Odd effects could occur as a result.
4581 The maildirsize information is now recalculated if the size or filecount
4584 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4585 support for a long time. Removed HAVE_SYS_VFS_H.
4587 28. Installed the latest version of exipick from John Jetmore.
4589 29. In an address list, if the pattern was not a regular expression, an empty
4590 subject address (from a bounce message) matched only if the pattern was an
4591 empty string. Non-empty patterns were not even tested. This was the wrong
4592 because it is perfectly reasonable to use an empty address as part of a
4593 database query. An empty address is now tested by patterns that are
4594 lookups. However, all the other forms of pattern expect the subject to
4595 contain a local part and a domain, and therefore, for them, an empty
4596 address still always fails if the pattern is not itself empty.
4598 30. Exim went into a mad DNS loop when attempting to do a callout where the
4599 host was specified on an smtp transport, and looking it up yielded more
4600 than one IP address.
4602 31. Re-factored the code for checking spool and log partition space into a
4603 function that finds that data and another that does the check. The former
4604 is then used to implement four new variables: $spool_space, $log_space,
4605 $spool_inodes, and $log_inodes.
4607 32. The RFC2047 encoding function was originally intended for short strings
4608 such as real names; it was not keeping to the 75-character limit for
4609 encoded words that the RFC imposes. It now respects the limit, and
4610 generates multiple encoded words if necessary. To be on the safe side, I
4611 have increased the buffer size for the ${rfc2047: expansion operator from
4614 33. It is now permitted to omit both strings after an "if" condition; if the
4615 condition is true, the result is "true". As before, when the second string
4616 is omitted, a false condition yields an empty string. This makes it less
4617 cumbersome to write custom ACL and router conditions.
4619 34. Failure to deliver a bounce message always caused it to be frozen, even if
4620 there was an errors_to setting on the router. The errors_to setting is now
4623 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4624 canonical form (fully expanded) before being placed in
4625 $sender_host_address.
4627 36. The table in the code that translates DNS record types into text (T_A to
4628 "A" for instance) was missing entries for NS and CNAME. It is just possible
4629 that this could have caused confusion if both these types were looked up
4630 for the same domain, because the text type is used as part of Exim's
4631 per-process caching. But the chance of anyone hitting this buglet seems
4634 37. The dnsdb lookup has been extended in a number of ways.
4636 (1) There is a new type, "zns", which walks up the domain tree until it
4637 finds some nameserver records. It should be used with care.
4639 (2) There is a new type, "mxh", which is like "mx" except that it returns
4640 just the host names, not the priorities.
4642 (3) It is now possible to give a list of domains (or IP addresses) to be
4643 looked up. The behaviour when one of the lookups defers can be
4644 controlled by a keyword.
4646 (4) It is now possible to specify the separator character for use when
4647 multiple records are returned.
4649 38. The dnslists ACL condition has been extended: it is now possible to supply
4650 a list of IP addresses and/or domains to be looked up in a particular DNS
4653 39. Added log_selector=+queue_time_overall.
4655 40. When running the queue in the test harness, wait just a tad after forking a
4656 delivery process, to get repeatability of debugging output.
4658 41. Include certificate and key file names in error message when GnuTLS fails
4659 to set them up, because the GnuTLS error message doesn't include the name
4660 of the failing file when there is a problem reading it.
4662 42. Allow both -bf and -bF in the same test run.
4664 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4666 44. The "Exiscan patch" is now merged into the mainline Exim source.
4668 45. Sometimes the final signoff response after QUIT could fail to get
4669 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4670 < 0 before doing a fflush(). This bug looks as though it goes back to the
4671 introduction of TLS in release 3.20, but "sometimes" must have been rare
4672 because the tests only now provoked it.
4674 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4675 (this can affect the format of dates).
4677 47. exim_tidydb, when checking for the continued existence of a message for
4678 which it has found a message-specific retry record, was not finding
4679 messages that were in split spool directories. Consequently, it was
4680 deleting retry records that should have stayed in existence.
4682 48. Steve fixed some bugs in eximstats.
4684 49. The SPA authentication driver was not abandoning authentication and moving
4685 on to the next authenticator when an expansion was forced to fail,
4686 contradicting the general specification for all authenticators. Instead it
4687 was generating a temporary error. It now behaves as specified.
4689 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4690 (the order specifies the preference for clients). The order is now AES256,
4691 AES128, 3DES, ARCFOUR128.
4693 51. Small patch to Sieve code - explicitly set From: when generating an
4696 52. Exim crashed if a remote delivery caused a very long error message to be
4697 recorded - for instance if somebody sent an entire SpamAssassin report back
4698 as a large number of 550 error lines. This bug was coincidentally fixed by
4699 increasing the size of one of Exim's internal buffers (big_buffer) that
4700 happened as part of the Exiscan merge. However, to be on the safe side, I
4701 have made the code more robust (and fixed the comments that describe what
4704 53. Now that there can be additional text after "Completed" in log lines (if
4705 the queue_time_overall log selector is set), a one-byte patch to exigrep
4706 was needed to allow it to recognize "Completed" as not the last thing in
4709 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4710 patch that reportedly fixes this has been added. I am not expert enough to
4711 create a test for it. This is what the patch creator wrote:
4713 "I found a little strange behaviour of ldap code when working with
4714 Windows 2003 AD Domain, where users was placed in more than one
4715 Organization Units. When I tried to give exim partial DN, the exit code
4716 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4717 But simultaneously result of request was absolutely normal ldap result,
4718 so I produce this patch..."
4720 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4721 REFERENCE, so I have modified the code to exclude the patch when that macro
4724 55. Some experimental protocols are using DNS PTR records for new purposes. The
4725 keys for these records are domain names, not reversed IP addresses. The
4726 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4727 leaves it alone. Component reversal etc. now happens only for IP addresses.
4730 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4732 57. Double the size of the debug message buffer (to 2048) so that more of very
4733 long debug lines gets shown.
4735 58. The exicyclog utility now does better if the number of log files to keep
4736 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4738 59. Two changes related to the smtp_active_hostname option:
4740 (1) $smtp_active_hostname is now available as a variable.
4741 (2) The default for smtp_banner uses $smtp_active_hostname instead
4742 of $primary_hostname.
4744 60. The host_aton() function is supposed to be passed a string that is known
4745 to be a valid IP address. However, in the case of IPv6 addresses, it was
4746 not checking this. This is a hostage to fortune. Exim now panics and dies
4747 if the condition is not met. A case was found where this could be provoked
4748 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4749 components; fortuitously, this particular loophole had already been fixed
4750 by change 4.50/55 above.
4752 If there are any other similar loopholes, the new check in host_aton()
4753 itself should stop them being exploited. The report I received stated that
4754 data on the command line could provoke the exploit when Exim was running as
4755 exim, but did not say which command line option was involved. All I could
4756 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4757 running as the user.
4760 61. There was a buffer overflow vulnerability in the SPA authentication code
4761 (which came originally from the Samba project). I have added a test to the
4762 spa_base64_to_bits() function which I hope fixes it.
4765 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4766 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4768 63. The daemon start-up calls getloadavg() while still root for those OS that
4769 need the first call to be done as root, but it missed one case: when
4770 deliver_queue_load_max is set with deliver_drop_privilege. This is
4771 necessary for the benefit of the queue runner, because there is no re-exec
4772 when deliver_drop_privilege is set.
4774 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4775 This has been fixed.
4777 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4778 was in use, was not putting the data itself into the right store pool;
4779 consequently, it could be overwritten for a subsequent message in the same
4780 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4783 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4785 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4786 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4787 the function were treating the return as a boolean value, which happened to
4788 work because 0=false and not-0=true, but is not correct code.
4790 68. The host_aton() function was not handling scoped IPv6 addresses (those
4791 with, for example, "%eth0" on the end) correctly.
4793 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4794 negated items (that is, ~something) in unsigned ints. Some compilers
4795 apparently mutter when there is no cast.
4797 70. If an address verification called from an ACL failed, and did not produce a
4798 user-specific message (i.e. there was only a "system" message), nothing was
4799 put in $acl_verify_message. In this situation, it now puts the system
4802 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4803 session; change 4.31/43 added the unwanted input to the log line - except
4804 that it did not do this in the start of session case. It now does.
4806 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4807 This is wrong and can cause the other end to generate a synchronization
4808 error if it is another Exim or anything else that does the synchronization
4809 check. A QUIT command is no longer sent after a timeout.
4811 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4812 during host lookups.
4814 74. The defer_ok option of callout verification was not working if it was used
4815 when verifying addresses in header lines, that is, for this case:
4817 verify = header_sender/callout=defer_ok
4819 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4820 those file descriptors could be used for SMTP connections. If anything
4821 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4822 could be sent to the SMTP client, causing chaos. The daemon now opens
4823 stdin, stdout, and stderr to /dev/null when it puts itself into the
4826 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4827 log by default. The user can override this with suitable Perl magic.
4829 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4830 the log message when discard triggers, was not working for the DATA ACL or
4831 for the non-SMTP ACL.
4833 78. Error message wording change in sieve.c.
4835 79. If smtp_accept_max_per_host was set, the number of connections could be
4836 restricted to fewer than expected, because the daemon was trying to set up
4837 a new connection before checking whether the processes handling previous
4838 connections had finished. The check for completed processes is now done
4839 earlier. On busy systems, this bug wouldn't be noticed because something
4840 else would have woken the daemon, and it would have reaped the completed
4843 80. If a message was submitted locally by a user whose login name contained one
4844 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4845 It caused a spool format error. I have fixed the spool reading code. A
4846 related problem was that the "from" clause in the Received: line became
4847 illegal because of the space(s). It is now covered by ${quote_local_part.
4849 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4852 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4854 83. Patch from Sieve maintainer to fix unterminated string problem in
4855 "vacation" handling.
4857 84. Some minor changes to the Linux configuration files to help with other
4858 OS variants using glibc.
4860 85. One more patch for Sieve to update vacation handling to latest spec.
4863 ----------------------------------------------------
4864 See the note above about the 4.44 and 4.50 releases.
4865 ----------------------------------------------------
4871 1. Change 4.43/35 introduced a bug that caused file counts to be
4872 incorrectly computed when quota_filecount was set in an appendfile
4875 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4876 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4879 3. Give more explanation in the error message when the command for a transport
4880 filter fails to execute.
4882 4. There are several places where Exim runs a non-Exim command in a
4883 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4884 was being done only for the command run by the queryprogram router. It is
4885 now done for all such subprocesses. The other cases are: ${run, transport
4886 filters, and the commands run by the lmtp and pipe transports.
4888 5. Some older OS have a limit of 256 on the maximum number of file
4889 descriptors. Exim was using setrlimit() to set 1000 as a large value
4890 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4891 systems. I've change it so that if it can't get 1000, it tries for 256.
4893 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4894 was an oversight, and furthermore, ever since the addition of extra
4895 controls (e.g. 4.43/32), the checks on when to allow different forms of
4896 "control" were broken. There should now be diagnostics for all cases when a
4897 control that does not make sense is encountered.
4899 7. $recipients is now available in the predata ACL (oversight).
4901 8. Tidy the search cache before the fork to do a delivery from a message
4902 received from the command line. Otherwise the child will trigger a lookup
4903 failure and thereby defer the delivery if it tries to use (for example) a
4904 cached ldap connection that the parent has called unbind on.
4906 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4907 of $address_data from the recipient verification was clobbered by the
4908 sender verification.
4910 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4911 was its contents. (It was OK if the option was not defined at all.)
4913 11. A "Completed" log line is now written for messages that are removed from
4914 the spool by the -Mrm option.
4916 12. $host_address is now set to the target address during the checking of
4917 ignore_target_hosts.
4919 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4920 being passed; this would have caused $sender_host_name to have been used if
4921 matching the list had actually called for a host name (not very likely,
4922 since this list is usually IP addresses). A host name is now passed as
4925 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4926 code that specifies a non-restarting handler (typically sigaction() in
4927 modern systems) in an attempt to fix a rare and obscure crash bug.
4929 15. Narrowed the window for a race in the daemon that could cause it to ignore
4930 SIGCHLD signals. This is not a major problem, because they are used only to
4931 wake it up if nothing else does.
4933 16. A malformed maildirsize file could cause Exim to calculate negative values
4934 for the mailbox size or file count. Odd effects could occur as a result.
4935 The maildirsize information is now recalculated if the size or filecount
4938 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4939 support for a long time. Removed HAVE_SYS_VFS_H.
4941 18. Updated exipick to current release from John Jetmore.
4943 19. Allow an empty sender to be matched against a lookup in an address list.
4944 Previously the only cases considered were a regular expression, or an
4947 20. Exim went into a mad DNS lookup loop when doing a callout where the
4948 host was specified on the transport, if the DNS lookup yielded more than
4951 21. The RFC2047 encoding function was originally intended for short strings
4952 such as real names; it was not keeping to the 75-character limit for
4953 encoded words that the RFC imposes. It now respects the limit, and
4954 generates multiple encoded words if necessary. To be on the safe side, I
4955 have increased the buffer size for the ${rfc2047: expansion operator from
4958 22. Failure to deliver a bounce message always caused it to be frozen, even if
4959 there was an errors_to setting on the router. The errors_to setting is now
4962 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4963 canonical form (fully expanded) before being placed in
4964 $sender_host_address.
4966 24. Updated eximstats to version 1.33
4968 25. Include certificate and key file names in error message when GnuTLS fails
4969 to set them up, because the GnuTLS error message doesn't include the name
4970 of the failing file when there is a problem reading it.
4972 26. Expand error message when OpenSSL has problems setting up cert/key files.
4975 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4976 (this can affect the format of dates).
4978 28. exim_tidydb, when checking for the continued existence of a message for
4979 which it has found a message-specific retry record, was not finding
4980 messages that were in split spool directories. Consequently, it was
4981 deleting retry records that should have stayed in existence.
4983 29. eximstats updated to version 1.35
4984 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4985 1.35 - bugfix such that pie charts by volume are generated correctly
4987 30. The SPA authentication driver was not abandoning authentication and moving
4988 on to the next authenticator when an expansion was forced to fail,
4989 contradicting the general specification for all authenticators. Instead it
4990 was generating a temporary error. It now behaves as specified.
4992 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4993 (the order specifies the preference for clients). The order is now AES256,
4994 AES128, 3DES, ARCFOUR128.
4996 31. Small patch to Sieve code - explicitly set From: when generating an
4999 32. Exim crashed if a remote delivery caused a very long error message to be
5000 recorded - for instance if somebody sent an entire SpamAssassin report back
5001 as a large number of 550 error lines. This bug was coincidentally fixed by
5002 increasing the size of one of Exim's internal buffers (big_buffer) that
5003 happened as part of the Exiscan merge. However, to be on the safe side, I
5004 have made the code more robust (and fixed the comments that describe what
5007 33. Some experimental protocols are using DNS PTR records for new purposes. The
5008 keys for these records are domain names, not reversed IP addresses. The
5009 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5010 leaves it alone. Component reversal etc. now happens only for IP addresses.
5013 34. The host_aton() function is supposed to be passed a string that is known
5014 to be a valid IP address. However, in the case of IPv6 addresses, it was
5015 not checking this. This is a hostage to fortune. Exim now panics and dies
5016 if the condition is not met. A case was found where this could be provoked
5017 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5018 components; fortuitously, this particular loophole had already been fixed
5019 by change 4.50/55 or 4.44/33 above.
5021 If there are any other similar loopholes, the new check in host_aton()
5022 itself should stop them being exploited. The report I received stated that
5023 data on the command line could provoke the exploit when Exim was running as
5024 exim, but did not say which command line option was involved. All I could
5025 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5026 running as the user.
5029 35. There was a buffer overflow vulnerability in the SPA authentication code
5030 (which came originally from the Samba project). I have added a test to the
5031 spa_base64_to_bits() function which I hope fixes it.
5034 36. The daemon start-up calls getloadavg() while still root for those OS that
5035 need the first call to be done as root, but it missed one case: when
5036 deliver_queue_load_max is set with deliver_drop_privilege. This is
5037 necessary for the benefit of the queue runner, because there is no re-exec
5038 when deliver_drop_privilege is set.
5040 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5041 was in use, was not putting the data itself into the right store pool;
5042 consequently, it could be overwritten for a subsequent message in the same
5043 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5046 38. Sometimes the final signoff response after QUIT could fail to get
5047 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5048 < 0 before doing a fflush(). This bug looks as though it goes back to the
5049 introduction of TLS in release 3.20, but "sometimes" must have been rare
5050 because the tests only now provoked it.
5056 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5057 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5058 Now it returns an integer. A number of calls were still expecting a T/F
5059 return. Fortuitously, in all cases, the tests worked in OK situations,
5060 which is the norm. However, things would have gone wrong on any write
5061 failures on the smtp file descriptor. This function is used when sending
5062 messages over SMTP and also when doing verify callouts.
5064 2. When Exim is called to do synchronous delivery of a locally submitted
5065 message (the -odf or -odi options), it no longer closes stderr before doing
5068 3. Implemented the mua_wrapper option.
5070 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5072 5. Implemented the functions header_remove(), header_testname(),
5073 header_add_at_position(), and receive_remove_recipient(), and exported them
5076 6. If an ACL "warn" statement specified the addition of headers, Exim already
5077 inserted X-ACL-Warn: at the start if there was no header name. However, it
5078 was not making this test for the second and subsequent header lines if
5079 there were newlines in the string. This meant that an invalid header could
5080 be inserted if Exim was badly configured.
5082 7. Allow an ACL "warn" statement to add header lines at the start or after all
5083 the Received: headers, as well as at the end.
5085 8. Added the rcpt_4xx retry error code.
5087 9. Added postmaster_mailfrom=xxx to callout verification option.
5089 10. Added mailfrom=xxxx to the callout verification option, for verify=
5092 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5093 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5095 12. Inserted some casts to stop certain compilers warning when using pointer
5096 differences as field lengths or precisions in printf-type calls (mostly
5097 affecting debugging statements).
5099 13. Added optional readline() support for -be (dynamically loaded).
5101 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5102 same clock tick as a message's arrival, so that its received time was the
5103 same as the "first fail" time on the retry record, and that message
5104 remained on the queue past the ultimate address timeout, every queue runner
5105 would try a delivery (because it was past the ultimate address timeout) but
5106 after another failure, the ultimate address timeout, which should have then
5107 bounced the address, did not kick in. This was a "< instead of <=" error;
5108 in most cases the first failure would have been in the next clock tick
5109 after the received time, and all would be well.
5111 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5112 being recognized when the domain list was tested by the match_domain
5113 condition in an expansion string.
5115 16. Added the ${str2b64: operator.
5117 17. Exim was always calling setrlimit() to set a large limit for the number of
5118 processes, without checking whether the existing limit was already
5119 adequate. (It did check for the limit on file descriptors.) Furthermore,
5120 errors from getrlimit() and setrlimit() were being ignored. Now they are
5121 logged to the main and panic logs, but Exim does carry on, to try to do its
5122 job under whatever limits there are.
5124 18. Imported PCRE 5.0.
5126 19. Trivial typo in log message " temporarily refused connection" (the leading
5129 20. If the log selector return_path_on_delivery was set and an address was
5130 redirected to /dev/null, the delivery process crashed because it assumed
5131 that a return path would always be set for a "successful" delivery. In this
5132 case, the whole delivery is bypassed as an optimization, and therefore no
5135 21. Internal re-arrangement: the function for sending a challenge and reading
5136 a response while authentication was assuming a zero-terminated challenge
5137 string. It's now changed to take a pointer and a length, to allow for
5138 binary data in such strings.
5140 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5142 23. Exim was not respecting finduser_retries when seeking the login of the
5143 uid under which it was called; it was always trying 10 times. (The default
5144 setting of finduser_retries is zero.) Also, it was sleeping after the final
5145 failure, which is pointless.
5147 24. Implemented tls_on_connect_ports.
5149 25. Implemented acl_smtp_predata.
5151 26. If the domain in control=submission is set empty, Exim assumes that the
5152 authenticated id is a complete email address when it generates From: or
5153 Sender: header lines.
5155 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5156 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5157 chown and chgrp in /bin and hostname in /usr/bin.
5159 28. Exim was keeping the "process log" file open after each use, just as it
5160 does for the main log. This opens the possibility of it remaining open for
5161 long periods when the USR1 signal hits a daemon. Occasional processlog
5162 errors were reported, that could have been caused by this. Anyway, it seems
5163 much more sensible not to leave this file open at all, so that is what now
5166 29. The long-running daemon process does not normally write to the log once it
5167 has entered its main loop, and it closes the log before doing so. This is
5168 so that log files can straightforwardly be renamed and moved. However,
5169 there are a couple of unusual error situations where the daemon does write
5170 log entries, and I had neglected to close the log afterwards.
5172 30. The text of an SMTP error response that was received during a remote
5173 delivery was being truncated at 512 bytes. This is too short for some of
5174 the long messages that one sometimes sees. I've increased the limit to
5177 31. It is now possible to make retry rules that apply only when a message has a
5178 specific sender, in particular, an empty sender.
5180 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5181 it possible to be selective about when SMTP synchronization is enforced.
5183 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5185 32. Implemented hosts_connection_nolog.
5187 33. Added an ACL for QUIT.
5189 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5192 35. Added mailbox_size and mailbox_filecount to appendfile.
5194 36. Added control = no_multiline_responses to ACLs.
5196 37. There was a bug in the logic of the code that waits for the clock to tick
5197 in the case where the clock went backwards by a substantial amount such
5198 that the microsecond fraction of "now" was more than the microsecond
5199 fraction of "then" (but the whole seconds number was less).
5201 38. Added support for the libradius Radius client library this is found on
5202 FreeBSD (previously only the radiusclient library was supported).
5208 1. When certain lookups returned multiple values in the form name=value, the
5209 quoting of the values was not always being done properly. Specifically:
5210 (a) If the value started with a double quote, but contained no whitespace,
5212 (b) If the value contained whitespace other than a space character (i.e.
5213 tabs or newlines or carriage returns) it was not quoted.
5214 This fix has been applied to the mysql and pgsql lookups by writing a
5215 separate quoting function and calling it from the lookup code. The fix
5216 should probably also be applied to nisplus, ibase and oracle lookups, but
5217 since I cannot test any of those, I have not disturbed their existing code.
5219 2. A hit in the callout cache for a specific address caused a log line with no
5220 reason for rejecting RCPT. Now it says "Previous (cached) callout
5221 verification failure".
5223 3. There was an off-by-one bug in the queryprogram router. An over-long
5224 return line was truncated at 256 instead of 255 characters, thereby
5225 overflowing its buffer with the terminating zero. As well as fixing this, I
5226 have increased the buffer size to 1024 (and made a note to document this).
5228 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5229 when Exim is waiting for an SMTP response from a remote server, Exim
5230 restarts its select() call on the socket, thereby resetting its timeout.
5231 This is not a problem when such interrupts are rare. Somebody set up a cron
5232 job to run exiwhat every 2 minutes, which is less than the normal select()
5233 timeout (5 or 10 minutes). This meant that the select() timeout never
5234 kicked in because it was always reset. I have fixed this by comparing the
5235 time when an interrupt arrives with the time at the start of the first call
5236 to select(). If more time than the timeout has elapsed, the interrupt is
5237 treated as a timeout.
5239 5. Some internal re-factoring in preparation for the addition of Sieve
5240 extensions (by MH). In particular, the "personal" test is moved to a
5241 separate function, and given an option for scanning Cc: and Bcc: (which is
5242 not set for Exim filters).
5244 6. When Exim created an email address using the login of the caller as the
5245 local part (e.g. when creating a From: or Sender: header line), it was not
5246 quoting the local part when it contained special characters such as @.
5248 7. Installed new OpenBSD configuration files.
5250 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5251 try to make them clearer.
5253 9. Callout options, other than the timeout value, were being ignored when
5254 verifying sender addresses in header lines. For example, when using
5256 verify = header_sender/callout=no_cache
5258 the cache was (incorrectly) being used.
5260 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5261 only the Cygwin environment.
5263 11. When return_path_on_delivery was set as a log selector, if different remote
5264 addresses in the same message used different return paths and parallel
5265 remote delivery occurred, the wrong values would sometimes be logged.
5266 (Whenever a remote delivery process finished, the return path value from
5267 the most recently started remote delivery process was logged.)
5269 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5270 lines when AUTH and/or TLS are in use. This is the "received protocol"
5271 field. Exim used to use "asmtp" for authenticated SMTP, without any
5272 indication (in the protocol name) for TLS use. Now it follows the RFC and
5273 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5274 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5275 names appear in log lines as well as in Received: header lines.
5277 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5278 extensions, and comparison tests, and to fix some bugs.
5280 14. Changes to the "personal" filter test:
5282 (1) The test was buggy in that it was just doing the equivalent of
5283 "contains" tests on header lines. For example, if a user's address was
5284 anne@some.where, the "personal" test would incorrectly be true for
5286 To: susanne@some.where
5288 This test is now done by extracting each address from the header in turn,
5289 and checking the entire address. Other tests that are part of "personal"
5290 are now done using regular expressions (for example, to check local parts
5291 of addresses in From: header lines).
5293 (2) The list of non-personal local parts in From: addresses has been
5294 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5295 taken from the Sieve specification recommendations.
5297 (3) If the message contains any header line starting with "List-" it is
5298 treated as non-personal.
5300 (4) The test for "circular" in the Subject: header line has been removed
5301 because it now seems ill-conceived.
5303 15. Minor typos in src/EDITME comments corrected.
5305 16. Installed latest exipick from John Jetmore.
5307 17. If headers_add on a router specified a text string that was too long for
5308 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5309 of string_sprintf() is now avoided.
5311 18. $message_body_size was not set (it was always zero) when running the DATA
5312 ACL and the local_scan() function.
5314 19. For the "mail" command in an Exim filter, no default was being set for
5315 the once_repeat time, causing a random time value to be used if "once" was
5316 specified. (If the value happened to be <= 0, no repeat happened.) The
5317 default is now 0s, meaning "never repeat". The "vacation" command was OK
5318 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5319 (I found it when inspecting the code).
5321 20. There is now an overall timeout for performing a callout verification. It
5322 defaults to 4 times the callout timeout, which applies to individual SMTP
5323 commands during the callout. The overall timeout applies when there is more
5324 than one host that can be tried. The timeout is checked before trying the
5325 next host. This prevents very long delays if there are a large number of
5326 hosts and all are timing out (e.g. when the network connections are timing
5327 out). The value of the overall timeout can be changed by specifying an
5328 additional sub-option for "callout", called "maxwait". For example:
5330 verify = sender/callout=5s,maxwait=20s
5332 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5333 to the end before writing, but this should make it even safer).
5335 22. Exim was forgetting that it had advertised PIPELINING for the second and
5336 subsequent messages on an SMTP connection. It was also not resetting its
5337 memory on STARTTLS and an internal HELO.
5339 23. When Exim logs an SMTP synchronization error within a session, it now
5340 records whether PIPELINING has been advertised or not.
5342 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5343 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5344 rather than long int.
5346 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5348 26. Added the never_mail option to autoreply.
5354 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5355 crash if the getsockname() call failed; this can happen if a connection is
5356 closed very soon after it is established. The problem was simply in the
5357 order in which certain operations were done, causing Exim to try to write
5358 to the SMTP stream before it had set up the file descriptor. The bug has
5359 been fixed by making things happen in the correct order.
5365 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5366 before the connection was closed, thus losing the rejection response.
5368 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5369 some early Solaris releases, but causes trouble in current releases where
5370 socklen_t is defined.
5372 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5375 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5378 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5379 that releasing the top of it at the end releases what was used for sub-
5380 expansions (unless the block got too big). However, discard this block if
5381 the first thing is a variable or header, so that we can use its block when
5382 it is dynamic (useful for very large $message_headers, for example).
5384 6. Lookups now cache *every* query, not just the most recent. A new, separate
5385 store pool is used for this. It can be recovered when all lookup caches are
5386 flushed. Lookups now release memory at the end of their result strings.
5387 This has involved some general refactoring of the lookup sources.
5389 7. Some code has been added to the store_xxx() functions to reduce the amount
5390 of flapping under certain conditions.
5392 8. log_incoming_interface used to affect only the <= reception log lines. Now
5393 it causes the local interface and port to be added to several more SMTP log
5394 lines, for example "SMTP connection from", and rejection lines.
5396 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5398 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5400 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5401 could be overwritten at the end of the current message (or the start of a
5402 new message if it was set in a HELO ACL). The value is now preserved for
5403 the duration of the SMTP connection.
5405 12. If a transport had a headers_rewrite setting, and a matching header line
5406 contained an unqualified address, that address was qualified, even if it
5407 did not match any rewriting rules. The underlying bug was that the values
5408 of the flags that permit the existence of unqualified sender and recipient
5409 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5410 non-local messages, and by -bnq for local messages) were not being
5411 preserved with the message after it was received.
5413 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5414 "next input=" as part of the text comprising the host identity instead of
5415 the correct text. The code was using the same buffer for two different
5416 strings. However, depending on which order the printing function evaluated
5417 its arguments, the bug did not always show up. Under Linux, for example, my
5418 test suite worked just fine.
5420 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5421 doesn't work with some older versions of Perl. It has been changed to "my",
5422 which in any case is probably the better facility to use.
5424 15. A really picky compiler found some instances of statements for creating
5425 error messages that either had too many or two few arguments for the format
5428 16. The size of the buffer for calls to the DNS resolver has been increased
5429 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5430 for addresses that have a lot of PTR records. This alleviates a problem; it
5431 does not fully solve it.
5433 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5434 buffer now truncates the list and logs the incident, which is the same
5435 action as happens when Exim is looking up a host name and its aliases.
5436 Previously in this situation something unpredictable would happen;
5437 sometimes it was "internal error: store_reset failed".
5439 18. If a server dropped the connection unexpectedly when an Exim client was
5440 using GnuTLS and trying to read a response, the client delivery process
5441 crashed while trying to generate an error log message.
5443 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5444 string, for example:
5446 warn message = H1: something\nH2: something
5448 the text was added as a single header line from Exim's point of view
5449 though it ended up OK in the delivered message. However, searching for the
5450 second and subsequent header lines using $h_h2: did not work. This has been
5451 fixed. Similarly, if a system filter added multiple headers in this way,
5452 the routers could not see them.
5454 20. Expanded the error message when iplsearch is called with an invalid key to
5455 suggest using net-iplsearch in a host list.
5457 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5458 ACLs are no longer actually imposed (and a message to that effect is
5461 22. If a "gecos" field in a passwd entry contained escaped characters, in
5462 particular, if it contained a \" sequence, Exim got it wrong when building
5463 a From: or a Sender: header from that name. A second bug also caused
5464 incorrect handling when an unquoted " was present following a character
5465 that needed quoting.
5467 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5468 was not being matched caselessly.
5470 24. Arranged for all hyphens in the exim.8 source to be escaped with
5473 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5474 properly. Recipient callout cache records were still being keyed to include
5475 the sender, even when use_sender was set false. This led to far more
5476 callouts that were necessary. The sender is no longer included in the key
5477 when use_sender is false.
5479 26. Added "control = submission" modifier to ACLs.
5481 27. Added the ${base62d: operator to decode base 62 numbers.
5483 28. dnsdb lookups can now access SRV records.
5485 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5486 the configuration file.
5488 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5489 (-v) mode. This makes the output for a verbose queue run more intelligible.
5491 31. Added a use_postmaster feature to recipient callouts.
5493 32. Added the $body_zerocount variable, containing the number of binary zero
5494 bytes in the message body.
5496 33. The time of last modification of the "new" subdirectory is now used as the
5497 "mailbox time last read" when there is a quota error for a maildir
5500 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5502 35. Added +ignore_unknown as a special item in host lists.
5504 36. Code for decoding IPv6 addresses in host lists is now included, even if
5505 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5506 address was recognized as an IP address, but was then not correctly decoded
5507 into binary, causing unexpected and incorrect effects when compared with
5514 1. Very minor rewording of debugging text in manualroute to say "list of
5515 hosts" instead of "hostlist".
5517 2. If verify=header_syntax was set, and a header line with an unqualified
5518 address (no domain) and a large number of spaces between the end of the
5519 name and the colon was received, the reception process suffered a buffer
5520 overflow, and (when I tested it) crashed. This was caused by some obsolete
5521 code that should have been removed. The fix is to remove it!
5523 3. When running in the test harness, delay a bit after writing a bounce
5524 message to get a bit more predictability in the log output.
5526 4. Added a call to search_tidyup() just before forking a reception process. In
5527 theory, someone could use a lookup in the expansion of smtp_accept_max_
5528 per_host which, without the tidyup, could leave open a database connection.
5530 5. Added the variables $recipient_data and $sender_data which get set from a
5531 lookup success in an ACL "recipients" or "senders" condition, or a router
5532 "senders" option, similar to $domain_data and $local_part_data.
5534 6. Moved the writing of debug_print from before to after the "senders" test
5537 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5538 problems for message scanning, either using a data ACL, or using
5539 local_scan() because the Received: header was not generated till after they
5540 were called (in order to set the time as the time of reception completion).
5541 I have revised the way this works. The header is now generated after the
5542 body is received, but before the ACL or local_scan() are called. After they
5543 are run, the timestamp in the header is updated.
5549 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5550 before starting a queue runner without re-exec. This happened only when
5551 deliver_drop_privilege was set or when the Exim user was set to root. The
5552 effect of the bug was that timeouts during subsequent deliveries caused
5553 crashes instead of being properly handled. The handler is now left at its
5554 default (and expected) setting.
5556 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5557 message, again when deliver_drop_privilege is set or Exim is run as root.
5558 The bug described in (1) was not present in this case, but the tidying up
5559 of the other signals was missing. I have made the two cases consistent.
5561 3. The ignore_target_hosts setting on a manualroute router was being ignored
5562 for hosts that were looked up using the /MX notation.
5564 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5567 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5568 operated on the sender address. After changing the $sender_address to <>
5569 for the sender address verify, Exim was re-instated it as the original
5570 (before rewriting) address, but remembering that it had rewritten it, so it
5571 wasn't rewriting it again. This bug also had the effect of breaking the
5572 sender address verification caching when the sender address was rewritten.
5574 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5575 This has been changed so that if the ip literal address matches
5576 ignore_target_hosts, the router declines.
5578 7. Added expansion conditions match_domain, match_address, and match_local_
5579 part (NOT match_host).
5581 8. The placeholder for the Received: header didn't have a length field set.
5583 9. Added code to Exim itself and to exim_lock to test for a specific race
5584 condition that could lead to file corruption when using MBX delivery. The
5585 issue is with the lockfile that is created in /tmp. If this file is removed
5586 after a process has opened it but before that process has acquired a lock,
5587 there is the potential for a second process to recreate the file and also
5588 acquire a lock. This could lead to two Exim processes writing to the file
5589 at the same time. The added code performs the same test as UW imapd; it
5590 checks after acquiring the lock that its file descriptor still refers to
5591 the same named file.
5593 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5594 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5597 11. Added the smtp_active_hostname option. If used, this will typically be made
5598 to depend on the incoming interface address. Because $interface_address is
5599 not set up until the daemon has forked a reception process, error responses
5600 that can happen earlier (such as "too many connections") no longer contain
5603 12. If an expansion in a condition on a "warn" statement fails because a lookup
5604 defers, the "warn" statement is abandoned, and the next ACL statement is
5605 processed. Previously this caused the whole ACL to be aborted.
5607 13. Added the iplsearch lookup type.
5609 14. Added ident_timeout as a log selector.
5611 15. Added tls_certificate_verified as a log selector.
5613 16. Added a global option tls_require_ciphers (compare the smtp transport
5614 option of the same name). This controls incoming TLS connections.
5616 17. I finally figured out how to make tls_require_ciphers do a similar thing
5617 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5618 before starting the TLS session.
5620 18. Tabs are now shown as \t in -bP output.
5622 19. If the log selector return_path_on_delivery was set, Exim crashed when
5623 bouncing a message because it had too many Received: header lines.
5625 20. If two routers both had headers_remove settings, and the first one included
5626 a superfluous trailing colon, the final name in the first list and the
5627 first name in the second list were incorrectly joined into one item (with a
5628 colon in the middle).
5634 1. Added -C and -D options to the exinext utility, mainly to make it easier
5635 to include in the automated testing, but these could be helpful when
5636 multiple configurations are in use.
5638 2. The exinext utility was not formatting the output nicely when there was
5639 an alternate port involved in the retry record key, nor when there was a
5640 message id as well (for retries that were specific to a specific message
5641 and a specific host). It was also confused by IPv6 addresses, because of
5642 the additional colons they contain. I have fixed the IPv4 problem, and
5643 patched it up to do a reasonable job for IPv6.
5645 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5646 delivery, the log line now contains "pipelined" if PIPELINING was used.
5648 4. An SMTP transport process used to panic and die if the bind() call to set
5649 an explicit outgoing interface failed. This has been changed; it is now
5650 treated in the same way as a connect() failure.
5652 5. A reference to $sender_host_name in the part of a conditional expansion
5653 that was being skipped was still causing a DNS lookup. This no longer
5656 6. The def: expansion condition was not recognizing references to header lines
5657 that used bh_ and bheader_.
5659 7. Added the _cache feature to named lists.
5661 8. The code for checking quota_filecount in the appendfile transport was
5662 allowing one more file than it should have been.
5664 9. For compatibility with Sendmail, the command line option
5672 and sets the incoming protocol and host name (for trusted callers). The
5673 host name and its colon can be omitted when only the protocol is to be set.
5674 Note the Exim already has two private options, -pd and -ps, that refer to
5675 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5676 "s", but I don't think that's a major issue.
5678 10. A number of refactoring changes to the code, none of which should affect
5681 (a) The number of logging options was getting close to filling up the
5682 32-bit word that was used as a bit map. I have split them into two classes:
5683 those that are passed in the argument to log_write(), and those that are
5684 only ever tested independently outside of that function. These are now in
5685 separate 32-bit words, so there is plenty of room for expansion again.
5686 There is no change in the user interface or the logging behaviour.
5688 (b) When building, for example, log lines, the code previously used a
5689 macro that called string_cat() twice, in order to add two strings. This is
5690 not really sufficiently general. Furthermore, there was one instance where
5691 it was actually wrong because one of the argument was used twice, and in
5692 one call a function was used. (As it happened, calling the function twice
5693 did not affect the overall behaviour.) The macro has been replaced by a
5694 function that can join an arbitrary number of extra strings onto a growing
5697 (c) The code for expansion conditions now uses a table and a binary chop
5698 instead of a serial search (which was left over from when there were very
5699 few conditions). Also, it now recognizes conditions like "pam" even when
5700 the relevant support is not compiled in: a suitably worded error message is
5701 given if an attempt is made to use such a condition.
5703 11. Added ${time_interval:xxxxx}.
5705 12. A bug was causing one of the ddress fields not to be passed back correctly
5706 from remote delivery subprocesses. The field in question was not being
5707 subsequently used, so this caused to problems in practice.
5709 13. Added new log selectors queue_time and deliver_time.
5711 14. Might have fixed a bug in maildirsizefile handling that threw up
5712 "unexpected character" debug warnings, and recalculated the data
5713 unnecessarily. In any case, I expanded the warning message to give more
5716 15. Added the message "Restricted characters in address" to the statements in
5717 the default ACL that block characters like @ and % in local parts.
5719 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5720 Three changes have been made:
5722 (a) There was a serious bug; a negative response to MAIL caused the whole
5723 recipient domain to be cached as invalid, thereby blocking all messages
5724 to all local parts at the same domain, from all senders. This bug has
5725 been fixed. The domain is no longer cached after a negative response to
5726 MAIL if the sender used is not empty.
5728 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5731 (c) A new callout option, "use_sender" has been added for people who want
5732 the modified behaviour.
5738 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5741 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5742 indeed breaks things for older releases.
5744 3. Added additional logging to the case where there is a problem reading data
5745 from a filter that is running in a subprocess using a pipe, in order to
5746 try to track down a specific problem.
5748 4. Testing facility fudge: when running in the test harness and attempting
5749 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5750 getting "No route to host". Convert this to a timeout.
5752 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5755 6. Some OS don't have socklen_t but use size_t instead. This affects the
5756 fifth argument of getsockopt() amongst other things. This is now
5757 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5758 can be set for individual OS. I have set it for SunOS5, OSF1, and
5759 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5760 some earlier ones do not.
5762 7. Change 4.30/15 was not doing the test caselessly.
5764 8. The standard form for an IPv6 address literal was being rejected by address
5765 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5766 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5767 this, as well as the form without the "IPv6" on the front (but only when
5768 address literals are enabled, of course).
5770 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5772 10. Exim crashed if a message with an empty sender address specified by -f
5773 encountered a router with an errors_to setting. This could be provoked only
5774 by a command such as
5778 where an empty string was supplied; "<>" did not hit this bug.
5780 11. Installed PCRE release 4.5.
5782 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5783 remained set. It is now erased.
5785 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5786 times from message ids (which are base 36 rather than the normal 62).
5788 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5789 were being counted as actual protocol errors, and logged if the log
5790 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5791 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5792 and DATA following a set of rejected RCPTs do not count as protocol errors.
5793 In other words, Exim assumes they were pipelined, though this may not
5794 actually be the case. Of course, in all cases the client gets an
5795 appropriate error code.
5797 15. If a lookup fails in an ACL condition, a message about the failure may
5798 be available; it is used if testing the ACL cannot continue, because most
5799 such messages specify what the cause of the deferral is. However, some
5800 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5801 that caused an old message to be retained and used if a later statement
5802 caused a defer, replacing the real cause of the deferral.
5804 16. If an IP address had so many PTR records that the DNS lookup buffer
5805 was not large enough to hold them, Exim could crash while trying to process
5806 the truncated data. It now detects and logs this case.
5808 17. Further to 4.21/58, another change has been made: if (and only if) the
5809 first line of a message (the first header line) ends with CRLF, a bare LF
5810 in a subsequent header line has a space inserted after it, so as not to
5811 terminate the header.
5813 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5814 unnecessarily, used atoi() instead of strtol(), and didn't check the
5815 termination when getting file sizes from file names by regex.
5817 19. Completely re-implemented the support for maildirsize files, in the light
5818 of a number of problems with the previous contributed implementation
5819 (4.30/29). In particular:
5821 . If the quota is zero, the maildirsize file is maintained, but no quota is
5824 . If the maildir directory does not exist, it is created before any attempt
5825 to write a maildirsize file.
5827 . The quota value in the file is just a cache; if the quota is changed in
5828 the transport, the new value overrides.
5830 . A regular expression is available for excluding directories from the
5833 20. The autoreply transport checks the characters in options that define the
5834 message's headers; it allows continued headers, but it was checking with
5835 isspace() after an embedded newline instead of explicitly looking for a
5838 21. If all the "regular" hosts to which an address was routed had passed their
5839 expiry times, and had not reached their retry times, the address was
5840 bounced, even if fallback hosts were defined. Now Exim should go on to try
5843 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5844 equivalent code in the SMTP transport. Some hosts send humungous responses
5845 to HELO/EHLO, more than 1024 it seems.
5847 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5848 gives compiler warnings in some environments. I've now done it "properly",
5851 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5852 (because of problems with the built-in one) was declared to return uschar *
5853 instead of char *, causing compiler failure.
5855 25. Fixed a file descriptor leak when processing alias/forward files.
5857 26. Fixed a minor format string issue in dbfn.c.
5859 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5861 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5862 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5865 29. When a sender address is verified, it is cached, to save repeating the test
5866 when there is more than one recipient in a message. However, when the
5867 verification involves a callout, it is possible for different callout
5868 options to be set for different recipients. It is too complicated to keep
5869 track of this in the cache, so now Exim always runs a verification when a
5870 callout is required, relying on the callout cache for the optimization.
5871 The overhead is duplication of the address routing, but this should not be
5874 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5875 to be verified with callout=postmaster, and the main callout worked but the
5876 postmaster check failed, the verification correctly failed. However, if a
5877 subsequent RCPT command asked for sender verification *without* the
5878 postmaster check, incorrect caching caused this verification also to fail,
5881 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5882 it was not caching the DNS options (qualify_single, search_parents) that
5883 were used when the lookup failed. A subsequent lookup with different
5884 options therefore always gave the same answer, though there were cases
5885 where it should not have. (Example: a "domains = !$mx_any" option on a
5886 dnslookup router: the "domains" option is always processed without any
5887 widening, but the router might have qualify_single set.) Now Exim uses the
5888 cached value only when the same options are set.
5890 32. Added John Jetmore's "exipick" utility to the distribution.
5892 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5893 than a timeout (e.g. a certificate is required, and is not provided), an
5894 Exim server now closes the connection immediately. Previously it waited for
5895 the client to close - but if the client is SSL, it seems that they each
5896 wait for each other, leading to a delay before one of them times out.
5898 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5899 maintained 0.8.x compatibility because I don't think many are using it, and
5900 it is clearly obsolete.
5902 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5905 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5906 client certificate was expired. A simple patch fixes this, though I don't
5907 understand the full logic of why the verify callback is called multiple
5910 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5911 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5912 which causes problems with some clients (such as the Certicom SSL Plus
5913 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5914 disables the coutermeasure allowing Eudora to connect."
5916 38. Exim was not checking that a write() to a log file succeeded. This could
5917 lead to Bad Things if a log got too big, in particular if it hit a file
5918 size limit. Exim now panics and dies if it cannot write to a log file, just
5919 as it does if it cannot open a log file.
5921 39. Modified OS/Makefile-Linux so that it now contains
5923 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5925 The two -D definitions ensure that Exim is compiled with large file
5926 support, which makes it possible to handle log files that are bigger than
5929 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5930 instance) a domain was checked against a named list that involved a lookup,
5931 causing $domain_data to be set, then another domain was checked against the
5932 same list, then the first domain was re-checked, the value of $domain_data
5933 after the final check could be wrong. In particular, if the second check
5934 failed, it could be set empty. This bug probably also applied to
5937 41. The strip_trailing_dot option was not being applied to the address given
5938 with the -f command-line option.
5940 42. The code for reading a message's header from the spool was incrementing
5941 $received_count, but never initializing it. This meant that the value was
5942 incorrect (doubled) while delivering a message in the same process in which
5943 it was received. In the most common configuration of Exim, this never
5944 happens - a fresh exec is done - but it can happen when
5945 deliver_drop_privilege is set.
5947 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5948 it now includes up to 150 characters of the unexpected data in the log
5951 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5952 and building data strings. The size of both of these buffers was 10 000
5953 bytes - far larger than anybody would *ever* want, thought I. Needless to
5954 say, somebody hit the limit. I have increased the maximum line length to
5955 20 000 and the maximum data length of concatenated lines to 100 000. I have
5956 also fixed two bugs, because there was no checking on these buffers. Tsk,
5957 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5958 buffer is too small.
5960 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5961 lsearch lookups. Now it does.
5963 46. When parsing a route_list item in a manualroute router, a fixed-length
5964 buffer was used for the list of hosts. I made this 1024 bytes long,
5965 thinking that nobody would ever have a list of hosts that long. Wrong.
5966 Somebody had a whole pile of complicated expansion conditions, and the
5967 string was silently truncated, leading to an expansion error. It turns out
5968 that it is easier to change to an unlimited length (owing to other changes
5969 that have happened since this code was originally written) than to build
5970 structure for giving a limitation error. The length of the item that
5971 expands into the list of hosts is now unlimited.
5973 47. The lsearch lookup could not handle data where the length of text line was
5974 more than 4095 characters. Such lines were truncated, leading to shortened
5975 data being returned. It should now handle lines of any length.
5977 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5978 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5981 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5982 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5983 no longer happens in comment lines. A list of replacements is now placed
5984 at the head of all of the source files, except those whose only change is
5985 to replace PERL_COMMAND in the very first #! line.
5987 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5988 messages on the queue, with a bottom-up merge sort, using code contributed
5989 by Michael Haardt. This should make operations like -bp somewhat faster on
5990 large queues. It won't affect queue runners, except when queue_run_in_order
5993 51. Installed eximstats 1.31 in the distribution.
5995 52. Added support for SRV lookups to the dnslookup router.
5997 53. If an ACL referred to $message_body or $message_body_end, the value was not
5998 reset for any messages that followed in the same SMTP session.
6000 54. The store-handling optimization for building very long strings was not
6001 differentiating between the different store pools. I don't think this
6002 actually made any difference in practice, but I've tidied it.
6004 55. While running the routers to verify a sender address, $sender_address
6005 was still set to the sender address. This is wrong, because when routing to
6006 send a bounce to the sender, it would be empty. Therefore, I have changed
6007 it so that, while verifying a sender address, $sender_address is set to <>.
6008 (There is no change to what happens when verifying a recipient address.)
6010 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6011 target A or AAAA records (if not already returned) without resetting the
6012 qualify_single or search_parents options of the DNS resolver. These are
6013 inappropriate in this case because the targets of MX and SRV records must
6014 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6015 target that, when qualified, matched something in the local domain. These
6016 two options are now turned off when doing these lookups.
6018 57. It seems that at least some releases of Reiserfs (which does not have the
6019 concept of a fixed number of inodes) returns zero and not -1 for the
6020 number of available inodes. This interacted badly with check_spool_inodes,
6021 which assumed that -1 was the "no such thing" setting. What I have done is
6022 to check that the total number of inodes is greater than zero before doing
6023 the test of how many are available.
6025 58. When a "warn" ACL statement has a log_message modifier, the message is
6026 remembered, and not repeated. This is to avoid a lot of repetition when a
6027 message has many recipients that cause the same warning to be written.
6028 However, Exim was preserving the list of already written lines for an
6029 entire SMTP session, which doesn't seem right. The memory is now reset if a
6030 new message is started.
6032 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6033 debugging output unless log_rewrite was also set.
6035 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6036 of (int)(handle) when we know that handle contains (void *)(-1).
6038 61. The Exim daemon panic-logs an error return when it closes the incoming
6039 connection. However "connection reset by peer" seems to be common, and
6040 isn't really an error worthy of noting specially, so that particular error
6043 62. When Exim is trying to find all the local interfaces, it used to panic and
6044 die if the ioctl to get the interface flags failed. However, it seems that
6045 on at least one OS (Solaris 9) it is possible to have an interface that is
6046 included in the list of interfaces, but for which you get a failure error
6047 for this call. This happens when the interface is not "plumbed" into a
6048 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6049 failure of the "get flags" call assumes that the interface is down.
6051 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6052 makes life easier for people who are doing arithmetic on fields extracted
6053 from dates, where you often get leading zeros that should not be
6054 interpreted as octal.
6056 64. Added qualify_domain to the redirect router, to override the global
6059 65. If a pathologically long header line contained very many addresses (the
6060 report of this problem mentioned 10 000) and each of them was rewritten,
6061 Exim could use up a very large amount of memory. (It kept on making new
6062 copies of the header line as it rewrote, and never released the old ones.)
6063 At the expense of a bit more processing, the header rewriting function has
6064 been changed so that it no longer eats memory in this way.
6066 66. The generation of the Received: header has been moved from the time that a
6067 message starts to be received, to the time that it finishes. The timestamp
6068 in the Received: header should now be very close to that of the <= log
6069 line. There are two side-effects of this change:
6071 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6072 logged header lines no longer include the local Received: line, because
6073 it has not yet been created. The same applies to a copy of the message
6074 that is returned to a non-SMTP sender when a message is rejected.
6076 (b) When a filter file is tested using -bf, no additional Received: header
6077 is added to the test message. After some thought, I decided that this
6080 This change does not affect the value of $received_for. It is still set
6081 after address rewriting, but before local_scan() is called.
6083 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6085 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6086 gave an unhelpful panic error message, and a defer error. I have managed to
6087 change this behaviour so that it now rejects any supplied certificate,
6088 which seems right, as the list of acceptable certificates is empty.
6090 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6091 gave an unhelpful defer error. I have not managed to make this reject any
6092 supplied certificates, but the error message it gives is "no certificate
6093 supplied", which is not helpful.
6095 70. exigrep's output now also includes lines that are not associated with any
6096 message, but which match the given pattern. Implemented by a patch from
6097 Martin Sluka, which also tidied up the Perl a bit.
6099 71. Recipient callout verification, like sender verification, was using <> in
6100 the MAIL FROM command. This isn't really the right thing, since the actual
6101 sender may affect whether the remote host accepts the recipient or not. I
6102 have changed it to use the actual sender in the callout; this means that
6103 the cache record is now keyed on a recipient/sender pair, not just the
6104 recipient address. There doesn't seem to be a real danger of callout loops,
6105 since a callout by the remote host to check the sender would use <>.
6106 [SEE ABOVE: changed after hitting problems.]
6108 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6109 temporary errors. However, in the case of such a code being given after
6110 the end of a data transmission (i.e. after ".") Exim was failing to write
6111 a retry record for the message. (Yes, there was some broken host that was
6112 actually sending 8xx at this point.)
6114 73. An unknown lookup type in a host list could cause Exim to panic-die when
6115 the list was checked. (An example that provoked this was putting <; in the
6116 middle of a list instead of at the start.) If this happened during a DATA
6117 ACL check, a -D file could be left lying around. This kind of configuration
6118 error no longer causes Exim to die; instead it causes a defer error. The
6119 incident is still logged to the main and panic logs.
6121 74. Buglet left over from Exim 3 conversion. The message "too many messages
6122 in one connection" was written to the rejectlog but not the mainlog, except
6123 when address rewriting (yes!) was being logged.
6125 75. Added write_rejectlog option.
6127 76. When a system filter was run not as root (that is, when system_filter_user
6128 was set), the values of the $n variables were not being returned to the
6129 main process; thus, they were not subsequently available in the $sn
6132 77. Added +return_path_on_delivery log selector.
6134 78. A connection timeout was being treated differently from recipients deferred
6135 when testing hosts_max_try with a message that was older than the host's
6136 retry timeout. (The host should not be counted, thus allowing all hosts to
6137 be tried at least once before bouncing.) This may have been the cause of an
6138 occasionally reported bug whereby a message would remain on the queue
6139 longer than the retry timeout, but would be bounced if a delivery was
6140 forced. I say "may" because I never totally pinned down the problem;
6141 setting up timeout/retry tests is difficult. See also the next item.
6143 79. The ultimate address timeout was not being applied to errors that involved
6144 a combination of host plus message (for example, a timeout on a MAIL
6145 command). When an address resolved to a number of possible hosts, and they
6146 were not all tried for each delivery (e.g. because of hosts_max_try), a
6147 message could remain on the queue longer than the retry timeout.
6149 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6152 81. Fixed an obscure SMTP outgoing bug which required at least the following
6153 conditions: (a) there was another message waiting for the same server;
6154 (b) the server returned 5xx to all RCPT commands in the first message so
6155 that the message was not completed; (c) the server dropped the connection
6156 or gave a negative response to the RSET that Exim sends to abort the
6157 transaction. The observed case was a dropped connection after DATA that had
6158 been sent in pipelining mode. That is, the server had advertised PIPELINING
6159 but was not implementing it correctly. The effect of the bug was incorrect
6160 behaviour, such as trying another host, and this could lead to a crash.
6166 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6167 and daemon.c were passed as pointers to ints; they should have been
6168 pointers to socklen_t variables (which are typically unsigned ints).
6170 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6173 3. Fixed a really odd bug that affected only the testing scheme; patching a
6174 certain fixed string in the binary changed the value of another string that
6175 happened to be identical to the end of the original first string.
6177 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6178 name", it returns that address as the IP address. On some operating
6179 systems (e.g. Solaris), it also passes back the IP address string as the
6180 "host name". However, on others (e.g. Linux), it passes back an empty
6181 string. Exim wasn't checking for this, and was changing the host name to an
6182 empty string, assuming it had been canonicalized.
6184 5. Although rare, it is permitted to have more than one PTR record for a given
6185 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6186 all the names associated with an address, because they do in Solaris.
6187 However, it seems that they do not in Linux for data that comes from the
6188 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6189 I found this out when I moved to a new Linux workstation and tried to run
6190 the Exim test suite.
6192 To get round this problem I have changed the code so that it now does its
6193 own call to the DNS to look up PTR records when searching for a host name.
6194 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6195 addresses that are only in /etc/hosts are still found.
6197 This behaviour is, however, controlled by an option called host_lookup_
6198 order, which defaults to "bydns:byaddr". If people want to use the other
6199 order, or indeed, just use one or the other means of lookup, they can
6200 specify it in this variable.
6202 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6203 some operating systems, this comes back from gethostbyaddr() as an empty
6204 string, and this is what Exim used to test for. However, it seems that in
6205 other systems, "." is yielded. Exim now tests for this case too.
6207 7. The values of check_spool_space and check_log_space are now held internally
6208 as a number of kilobytes instead of an absolute number of bytes. If a
6209 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6210 kilobyte. This means that much larger values can be stored.
6212 8. Exim monitor: an attempt to get the action menu when not actually pointing
6213 at a message produces an empty menu entitled "No message selected". This
6214 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6215 no entries in it ("Shell widget menu has zero width and/or height"). So I
6216 have added a single, blank menu entry in this case.
6218 9. Added ${quote_local_part.
6220 10. MIME decoding is now applied to the contents of Subject: header lines when
6223 11. Now that a reference to $sender_host_address automatically causes a reverse
6224 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6225 host lookup before query-style lookups in lists that might use this
6226 variable. This has therefore been abolished, and the "net-" prefix is no
6227 longer necessary for query-style lookups.
6229 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6230 have been a typo for LFLAGS, so it has been changed.
6232 13. The install script calls Exim with "-C /dev/null" in order to find the
6233 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6234 to be output. However, since Exim outputs its version number before the
6235 error, it didn't break the script. It just looked ugly. I fixed this by
6236 always allowing "-C /dev/null" if the caller is root.
6238 14. Ignore overlarge ACL variable number when reading spool file - insurance
6239 against a later release with more variables having written the file.
6241 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6242 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6243 this, as well as the form without the "IPv6" on the front.
6245 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6246 OS/Makefile-Darwin file.
6248 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6249 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6251 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6252 variable. However, some OS use ints for inodes. I've added cast to long int
6253 to get rid of the compiler warning.
6255 19. I had forgotten to lock out "/../" in configuration file names when
6256 ALT_CONFIG_PREFIX was set.
6258 20. Routers used for verification do not need to specify transports. However,
6259 if such a router generated a host list, and callout was configured, Exim
6260 crashed, because it could not find a port number from the (non-existent)
6261 transport. It now assumes port 25 in this circumstance.
6263 21. Added the -t option to exigrep.
6265 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6266 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6267 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6268 them is set, LOOKUP_LSEARCH is forced.
6270 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6272 24. Added sender and host information to the "rejected by local_scan()" log
6273 line; previously there was no indication of these.
6275 25. Added .include_if_exists.
6277 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6278 Linux. It turns out that not all file systems support this. Apparently some
6279 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6280 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6281 not supported on the file descriptor, is now ignored when Exim is trying to
6282 sync a directory. This applies only to Linux.
6284 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6286 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6287 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6290 29. Added support for maildirsize files from supplied patch (modified a bit).
6292 30. The use of :fail: followed by an empty string could lead Exim to respond to
6293 sender verification failures with (e.g.):
6295 550 Verification failed for <xxx>
6296 550 Sender verify failed
6298 where the first response line was missing the '-' that indicates it is not
6299 the final line of the response.
6301 31. The loop for finding the name of the user that called Exim had a hardwired
6302 limit of 10; it now uses the value of finduser_retries, which is used for
6303 all other user lookups.
6305 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6308 33. Exim was neglecting to zero errno before one call of strtol() when
6309 expanding a string and expecting an integer value. On some systems this
6310 resulted in spurious "integer overflow" errors. Also, it was casting the
6311 result into an int without checking.
6313 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6314 did not work. The code looks as if it has *never* worked, though it appears
6315 to have been documented since at least release 1.62. I have made it work.
6317 35. The "timeout_DNS" error in retry rules, also documented since at least
6318 1.62, also never worked. As it isn't clear exactly what this means, and
6319 clearly it isn't a major issue, I have abolished the feature by treating it
6320 as "timeout", and writing a warning to the main and panic logs.
6322 36. The display of retry rules for -brt wasn't always showing the error code
6325 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6326 timeout_connect_A, timeout_connect_MX.
6328 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6329 to the empty sender.
6331 39. The daemon was not analysing the content of -oX till after it had closed
6332 stderr and disconnected from the controlling terminal. This meant that any
6333 syntax errors were only noted on the panic log, and the return code from
6334 the command was 0. By re-arranging the code a little, I've made the
6335 decoding happen first, so such errors now appear on stderr, and the return
6336 code is 1. However, the actual setting up of the sockets still happens in
6337 the disconnected process, so errors there are still only recorded on the
6340 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6341 connections (as happens on some IP stacks) was logged at start up time as
6342 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6343 it from "IPv6 and IPv4", which means that two separate sockets are being
6346 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6347 says whether AF_INET or AF_INET6 was passed as an argument.
6349 42. Exiwhat output was messed up when time zones were included in log
6352 43. Exiwhat now gives more information about the daemon's listening ports,
6353 and whether -tls-on-connect was used.
6355 44. The "port" option of the smtp transport is now expanded.
6357 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6358 silently ignored. Now an error message is written to the main and panic
6361 46. There's a new ACL modifier called "logwrite" which writes to a log file
6362 as soon as it is encountered.
6364 47. Added $local_user_uid and $local_user_gid at routing time.
6366 48. Exim crashed when trying to verify a sender address that was being
6369 49. Exim was recognizing only a space character after ".include". It now also
6370 recognizes a tab character.
6372 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6373 extracting the relevant information from the specification. The man page no
6374 longer contains scrambled data for the -d option, and I've added a section
6375 at the front about calling Exim under different names.
6377 51. Added "extra_headers" argument to the "mail" command in filter files.
6379 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6382 53. Installed eximstats 1.29.
6384 54. Added transport_filter_timeout as a generic transport option.
6386 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6387 Cc: header lines. This was required by RFC 822, but it not required by RFC
6390 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6391 incoming messages that did not have them. Now it does so only if the
6392 message originates locally, that is, if there is no associated remote host
6393 address. When Resent- header lines are present, this applies to the Resent-
6394 lines rather than the non-Resent- lines.
6396 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6397 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6399 58. Messages for configuration errors now include the name of the main
6400 configuration file - useful now that there may be more than one file in a
6401 list (.included file names were always shown).
6403 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6404 for those rare installations that do not start the daemon as root or run it
6405 setuid root. I've cut out the call to initgroups() if the daemon is not
6408 60. The Exim user and group can now be bound into the binary as text strings
6409 that are looked up at the start of Exim's processing.
6411 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6413 62. Added $mailstore_basename variable.
6415 63. Installed patch to sieve.c from Michael Haardt.
6417 64. When Exim failed to open the panic log after failing to open the main log,
6418 the original message it was trying to log was written to stderr and debug
6419 output, but if they were not available (the usual case in production), it
6420 was lost. Now it is written to syslog before the two lines that record the
6421 failures to open the logs.
6423 65. Users' Exim filters run in subprocesses under the user's uid. It is
6424 possible for a "deliver" command or an alias in a "personal" command to
6425 provoke an address rewrite. If logging of address rewriting is configured,
6426 this fails because the process is not running as root or exim. There may be
6427 a better way of dealing with this, but for the moment (because 4.30 needs
6428 to be released), I have disabled address rewrite logging when running a
6429 filter in a non-root, non-exim process.
6435 1. The buildconfig auxiliary program wasn't quoting the value set for
6436 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6437 not defined. This bug was masked in 4.22 by the effect that was fixed in
6440 2. Some messages that were rejected after a message id was allocated were
6441 shown as "incomplete" by exigrep. It no longer does this for messages that
6442 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6444 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6445 have allow_domain_literals set, the ID did not get logged in the <= line.
6446 Domain literals are now always recognized in Message-ID: header lines.
6448 4. The first argument for a ${extract expansion item is the key name or field
6449 number. Leading and trailing spaces in this item were not being ignored,
6450 causing some misleading effects.
6452 5. When deliver_drop_privilege was set, single queue runner processes started
6453 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6454 same command in the process it spins off) were not dropping privilege.
6456 6. When the daemon running as "exim" started a queue runner, it always
6457 re-executed Exim in the spun-off process. This is a waste of effort when
6458 deliver_drop_privilege is set. The new process now just calls the
6459 queue-runner function directly.
6465 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6468 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6469 ignored. Though the use of -oP was forcing the writing of a pid file, it
6470 was always written to the default place.
6472 3. If the message "no IP address found for host xxxx" is generated during
6473 incoming verification, it is now followed by identification of the incoming
6474 connection (so you can more easily find what provoked it).
6476 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6478 5. Added some features to "harden" Exim a bit more against certain attacks:
6480 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6481 be put in Local/Makefile. This is like the never_users runtime option,
6482 but it cannot be overridden. The default setting is "root".
6484 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6485 prefix string with which any file named in a -C command line option
6488 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6489 is retained for -C and -D only if the caller of Exim is root. Without
6490 it, the exim user may also use -C and -D and retain privilege.
6492 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6493 command line option is disabled.
6495 6. Macro names set by the -D option must start with an upper case letter, just
6496 like macro names defined in the configuration file.
6498 7. Added "dereference=" facility to LDAP.
6500 8. Two instances of the typo "uknown" in the source files are fixed.
6502 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6503 the Configure-Makefile script screwed up while processing it.
6505 10. Incorporated PCRE 4.4.
6507 11. The SMTP synchronization check was not operating right at the start of an
6508 SMTP session. For example, it could not catch a HELO sent before the client
6509 waited for the greeting. There is now a check for outstanding input at the
6510 point when the greeting is written. Because of the duplex, asynchronous
6511 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6512 way, but not yet received, when the check is performed.
6514 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6515 on TCP/IP sockets, because this apparently causes some broken clients to
6518 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6519 unchanged) from the Cygwin maintainer.
6521 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6522 was supported instead of testing for mbx. Effectively a typo.
6524 15. The spa authenticator server code was not checking that the input it
6525 received was valid base64.
6527 16. The debug output line for the "set" modifier in ACLs was not showing the
6528 name of the variable that was being set.
6530 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6532 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6533 lookup on its own. Something else was needed to trigger the lookup. For
6534 example, a match in host_lookup or the need for a host name in a host list.
6535 Now, if $sender_host_name is referenced and the host name has not yet been
6536 looked up, a lookup is performed. If the lookup fails, the variable remains
6537 empty, and $host_lookup_failed is set to "1".
6539 19. Added "eqi" as a case-independent comparison operator.
6541 20. The saslauthd authentication condition could segfault if neither service
6542 nor realm was specified.
6544 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6545 error message that was logged was misleading, and incoming SMTP
6546 connections were dropped. The message is now more accurate, and temporary
6547 errors are given to SMTP connections.
6549 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6550 (except RSET) with a 421 error, until QUIT is received. However, it was
6551 failing to send a response to QUIT.
6553 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6554 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6555 ACL is now run after the helo_try_verify_hosts code.
6557 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6558 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6559 case-independent, so other case variants are also recognized). Apparently
6560 some systems use these upper case variants.
6562 25. If more than two messages were waiting for the same host, and a transport
6563 filter was specified for the transport, Exim sent two messages over the
6564 same TCP/IP connection, and then failed with "socket operation on non-
6565 socket" when it tried to send the third.
6567 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6569 27. The extern definition of crypt16() in expand.c was not being excluded when
6570 the OS had its own crypt16() function.
6572 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6573 as a preferred synonym for return_size_limit, both as an option and as an
6576 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6578 30. Changed the default configuration ACL to relax the local part checking rule
6579 for addresses that are not in any local domains. For these addresses,
6580 slashes and pipe symbols are allowed within local parts, but the sequence
6581 /../ is explicitly forbidden.
6583 31. SPA server authentication was not clearing the challenge buffer before
6586 32. log_message in a "warn" ACL statement was writing to the reject log as
6587 well as to the main log, which contradicts the documentation and doesn't
6588 seem right (because no rejection is happening). So I have stopped it.
6590 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6591 However, I am unable to do any testing of this.
6593 34. Fixed an infelicity in the appendfile transport. When checking directories
6594 for a mailbox, to see if any needed to be created, it was accidentally
6595 using path names with one or more superfluous leading slashes; tracing
6596 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6598 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6599 contents are added to the log line that is written for every discarded
6600 recipient. (Previously a log_message setting was ignored.)
6602 36. The ${quote: operator now quotes the string if it is empty.
6604 37. The install script runs exim in order to find its version number. If for
6605 some reason other than non-existence or emptiness, which it checks, it
6606 could not run './exim', it was installing it with an empty version number,
6607 i.e. as "exim-". This error state is now caught, and the installation is
6610 38. An argument was missing from the function that creates an error message
6611 when Exim fails to connect to the socket for saslauthd authentication.
6612 This could cause Exim to crash, or give a corrupted message.
6614 39. Added isip, isip4, and isip6 to ${if conditions.
6616 40. The ACL variables $acl_xx are now saved with the message, and can be
6617 accessed later in routers, transports, and filters.
6619 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6620 strings in the file are not string-expanded.
6622 42. If a MAIL command specified a SIZE value that was too large to fit into an
6623 int variable, the check against message_size_limit failed. Such values are
6624 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6625 day this will have to be increased, but I don't think I want to be around
6626 when emails are that large.
6633 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6634 iconv() is not standard in FreeBSD.
6636 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6637 IPv6 enabled. The observed symptom was a segmentation fault on return from
6638 the function os_common_find_running_interfaces() in src/os.c.
6640 3. In the check_special_case() function in daemon.c I had used "errno" as an
6641 argument name, which causes warnings on some systems. This was basically a
6642 typo, since it was named "eno" in the comments!
6644 4. The code that waits for the clock to tick (at a resolution of some fraction
6645 of a second) so as to ensure message-id uniqueness was always waiting for
6646 at least one whole tick, when it could have waited for less. [This is
6647 almost certainly not relevant at current processor speeds, where it is
6648 unlikely to ever wait at all. But we try to future-proof.]
6650 5. The function that sleeps for a time interval that includes fractions of a
6651 second contained a race. It did not block SIGALRM between setting the
6652 timer, and suspending (a couple of lines later). If the interval was short
6653 and the sigsuspend() was delayed until after it had expired, the suspension
6654 never ended. On busy systems this could lead to processes getting stuck for
6657 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6658 process, before it forks any delivery processes. The open lookup caching
6659 mechanism meant that the open file or database connection was passed into
6660 the delivery process. The problem was that delivery processes always tidy
6661 up cached lookup data. This could cause a problem for the next delivery
6662 process started by the queue runner, because the external queue runner
6663 process does not know about the closure. So the next delivery process
6664 still has data in the lookup cache. In the case of a file lookup, there was
6665 no problem because closing a file descriptor in a subprocess doesn't affect
6666 the parent. However, if the lookup was caching a connection to a database,
6667 the connection was closed, and the second delivery process was likely to
6668 see errors such as "PGSQL: query failed: server closed the connection
6669 unexpectedly". The problem has been fixed by closing all cached lookups
6670 in a queue runner before running a delivery process.
6672 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6673 seem to have the "const" qualifier which it has on other OS. I've
6676 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6677 *of the same type* (client or server) with the same public name that an
6678 error should be diagnosed.
6680 9. When Exim looked up a host name for an IP address, but failed to find the
6681 original IP address when looking up the host name (a safety check), it
6682 output the message "<ip address> does not match any IP for NULL", which was
6683 confusing, to say the least. The bug was that the host name should have
6684 appeared instead of "NULL".
6686 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6687 user that is built into the binary, and the -C or -D options is used, root
6688 privilege is dropped before the configuration file is read. In addition,
6689 logging is switched to stderr instead of the normal log files. If the
6690 configuration then re-defines the Exim user, the unprivileged environment
6691 is probably not what is expected, so Exim logs a panic warning message (but
6694 However, if deliver_drop_privilege is set, the unprivileged state may well
6695 be exactly what is intended, so the warning has been cut out in that case,
6696 and Exim is allowed to try to write to its normal log files.
6702 1. smtp_return_error_details was not giving details for temporary sender
6703 or receiver verification errors.
6705 2. Diagnose a configuration error if two authenticators have the same public
6708 3. Exim used not to create the message log file for a message until the first
6709 delivery attempt. This could be confusing when incoming messages were held
6710 for policy or load reasons. The message log file is now created at the time
6711 the message is received, and an initial "Received" line is written to it.
6713 4. The automatically generated man page for command line options had a minor
6714 bug that caused no ill effects; however, a more serious problem was that
6715 the procedure for building the man page automatically didn't always
6716 operate. Consequently, release 4.20 contains an out-of-date version. This
6717 shouldn't happen again.
6719 5. When building Exim with embedded Perl support, the script that builds the
6720 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6721 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6723 6. The freeze_tell option was not being used for messages that were frozen on
6724 arrival, either by an ACL or by local_scan().
6726 7. Added the smtp_incomplete_transaction log selector.
6728 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6729 was accepting AUTH without a new EHLO.
6731 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6732 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6735 10. Logging of TCP/IP connections (when configured) now happens in the main
6736 daemon process instead of the child process, so that the TCP/IP connection
6737 count is more accurate (but it can never be perfect).
6739 11. The use of "drop" in a nested ACL was not being handled correctly in the
6740 outer ACL. Now, if condition failure induced by the nested "drop" causes
6741 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6742 or "require"), the connection is dropped.
6744 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6745 that yield "discard" can now be used with an "accept" or a "discard" verb,
6746 but an error is generated for any others (because I can't see a useful way
6747 to define what should happen).
6749 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6750 are now processed in the same way as lines in the Exim configuration file.
6751 In particular, continuation lines are supported.
6753 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6755 15. Added -ti meaning -t -i.
6757 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6758 domains, and warn by logging if others are found.
6760 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6761 returned from GIFCONF when Exim is trying to find the list of interfaces on
6762 a host. The code in os.c has been modified to copy each ifreq to an aligned
6763 structure in all cases.
6765 Also, in some cases, the returned ifreq's were being copied to a 'struct
6766 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6767 means the last couple of bytes of an IPv6 address could be chopped if the
6768 ifreq contained only a normal sockaddr (14 bytes storage).
6770 18. Named domain lists were not supported in the hosts_treat_as_local option.
6771 An entry such as +xxxx was not recognized, and was treated as a literal
6774 19. Ensure that header lines added by a DATA ACL are included in the reject log
6775 if the ACL subsequently rejects the message.
6777 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6778 MD5 (which is deprecated).
6780 21. When testing a filter file using -bf, Exim was writing a message when it
6781 took the sender from a "From " line in the message, but it was not doing so
6782 when it took $return_path from a Return-Path: header line. It now does.
6784 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6785 with a valid header line field name (a series of printing characters
6786 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6788 23. Changed "disc" in the source to "disk" to conform to the documentation and
6789 the book and for uniformity.
6791 24. Ignore Sendmail's -Ooption=value command line item.
6793 25. When execve() failed while trying to run a command in a pipe transport,
6794 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6795 could be confused with a return value of 69 from the command itself. This
6796 has been changed to 127, the value the shell returns if it is asked to run
6797 a non-existent command. The wording for the related log line suggests a
6798 non-existent command as the problem.
6800 26. If received_header_text expands to an empty string, do not add a Received:
6801 header line to the message. (Well, it adds a token one on the spool, but
6802 marks it "old" so that it doesn't get used or transmitted.)
6804 27. Installed eximstats 1.28 (addition of -nt option).
6806 28. There was no check for failure on the call to getsockname() in the daemon
6807 code. This can fail if there is a shortage of resources on the system, with
6808 ENOMEM, for example. A temporary error is now given on failure.
6810 29. Contrary to the C standard, it seems that in some environments, the
6811 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6812 program. Exim now does this explicitly; it affects the formatting of
6813 timestamps using strftime().
6815 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6816 complaints. I've now initialized all the variables, to avoid this.
6818 32. Header lines added by a system filter were not being "seen" during
6819 transport-time rewrites.
6821 33. The info_callback() function passed to OpenSSL is set up with type void
6822 (*)(SSL *, int, int), as described somewhere. However, when calling the
6823 function (actually a macro) that sets it up, the type void(*)() is
6824 expected. I've put in a cast to prevent warnings from picky compilers.
6826 34. If a DNS black list lookup found a CNAME record, but there were no A
6827 records associated with the domain it pointed at, Exim crashed.
6829 35. If a DNS black list lookup returned more than one A record, Exim ignored
6830 all but the first. It now scans all returned addresses if a particular IP
6831 value is being sought. In this situation, the contents of the
6832 $dnslist_value variable are a list of all the addresses, separated by a
6835 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6836 to accept a host name and all its aliases if the forward lookup for any of
6837 them yielded the IP address of the incoming connection. Now it accepts only
6838 those names whose forward lookup yields the correct IP address. Any other
6839 names are discarded. This closes a loophole whereby a rogue DNS
6840 administrator could create reverse DNS records to break through a
6841 wildcarded host restriction in an ACL.
6843 37. If a user filter or a system filter that ran in a subprocess used any of
6844 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6845 the wrong values were passed to the pipe command ($thisaddress had the
6846 value of $0, $0 had the value of $1, etc). This bug was introduced by
6847 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6849 38. Improved the line breaking for long SMTP error messages from ACLs.
6850 Previously, if there was no break point between 40 and 75 characters, Exim
6851 left the rest of the message alone. Two changes have been made: (a) I've
6852 reduced the minimum length to 35 characters; (b) if it can't find a break
6853 point between 35 and 75 characters, it looks ahead and uses the first one
6854 that it finds. This may give the occasional overlong line, but at least the
6855 remaining text gets split now.
6857 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6858 file descriptors might be low, and that setting 1000 would always raise it.
6859 It turns out that in some environments, the limit is already over 1000 and
6860 that lowering it causes trouble. So now Exim takes care not to decrease it.
6862 40. When delivering a message, the value of $return_path is set to $sender_
6863 address at the start of routing (routers may change the value). By an
6864 oversight, this default was not being set up when an address was tested by
6865 -bt or -bv, which affected the outcome if any router or filter referred to
6868 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6869 log only when "message" or "log_message" are set. However, if one of the
6870 conditions was an address verification, or a call to a nested ACL, the
6871 messages generated by the underlying test were being passed through. This
6872 no longer happens. The underlying message is available in $acl_verify_
6873 message for both "message" and "log_message" expansions, so it can be
6874 passed through if needed.
6876 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6877 new expansion $bh_ to give the encoded byte string without charset
6878 translation. Translation happens only if iconv() is available; HAVE_ICONV
6879 indicates this at build time. HEADERS_CHARSET gives the charset to
6880 translate to; headers_charset can change it in the configuration, and
6881 "headers charset" can change it in an individual filter file.
6883 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6884 that creates RFC 2047 encoded "words" labels them as that charset instead
6885 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6886 expansion operator; (ii) when Exim creates a From: line for a local
6887 message; (iii) when a header line is rewritten to include a "phrase" part.
6889 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6890 buggy, causing it to skip the first lines of messages whose message ID
6891 ended in 'D'. This would not have bitten before Exim release 4.14, because
6892 message IDs were unlikely to end in 'D' before then. The effect was to have
6893 incorrect size information for certain domains.
6895 45. #include "config.h" was missing at the start of the crypt16.c module. This
6896 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6899 46. If there was a timeout during a "random" callout check, Exim treated it as
6900 a failure of the random address, and carried on sending RSET and the real
6901 address. If the delay was just some slowness somewhere, the response to the
6902 original RCPT would be taken as a response to RSET and so on, causing
6903 mayhem of various kinds.
6905 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6906 when I implemented it. It didn't allow for the fact that some option values
6907 may legitimately be negative (e.g. size_addition), and it didn't even do
6908 the right test for positive values.
6910 48. Domain names in DNS records are case-independent. Exim always looks them up
6911 in lower case. Some resolvers return domain names in exactly the case they
6912 appear in the zone file, that is, they may contain uppercase letters. Not
6913 all resolvers do this - some return always lower case. Exim was treating a
6914 change of case by a resolver as a change of domain, similar to a widening
6915 of a domain abbreviation. This triggered its re-routing code and so it was
6916 trying to route what was effectively the same domain again. This normally
6917 caused routing to fail (because the router wouldn't handle the domain
6918 twice). Now Exim checks for this case specially, and just changes the
6919 casing of the domain that it ultimately uses when it transmits the message
6922 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6925 50. If a filter generated a file delivery with a non-absolute name (possible if
6926 no home directory exists for the router), the forbid_file option was not
6929 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6930 the existing equality matching.
6932 52. Exim was using ints instead of ino_t variables in some places where it was
6933 dealing with inode numbers.
6935 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6936 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6937 called TMPDIR, and if it finds it is different, it changes its value.
6939 54. The smtp_printf() function is now made available to local_scan() so
6940 additional output lines can be written before returning. There is also an
6941 smtp_fflush() function to enable the detection of a dropped connection.
6942 The variables smtp_input and smtp_batched_input are exported to
6945 55. Changed the default runtime configuration: the message "Unknown user"
6946 has been removed from the ACL, and instead placed on the localuser router,
6947 using the cannot_route_message feature. This means that any verification
6948 failures that generate their own messages won't get overridden. Similarly,
6949 the "Unrouteable address" message that was in the ACL for unverifiable
6950 relay addresses has also been removed.
6952 56. Added hosts_avoid_esmtp to the smtp transport.
6954 57. The exicyclog script was not checking for the esoteric option
6955 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6956 will work only if exicyclog is run under the appropriate euid.
6958 58. Following a discussion on the list, the rules by which Exim recognises line
6959 endings on incoming messages have been changed. The -dropcr and drop_cr
6960 options are now no-ops, retained only for backwards compatibility. The
6961 following line terminators are recognized: LF CRLF CR. However, special
6962 processing applies to CR:
6964 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6965 nor a local message in the state where . is a terminator.
6967 (ii) If a bare CR is encountered in a header line, an extra space is added
6968 after the line terminator so as not to end the header. The reasoning
6969 behind this is that bare CRs in header lines are most likely either
6970 to be mistakes, or people trying to play silly games.
6972 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6973 was being incorrectly given as 18 bytes larger than it should have been.
6974 This is a VOB (very old bug).
6976 60. This may never have affected anything current, but just in case it has:
6977 When the local host is found other than at the start of a list of hosts,
6978 the local host, those with the same MX, and any that follow, are discarded.
6979 When the list in question was part of a longer list of hosts, the following
6980 hosts (not currently being processed) were also being discarded. This no
6981 longer happens. I'm not sure if this situation could ever has previously
6984 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6987 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6988 header. This is something that is recommended in a new Internet Draft, and
6989 is something that is documented as being done by Sendmail. There are two
6990 possible values. For messages generated by the autoreply transport, Exim
6993 Auto-Submitted: auto-replied
6995 whereas for all other generated messages (e.g. bounces) it adds
6997 Auto-Submitted: auto-generated
6999 63. The "personal" condition in filters now includes a test for the
7000 Auto-Submitted: header. If it contains the string "auto-" the message it
7001 not considered personal.
7003 64. Added rcpt_include_affixes as a generic transport option.
7005 65. Added queue_only_override (default true).
7007 66. Added the syslog_duplication option.
7009 67. If what should have been the first header line of a message consisted of
7010 a space followed by a colon, Exim was mis-interpreting it as a header line.
7011 It isn't of course - it is syntactically invalid and should therefore be
7012 treated as the start of the message body. The misbehaviour could have
7013 caused a number of strange effects, including loss of data in subsequent
7014 header lines, and spool format errors.
7016 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7017 client host had authenticated. This control can now be exercised by an ACL
7018 for more flexibility.
7020 69. By default, callouts do not happen when testing with -bh. There is now a
7021 variant, -bhc, which does actually run the callout code, including
7022 consulting and updating the callout cache.
7024 70. Added support for saslauthd authentication, courtesy of Alexander
7027 71. If statvfs() failed on the spool or log directories while checking their
7028 size for availability, Exim confusingly gave the error "space shortage".
7029 Furthermore, in debugging mode it crashed with a floating point exception.
7030 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7031 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7032 serious problem, Exim now writes to the main and panic logs when this
7033 happens, with details of the failure. It then refuses to accept the
7034 incoming message, giving the message "spool directory problem" or "log
7035 directory problem" with a 421 code for SMTP messages.
7037 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7038 0, 1, and 2 exist, because some OS complain for execs without them (see
7039 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7040 descriptors. However, the code omitted to check that the open succeeded,
7041 causing mysterious errors if for some reason the permissions on /dev/null
7042 got screwed. Now Exim writes a message to the main and panic logs, and
7043 bombs out if it can't open /dev/null.
7045 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7046 interact so that it is all more flexible. It is supposed to remain
7047 backwards compatible. Also added extra_local_interfaces.
7049 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7050 to bomb out with an assertion failure - to the client this appears as a
7051 connection drop. This problem occurs in the part of the code that was taken
7052 from the Samba project. Fortunately, the assertion is in a very simple
7053 function, so I have fixed this by reproducing the function inline in the
7054 one place where it is called, and arranging for authentication to fail
7055 instead of killing the process with assert().
7057 75. The SPA client code was not working when the server requested OEM rather
7058 than Unicode encoding.
7060 76. Added code to make require_files with a specific uid setting more usable in
7061 the case where statting the file as root fails - usually a non-root-mounted
7062 NFS file system. When this happens and the failure is EACCES, Exim now
7063 forks a subprocess and does the per-uid checking as the relevant uid.
7065 77. Added process_log_path.
7067 78. If log_file_path was not explicitly set, a setting of check_log_space or
7068 check_log_inodes was ignored.
7070 79. If a space check for the spool or log partitions fails, the incident is now
7071 logged. Of course, in the latter case the data may get lost...
7073 80. Added the %p formatting code to string_format() so that it can be used to
7074 print addresses in debug_print(). Adjusted all the address printing in the
7075 debugging in store.c to use %p rather than %d.
7077 81. There was a concern that a line of code in smtp_in.c could overflow a
7078 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7079 initially expressed, the concern was not well-founded, because trailing
7080 spaces are removed early. However, if the trailing spaces were followed by
7081 a NULL, they did not get removed, so the overflow was possible. Two fixes
7084 (a) I re-wrote the offending code in a cleaner fashion.
7085 (b) If an incoming SMTP command contains a NULL character, it is rejected
7088 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7089 runs initgroups(), so that if the Exim user is in any additional groups,
7090 they will be used during message reception.
7096 The change log for 4.20 and earlier releases has been archived.