1 Change log file for Exim from version 4.21
2 ------------------------------------------
3 This document describes *changes* to previous versions, that might
4 affect Exim's operation, with an unchanged configuration file. For new
5 options, and new features, see the NewStuff file next to this ChangeLog.
8 Since Exim version 4.90
9 -----------------------
11 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
12 When redis_servers is set to a list of > 1 element, and the Redis servers
13 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
14 case of MOVED into a DEFER case instead, thus moving the query onto the
15 next server in the list. For a cluster of N elements, all N servers must
16 be defined in redis_servers.
18 JH/01 Replace the store_release() internal interface with store_newblock(),
19 which internalises the check required to safely use the old one, plus
20 the allocate and data copy operations duplicated in both (!) of the
23 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
24 modifier. This matches the restriction on the commandline.
26 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
27 Previously only the last row was returned.
29 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
30 we assumed that tags in the header were well-formed, and parsed the
31 element content after inspecting only the first char of the tag.
32 Assumptions at that stage could crash the receive process on malformed
35 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
36 While running the DKIM ACL we operate on the Permanent memory pool so that
37 variables created with "set" persist to the DATA ACL. Also (at any time)
38 DNS lookups that fail create cache records using the Permanent pool. But
39 expansions release any allocations made on the current pool - so a dnsdb
40 lookup expansion done in the DKIM ACL releases the memory used for the
41 DNS negative-cache, and bad things result. Solution is to switch to the
42 Main pool for expansions.
43 While we're in that code, add checks on the DNS cache during store_reset,
44 active in the testsuite.
45 Problem spotted, and debugging aided, by Wolfgang Breyha.
47 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
48 When none of the hosts presented to a transport match an already-open
49 connection, close it and proceed with the list. Previously we would
50 queue the message. Spotted by Lena with Yahoo, probably involving
53 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
54 Previously a spurious "250 OK id=" response was appended to the proper
57 JH/08 The "support for" informational output now, which built with Content
58 Scanning support, has a line for the malware scanner interfaces compiled
59 in. Interface can be individually included or not at build time.
61 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
62 by the template makefile "src/EDITME". The "STREAM" support for an older
63 ClamAV interface method is removed.
65 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
66 rows affected is given instead).
68 JH/11 The runtime Berkeley DB library version is now additionally output by
69 "exim -d -bV". Previously only the compile-time version was shown.
71 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
72 SMTP connection. Previously, when one had more receipients than the
73 first, an abortive onward connection was made. Move to full support for
74 multiple onward connections in sequence, handling cutthrough connection
75 for all multi-message initiating connections.
77 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
78 routers. Previously, a multi-recipient message would fail to match the
79 onward-connection opened for the first recipient, and cause its closure.
81 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
82 a timeout on read on a GnuTLS initiating connection, resulting in the
83 initiating connection being dropped. This mattered most when the callout
84 was marked defer_ok. Fix to keep the two timeout-detection methods
87 JH/15 Relax results from ACL control request to enable cutthrough, in
88 unsupported situations, from error to silently (except under debug)
89 ignoring. This covers use with PRDR, frozen messages, queue-only and
92 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
94 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
95 metadata, resulting in a crash in free().
97 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
98 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
99 Broken also in d185889f4, with init system revamp.
101 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
102 Previously we abruptly closed the connection after reading a malware-
103 found indication; now we go on to read the "scan ok" response line,
106 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
107 ACL. Previously, a crash would result.
109 JH/19 Speed up macro lookups during configuration file read, by skipping non-
110 macro text after a replacement (previously it was only once per line) and
111 by skipping builtin macros when searching for an uppercase lead character.
113 JH/20 DANE support moved from Experimental to mainline. The Makefile control
114 for the build is renamed.
116 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
117 was allocated for every new TLS startup, meaning one per message. Fix
118 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
120 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
121 reported the original. Fix to report (as far as possible) the ACL
122 result replacing the original.
124 JH/23 Fix memory leak during multi-message connections using STARTTLS under
125 OpenSSL. Certificate information is loaded for every new TLS startup,
126 and the resources needed to be freed.
128 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
130 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
133 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
134 DATA response info to the (existing) per-recipient response info for
135 the "C=" log element. It can have useful tracking info from the
136 destination system. Patch from Simon Arlott.
138 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
139 length value. Previously this would segfault.
141 HS/02 Support Avast multiline protoocol, this allows passing flags to
142 newer versions of the scanner.
148 JH/01 Rework error string handling in TLS interface so that the caller in
149 more cases is responsible for logging. This permits library-sourced
150 string to be attached to addresses during delivery, and collapses
151 pairs of long lines into single ones.
153 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
154 during configuration. Wildcards are allowed and expanded.
156 JH/02 Rework error string handling in DKIM to pass more info back to callers.
157 This permits better logging.
159 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
160 do not close it down and have the child transport start it up again on
161 the passed-on TCP connection. Instead, proxy the child (and any
162 subsequent ones) for TLS via a unix-domain socket channel. Logging is
163 affected: the continued delivery log lines do not have any DNSSEC, TLS
164 Certificate or OCSP information. TLS cipher information is still logged.
166 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
167 identical IP addresses on different listening ports. Will also affect
170 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
171 add noisy ifdef guards to special-case this sillyness.
172 Patch from Bernd Kuhls.
174 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
175 than 255 are no longer allowed.
177 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
178 Disable the session-cache too, which might reduce our load. Since we
179 currrectly use a new context for every connection, both as server and
180 client, there is no benefit for these.
181 GnuTLS appears to not support tickets server-side by default (we don't
182 call gnutls_session_ticket_enable_server()) but client side is enabled
183 by default on recent versions (3.1.3 +) unless the PFS priority string
186 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
187 <https://reproducible-builds.org/specs/source-date-epoch/>.
189 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
190 the check for any unsuccessful recipients did not notice the limit, and
191 erroneously found still-pending ones.
193 JH/08 Pipeline CHUNKING command and data together, on kernels that support
194 MSG_MORE. Only in-clear (not on TLS connections).
196 JH/09 Avoid using a temporary file during transport using dkim. Unless a
197 transport-filter is involved we can buffer the headers in memory for
198 creating the signature, and read the spool data file once for the
199 signature and again for transmission.
201 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
202 4.77 as the kernel support then wasn't solid, having issues in 64bit
203 mode. Now, it's been long enough. Add support for FreeBSD also.
205 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
206 case where the routing stage had gathered several addresses to send to
207 a host before calling the transport for the first, we previously failed
208 to close down TLS in the old transport process before passing the TCP
209 connection to the new process. The new one sent a STARTTLS command
210 which naturally failed, giving a failed delivery and bloating the retry
211 database. Investigation and fix prototype from Wolfgang Breyha.
213 JH/12 Fix check on SMTP command input synchronisation. Previously there were
214 false-negatives in the check that the sender had not preempted a response
215 or prompt from Exim (running as a server), due to that code's lack of
216 awareness of the SMTP input buffering.
218 PP/04 Add commandline_checks_require_admin option.
219 Exim drops privileges sanely, various checks such as -be aren't a
220 security problem, as long as you trust local users with access to their
221 own account. When invoked by services which pass untrusted data to
222 Exim, this might be an issue. Set this option in main configuration
223 AND make fixes to the calling application, such as using `--` to stop
226 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
227 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
228 responses to those, into a single TLS record each way (this usually means
229 a single packet). As a side issue, smtp_enforce_sync now works on TLS
232 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
233 affects you only if you're dancing at the edge of the param size limits.
234 If you are, and this message makes sense to you, then: raise the
235 configured limit or use OpenSSL 1.1. Nothing we can do for older
238 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
239 cmdline element to get the documented default one. Previously it was
242 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
243 are unsuable for non-message-specific reasons - eg. connection timeout,
244 banner-time rejection.
246 JH/16 Fix logging of delivery remote port, when specified by router, under
249 PP/06 Repair manualroute's ability to take options in any order, even if one
250 is the name of a transport.
253 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
255 JH/17 Change the list-building routines interface to use the expanding-string
256 triplet model, for better allocation and copying behaviour.
258 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
259 Previously it was constructed the first time a possibly-matching string
260 was met in the configuration file input during startup; now it is done
263 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
264 compatible one, to avoid the (poorly documented) possibility of a config
265 file in the working directory redirecting the DB files, possibly correpting
266 some existing file. CVE-2017-10140 assigned for BDB.
268 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
269 cache-hot. Previously, although the result was properly cached, the
270 initial verify call returned a defer.
272 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
273 the main verify for receipient in uncached-mode.
275 JH/22 Retire historical build files to an "unsupported" subdir. These are
276 defined as "ones for which we have no current evidence of testing".
278 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
279 if present. Previously it was ignored.
281 JH/24 Start using specified-initialisers in C structure init coding. This is
282 a C99 feature (it's 2017, so now considered safe).
284 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
285 if was a fixed-sized field and bitmask ops via macros; it is now more
288 PP/07 GitHub PR 56: Apply MariaDB build fix.
289 Patch provided by Jaroslav Škarvada.
291 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
292 during Coverity cleanups [4.87 JH/47]
293 Diagnosis and fix provided by Michael Fischer v. Mollard.
295 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
296 the right size to place the terminating semicolon on its own folded
297 line, the header hash was calculated to an incorrect value thanks to
298 the (relaxed) space the fold became.
300 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
301 and confused the parent.
303 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
304 which could crash as a result. This could lead to undeliverable messages.
306 JH/28 Logging: "next input sent too soon" now shows where input was truncated
309 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
310 matters on fast-turnover and PID-randomising systems, which were getting
311 out-of-order delivery.
313 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
314 a possibly-overlapping copy. The symptom was that "Remote host closed
315 connection in response to HELO" was logged instead of the actual 4xx
318 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
319 Previously only that bufferd was discarded, resulting in SYMTP command
322 JH/32 DKIM: when a message has multiple signatures matching an identity given
323 in dkim_verify_signers, run the dkim acl once for each. Previously only
324 one run was done. Bug 2189.
326 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
327 "panic the current process" to "deliberately defer". The panic log is
328 still written with the problem list name; the mail and reject logs now
329 get a temp-reject line for the message that was being handled, saying
330 something like "domains check lookup or other defer". The SMTP 451
331 message is still "Temporary local problem".
333 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
334 A crafted sequence of BDAT commands could result in in-use memory beeing
335 freed. CVE-2017-16943.
337 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
338 from SMTP input. Previously it was always done; now only done for DATA
339 and not BDAT commands. CVE-2017-16944.
341 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
342 to the message (such as an overlong header line). Previously this was
343 not done and we did not exit BDAT mode. Followon from the previous item
344 though a different problem.
350 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
351 than -2003 did; needs libidn2 in addition to libidn.
353 JH/02 The path option on a pipe transport is now expanded before use.
355 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
356 Patch provided by "Björn", documentation fix added too.
358 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
359 missing a wire-to-host endian conversion.
361 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
362 close after a BDAT command line could be taken as a following command,
363 giving a synch failure. Fix by only checking for synch immediately
364 before acknowledging the chunk.
366 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
367 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
368 macro. Patches provided by Josh Soref.
370 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
371 Previously we did not; the RFC seems ambiguous and VRFY is not listed
372 by IANA as a service extension. However, John Klensin suggests that we
375 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
376 the dkim code may be unix-mode line endings rather than smtp wire-format
377 CRLF, so prepend a CR to any bare LF.
379 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
380 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
382 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
383 the first were themselves being wrongly included in the feed into dkim
384 processing; with most chunk sizes in use this resulted in an incorrect
385 body hash calculated value.
387 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
388 DKIM signature block, for verification. Although advised against by
389 standards it is specifically not ruled illegal.
391 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
393 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
394 missing a body hash (the bh= tag).
396 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
397 It seems that HAProxy sends the Proxy Protocol information in clear and
398 only then does a TLS startup, so do the same.
400 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
401 TCP connections (such as for Spamd) unless the daemon successfully set
402 Fast Open mode on its listening sockets. This fixes breakage seen on
403 too-old kernels or those not configured for Fast Open, at the cost of
404 requiring both directions being enabled for TFO, and TFO never being used
405 by non-daemon-related Exim processes.
407 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
408 endings, at least on the first header line. Try to canonify any that get
409 past that check, despite the cost.
411 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
412 now limited to an arbitrary five deep, while parsing addresses with the
413 strip_excess_angle_brackets option enabled.
415 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
416 instead leave the unprompted TLS handshake in socket buffer for the
417 TLS library to consume.
419 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
421 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
423 JH/16 Drop variables when they go out of scope. Memory management drops a whole
424 region in one operation, for speed, and this leaves assigned pointers
425 dangling. Add checks run only under the testsuite which checks all
426 variables at a store-reset and panics on a dangling pointer; add code
427 explicitly nulling out all the variables discovered. Fixes one known
428 bug: a transport crash, where a dangling pointer for $sending_ip_address
429 originally assigned in a verify callout, is re-used.
431 PP/06 Drop '.' from @INC in various Perl scripts.
433 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
435 PP/08 Reduce a number of compilation warnings under clang; building with
436 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
437 should be warning-free.
439 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
441 HS/01 Fix portability problems introduced by PP/08 for platforms where
442 realloc(NULL) is not equivalent to malloc() [SunOS et al].
444 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
445 chunk. This allows us to accept broken chunked messages. We need a more
446 general solution here.
448 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
449 already-broken messages in the queue.
451 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
453 JH/19 Fix reference counting bug in routing-generated-address tracking.
459 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
460 supports it and a size is available (ie. the sending peer gave us one).
462 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
463 years of being deprecated). The replacements are the ACLs
464 acl_smtp_mime and acl_not_smtp_mime.
466 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
467 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
468 or even an in-clear connection were permitted. Now, if the host lookup
469 was dnssec and dane was requested then the host is only used if the
470 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
471 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
472 if one fails this test.
473 This means that a poorly-configured remote DNS will make it incommunicado;
474 but it protects against a DNS-interception attack on it.
476 JH/04 Bug 1810: make continued-use of an open smtp transport connection
477 non-noisy when a race steals the message being considered.
479 JH/05 If main configuration option tls_certificate is unset, generate a
480 self-signed certificate for inbound TLS connections.
482 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
483 in rewrites and routers.
485 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
486 and logged a warning sing 4.83; now they are a configuration file error.
488 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
489 (lacking @domain). Apply the same qualification processing as RCPT.
491 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
493 JH/10 Support ${sha256:} applied to a string (as well as the previous
496 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
497 a cutthrough deliver is pending, as we always want to make a connection.
498 This also avoids re-routing the message when later placing the cutthrough
499 connection after a verify cache hit.
500 Do not update it with the verify result either.
502 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
503 when routing results in more than one destination address.
505 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
506 signing (which inhibits the cutthrough capability). Previously only
507 the presence of an option was tested; now an expansion evaluating as
508 empty is permissible (obviously it should depend only on data available
509 when the cutthrough connection is made).
511 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
512 the relevant preceding SMTP command did not note the pipelining mode.
514 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
515 Previously they were not counted.
517 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
518 as one having no matching records. Previously we deferred the message
519 that needed the lookup.
521 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
522 distinguished as "(=".
524 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
525 for missing MX records. Previously it only worked for missing A records.
527 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
529 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
530 after the data-go-ahead and data-ack. Patch from Jason Betts.
532 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
533 even for a "none" policy. Patch from Tony Meyer.
535 JH/22 Fix continued use of a connection for further deliveries. If a port was
536 specified by a router, it must also match for the delivery to be
539 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
540 When one of the recipients of a message was unsuitable for the connection
541 (has no matching addresses), we lost track of needing to mark it
542 deferred. As a result mail would be lost.
544 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
546 JH/25 Decoding ACL controls is now done using a binary search; the source code
547 takes up less space and should be simpler to maintain. Merge the ACL
548 condition decode tables also, with similar effect.
550 JH/26 Fix problem with one_time used on a redirect router which returned the
551 parent address unchanged. A retry would see the parent address marked as
552 delivered, so not attempt the (identical) child. As a result mail would
555 JH/27 Fix a possible security hole, wherein a process operating with the Exim
556 UID can gain a root shell. Credit to http://www.halfdog.net/ for
557 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
560 JH/28 Enable {spool,log} filesystem space and inode checks as default.
561 Main config options check_{log,spool}_{inodes,space} are now
562 100 inodes, 10MB unless set otherwise in the configuration.
564 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
565 Previously it only applied to the main-section connection policy
568 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
570 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
571 by me. Added RFC7919 DH primes as an alternative.
573 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
574 are not in the system include path.
576 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
577 GnuTLS, when a session startup failed (eg because the client disconnected)
578 Exim did stdio operations after fclose. This was exposed by a recent
579 change which nulled out the file handle after the fclose.
581 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
582 signed directly by the cert-signing cert, rather than an intermediate
583 OCSP-signing cert. This is the model used by LetsEncrypt.
585 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
587 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
588 an incoming connection.
590 HS/02 Bug 1802: Do not half-close the connection after sending a request
593 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
594 fallback to "prime256v1".
596 JH/34 SECURITY: Use proper copy of DATA command in error message.
597 Could leak key material. Remotely exploitable. CVE-2016-9963.
603 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
604 and 3.4.4 - once the server is enabled to respond to an OCSP request
605 it does even when not requested, resulting in a stapling non-aware
606 client dropping the TLS connection.
608 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
609 support variable-length bit vectors. No functional change.
611 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
612 The I= interface field on outgoing lines is now after the H= remote
613 host field, same as incoming lines. There is a separate
614 outgoing_interface log selector which allows you to disable the
617 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
618 If not running log_selector +smtp_connection the mainlog would be held
619 open indefinitely after a "too many connections" event, including to a
620 deleted file after a log rotate. Leave the per net connection logging
621 leaving it open for efficiency as that will be quickly detected by the
622 check on the next write.
624 HS/01 Bug 1671: Fix post transport crash.
625 Processing the wait-<transport> messages could crash the delivery
626 process if the message IDs didn't exist for some reason. When
627 using 'split_spool_directory=yes' the construction of the spool
628 file name failed already, exposing the same netto behaviour.
630 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
631 mime_regex ACL conditions.
633 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
634 to DSN fail messages (bounces): remote IP, remote greeting, remote response
635 to HELO, local diagnostic string.
637 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
638 log line to debug. Even when configured with a tls authenticator many
639 client connections are expected to not authenticate in this way, so
640 an authenticate fail is not an error.
642 HS/02 Add the Exim version string to the process info. This way exiwhat
643 gives some more detail about the running daemon.
645 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
646 matter for fast-change records such as DNSBLs.
648 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
649 retry record, even if constant. There may be multiple transports with
650 different interface settings and the retry behaviour needs to be kept
653 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
655 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
657 JH/10 Bug 840: fix log_defer_output option of pipe transport
659 JH/11 Bug 830: use same host for all RCPTS of a message, even under
660 hosts_randomize. This matters a lot when combined with mua_wrapper.
662 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
663 ${quote_pgsql:<string>} operator.
665 JH/13 Bug 1708: avoid misaligned access in cached lookup.
667 JH/14 Change header file name for freeradius-client. Relevant if compiling
668 with Radius support; from the Gentoo tree and checked under Fedora.
670 JH/15 Bug 1712: Introduce $prdr_requested flag variable
672 JH/16 Bug 1714: Permit an empty string as expansion result for transport
673 option transport_filter, meaning no filtering.
675 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
677 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
678 defaults to "*" (all hosts). The variable is now available when not built
679 with TLS, default unset, mainly to enable keeping the testsuite sane.
680 If a server certificate is not supplied (via tls_certificate) an error is
681 logged, and clients will find TLS connections fail on startup. Presumably
682 they will retry in-clear.
683 Packagers of Exim are strongly encouraged to create a server certificate
684 at installation time.
686 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
687 with the $config_file variable.
689 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
690 in transport context, after the attempt, and per-recipient. The latter type
691 is per host attempted. The event data is the error message, and the errno
692 information encodes the lookup type (A vs. MX) used for the (first) host,
693 and the trailing two digits of the smtp 4xx response.
695 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
696 to write to mainlog (or rejectlog, paniclog) in the window between file
697 creation and permissions/ownership being changed. Particularly affects
698 installations where exicyclog is run as root, rather than exim user;
699 result is that the running daemon panics and dies.
701 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
703 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
704 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
705 "pri" and "weight". Note that the previous implicit priority given by the
706 list order is no longer honoured.
708 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
711 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
712 by defining SUPPORT_SOCKS.
714 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
715 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
716 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
717 variables are renamed to proxy_{local,external}_{address,port}.
719 JH/27 Move Internationalisation support from Experimental to mainline, enabled
720 for a build by defining SUPPORT_I18N
722 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
723 of the query string, and make ${quote_redis:} do that quoting.
725 JH/29 Move Events support from Experimental to mainline, enabled by default
726 and removable for a build by defining DISABLE_EVENT.
728 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
730 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
731 cached by the daemon.
733 JH/32 Move Redis support from Experimental to mainline, enabled for a build
734 by defining LOOKUP_REDIS. The libhiredis library is required.
736 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
737 keys are given for lookup.
739 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
740 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
741 only supported when built with TLS support. The PolarSSL SHA routines
742 are still used when the TLS library is too old for convenient support.
744 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
745 openssl_options), for security. OpenSSL forces this from version 1.1.0
746 server-side so match that on older versions.
748 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
749 allocation for $value could be released as the expansion processing
750 concluded, but leaving the global pointer active for it.
752 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
753 and to use the domains and local_parts ACL conditions.
755 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
756 incorrectly not doubled on cutthrough transmission, hence seen as a
757 body-termination at the receiving system - resulting in truncated mails.
758 Commonly the sender saw a TCP-level error, and retransmitted the message
759 via the normal store-and-forward channel. This could result in duplicates
760 received - but deduplicating mailstores were liable to retain only the
761 initial truncated version.
763 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
765 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
767 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
768 we're in there, support oversigning also; bug 1309.
770 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
772 HS/04 Add support for keep_environment and add_environment options.
774 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
775 either intentional arithmetic overflow during PRNG, or testing config-
778 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
779 delivery resulted in actual delivery. Cancel cutthrough before DATA
782 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
783 rejects a recipient: pass the reject to the originator.
785 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
786 Many were false-positives and ignorable, but it's worth fixing the
789 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
790 for the new environment-manipulation done at startup. Move the routines
791 from being local to tls.c to being global via the os.c file.
793 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
794 an extract embedded as result-arg for a map, the first arg for extract
795 is unavailable so we cannot tell if this is a numbered or keyed
796 extraction. Accept either.
802 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
805 JH/02 The smtp transport option "multi_domain" is now expanded.
807 JH/03 The smtp transport now requests PRDR by default, if the server offers
810 JH/04 Certificate name checking on server certificates, when exim is a client,
811 is now done by default. The transport option tls_verify_cert_hostnames
812 can be used to disable this per-host. The build option
813 EXPERIMENTAL_CERTNAMES is withdrawn.
815 JH/05 The value of the tls_verify_certificates smtp transport and main options
816 default to the word "system" to access the system default CA bundle.
817 For GnuTLS, only version 3.0.20 or later.
819 JH/06 Verification of the server certificate for a TLS connection is now tried
820 (but not required) by default. The verification status is now logged by
821 default, for both outbound TLS and client-certificate supplying inbound
824 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
827 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
828 Status Notification (bounce) messages are now MIME format per RFC 3464.
829 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
830 under the control of the dsn_advertise_hosts option, and routers may
831 have a dsn_lasthop option.
833 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
834 default, modifiable by a malware= option. The list separator for
835 the options can now be changed in the usual way. Bug 68.
837 JH/10 The smtp_receive_timeout main option is now expanded before use.
839 JH/11 The incoming_interface log option now also enables logging of the
840 local interface on delivery outgoing connections.
842 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
843 if the interface and destination host and port all match.
845 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
848 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
849 Patch from Andrew Lewis.
851 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
852 now supports optional time-restrictions, weighting, and priority
853 modifiers per server. Patch originally by <rommer@active.by>.
855 JH/16 The spamd_address main option now supports a mixed list of local
856 and remote servers. Remote servers can be IPv6 addresses, and
857 specify a port-range.
859 JH/17 Bug 68: The spamd_address main option now supports an optional
860 timeout value per server.
862 JH/18 Bug 1581: Router and transport options headers_add/remove can
863 now have the list separator specified.
865 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
868 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
871 JH/21 Support for the A6 type of dns record is withdrawn.
873 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
874 rather than the verbs used.
876 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
877 from 255 to 1024 chars.
879 JH/24 Verification callouts now attempt to use TLS by default.
881 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
882 are generic router options now. The defaults didn't change.
884 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
885 Original patch from Alexander Shikoff, worked over by JH.
887 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
888 files if ZCAT_COMMAND is not executable.
890 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
892 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
894 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
895 Normally benign, it bites when the pair was led to by a CNAME;
896 modern usage is to not canonicalize the domain to a CNAME target
897 (and we were inconsistent anyway for A-only vs AAAA+A).
899 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
901 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
902 when evaluating $sender_host_dnssec.
904 JH/31 Check the HELO verification lookup for DNSSEC, adding new
905 $sender_helo_dnssec variable.
907 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
909 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
911 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
913 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
914 documented as working, but never had. Support all but $spam_report.
916 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
917 added for tls authenticator.
919 HS/03 Add perl_taintmode main config option
925 TL/01 When running the test suite, the README says that variables such as
926 no_msglog_check are global and can be placed anywhere in a specific
927 test's script, however it was observed that placement needed to be near
928 the beginning for it to behave that way. Changed the runtest perl
929 script to read through the entire script once to detect and set these
930 variables, reset to the beginning of the script, and then run through
931 the script parsing/test process like normal.
933 TL/02 The BSD's have an arc4random API. One of the functions to induce
934 adding randomness was arc4random_stir(), but it has been removed in
935 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
936 function when detected.
938 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
939 cause callback expansion.
941 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
942 syntax errors in an expansion can be treated as a string instead of
943 logging or causing an error, due to the internal use of bool_lax
944 instead of bool when processing it.
946 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
947 server certificates when making smtp deliveries.
949 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
951 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
953 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
955 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
956 Merged patch from Sebastian Wiedenroth.
958 JH/05 Fix results-pipe from transport process. Several recipients, combined
959 with certificate use, exposed issues where response data items split
960 over buffer boundaries were not parsed properly. This eventually
961 resulted in duplicates being sent. This issue only became common enough
962 to notice due to the introduction of connection certificate information,
963 the item size being so much larger. Found and fixed by Wolfgang Breyha.
965 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
966 size buffer was used, resulting in syntax errors when an expansion
969 JH/07 Add support for directories of certificates when compiled with a GnuTLS
970 version 3.3.6 or later.
972 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
973 is EXPERIMENTAL_EVENT, the main-configuration and transport options
974 both become "event_action", the variables become $event_name, $event_data
975 and $event_defer_errno. There is a new variable $verify_mode, usable in
976 routers, transports and related events. The tls:cert event is now also
977 raised for inbound connections, if the main configuration event_action
980 TL/06 In test suite, disable OCSP for old versions of openssl which contained
981 early OCSP support, but no stapling (appears to be less than 1.0.0).
983 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
984 server certificate names available under the smtp transport option
985 "tls_verify_cert_hostname" now do not permit multi-component wildcard
988 JH/10 Time-related extraction expansions from certificates now use the main
989 option "timezone" setting for output formatting, and are consistent
990 between OpenSSL and GnuTLS compilations. Bug 1541.
992 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
993 encoded parameter in the incoming message. Bug 1558.
995 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
996 include certificate info, eximon was claiming there were spoolfile
999 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1001 JH/14 Log delivery-related information more consistently, using the sequence
1002 "H=<name> [<ip>]" wherever possible.
1004 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1005 are problematic for Debian distribution, omit them from the release
1008 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1010 JH/16 Fix string representation of time values on 64bit time_t architectures.
1013 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1014 output list separator was used.
1019 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1020 checkers that were complaining about end of non-void function with no
1023 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1024 This was a regression introduced in 4.83 by another bugfix.
1026 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1028 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1029 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1035 TF/01 Correctly close the server side of TLS when forking for delivery.
1037 When a message was received over SMTP with TLS, Exim failed to clear up
1038 the incoming connection properly after forking off the child process to
1039 deliver the message. In some situations the subsequent outgoing
1040 delivery connection happened to have the same fd number as the incoming
1041 connection previously had. Exim would try to use TLS and fail, logging
1042 a "Bad file descriptor" error.
1044 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1045 utilities have not been installed.
1047 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1048 temporary space as the ACL may create new global variables.
1050 TL/01 LDAP support uses per connection or global context settings, depending
1051 upon the detected version of the libraries at build time.
1053 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1054 to extract and use the src ip:port in logging and expansions as if it
1055 were a direct connection from the outside internet. PPv2 support was
1056 updated based on HAProxy spec change in May 2014.
1058 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1060 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1061 Properly escape header and check for NULL return.
1063 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1066 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1068 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1069 characters in header names, implemented as a verify condition.
1070 Contributed by Michael Fischer v. Mollard.
1072 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1073 results permerror and temperror. Previous values are deprecated but
1074 still accepted. In a future release, err_perm and err_temp will be
1075 completely removed, which will be a backward incompatibility if the
1076 ACL tests for either of these two old results. Patch contributed by
1077 user bes-internal on the mailing list.
1079 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1081 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1082 selectors, in both main and reject logs.
1084 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1087 JH/07 Add malware type "sock" for talking to simple daemon.
1089 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1091 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1092 routers/transports under cutthrough routing.
1094 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1095 numbers. Touch up "bool" conditional to keep the same definition.
1097 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1099 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1101 JH/12 Expand items in router/transport headers_add or headers_remove lists
1102 individually rather than the list as a whole. Bug 1452.
1104 Required for reasonable handling of multiple headers_ options when
1105 they may be empty; requires that headers_remove items with embedded
1106 colons must have them doubled (or the list-separator changed).
1108 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1109 view the policy declared in the DMARC record. Currently, $dmarc_status
1110 is a combined value of both the record presence and the result of the
1113 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1115 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1116 dnslookup router and the smtp transport (applying to the forward
1119 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1120 of ldap servers used for a specific lookup. Patch provided by Heiko
1123 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1124 New variable $lookup_dnssec_authenticated for observability.
1126 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1127 Patch submitted by Lars Timman.
1129 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1131 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1132 Requires trusted mode and valid format message id, aborts otherwise.
1133 Patch contributed by Heiko Schlichting.
1135 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1136 certextract with support for various fields. Bug 1358.
1138 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1139 is requested by default, modifiable by smtp transport option
1142 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1143 operate on certificate variables to give certificate fingerprints
1144 Also new ${sha256:cert_variable}.
1146 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1148 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1149 Christian Aistleitner.
1151 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1153 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1154 file. Patch from Wolfgang Breyha.
1156 JH/25 Expand the coverage of the delivery $host and $host_address to
1157 client authenticators run in verify callout. Bug 1476.
1159 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1160 align with daemon_smtp_ports. Bug 72.
1162 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1163 support and error reporting did not work properly.
1165 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1166 and is readable. Patch from Andrew Colin Kissa.
1168 TL/14 Enhance documentation of ${run expansion and how it parses the
1169 commandline after expansion, particularly in the case when an
1170 unquoted variable expansion results in an empty value.
1172 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1174 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1175 by renaming away from it. Addresses GH issue 15, reported by
1178 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1180 TL/15 SECURITY: prevent double expansion in math comparison functions
1181 (can expand unsanitized data). Not remotely exploitable.
1188 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1190 PP/02 Make -n do something, by making it not do something.
1191 When combined with -bP, the name of an option is not output.
1193 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1196 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1197 $sender_host_name and config options to manage this, and basic check
1200 PP/05 DSCP support for outbound connections and control modifier for inbound.
1202 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1203 (Only plugin which currently uses this is kerberos4, which nobody should
1204 be using, but we should make it available and other future plugins might
1205 conceivably use it, even though it would break NAT; stuff *should* be
1206 using channel bindings instead).
1208 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1209 name; added for Sendmail compatibility; requires admin caller.
1210 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1211 just ignore it); requires trusted caller.
1212 Also parse but ignore: -Ac -Am -X<logfile>
1215 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1217 TL/02 Add +smtp_confirmation as a default logging option.
1219 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1220 Patch by Magnus Holmgren from 2007-02-20.
1222 TL/04 Bugzilla 1281 - Spec typo.
1223 Bugzilla 1283 - Spec typo.
1224 Bugzilla 1290 - Spec grammar fixes.
1226 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1228 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1230 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1231 mailing list by by Dmitry Isaikin.
1233 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1235 JH/02 Support "G" suffix to numbers in ${if comparisons.
1237 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1239 NM/01 Bugzilla 1197 - Spec typo
1240 Bugzilla 1196 - Spec examples corrections
1242 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1244 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1245 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1248 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1249 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1251 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1252 "acl {{name}{arg}...}", and optional args on acl condition
1255 JH/05 Permit multiple router/transport headers_add/remove lines.
1257 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1259 JH/07 Avoid using a waiting database for a single-message-only transport.
1260 Performance patch from Paul Fisher. Bugzilla 1262.
1262 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1265 JH/09 Add $headers_added variable, with content from use of ACL modifier
1266 add_header (but not yet added to the message). Bugzilla 199.
1268 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1269 Pulled from Bugzilla 817 by Wolfgang Breyha.
1271 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1273 (nb: this is the same fix as in Exim 4.80.1)
1275 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1278 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1281 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1283 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1284 advertise SMTP AUTH mechanism to us, instead of a generic
1285 protocol violation error. Also, make Exim more robust to bad
1286 data from the Dovecot auth socket.
1288 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1290 When a queue runner is handling a message, Exim first routes the
1291 recipient addresses, during which it prunes them based on the retry
1292 hints database. After that it attempts to deliver the message to
1293 any remaining recipients. It then updates the hints database using
1296 So if a recipient address works intermittently, it can get repeatedly
1297 deferred at routing time. The retry hints record remains fresh so the
1298 address never reaches the final cutoff time.
1300 This is a fairly common occurrence when a user is bumping up against
1301 their storage quota. Exim had some logic in its local delivery code
1302 to deal with this. However it did not apply to per-recipient defers
1303 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1305 This change adds a proper retry rule check during routing so that the
1306 final cutoff time is checked against the message's age. We only do
1307 this check if there is an address retry record and there is not a
1308 domain retry record; this implies that previous attempts to handle
1309 the address had the retry_use_local_parts option turned on. We use
1310 this as an approximation for the destination being like a local
1311 delivery, as in LMTP.
1313 I suspect this new check makes the old local delivery cutoff check
1314 redundant, but I have not verified this so I left the code in place.
1316 TF/02 Correct gecos expansion when From: is a prefix of the username.
1318 Test 0254 submits a message to Exim with the header
1322 When I ran the test suite under the user fanf2, Exim expanded
1323 the header to contain my full name, whereas it should have added
1324 a Resent-Sender: header. It erroneously treats any prefix of the
1325 username as equal to the username.
1327 This change corrects that bug.
1329 GF/01 DCC debug and logging tidyup
1330 Error conditions log to paniclog rather than rejectlog.
1331 Debug lines prefixed by "DCC: " to remove any ambiguity.
1333 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1335 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1336 Bug spotted by Jeremy Harris; was flawed since initial commit.
1337 Would have resulted in OCSP responses post-SNI triggering an Exim
1338 NULL dereference and crash.
1340 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1342 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1343 Bug detection, analysis and fix by Samuel Thibault.
1344 Bugzilla 1331, Debian bug #698092.
1346 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1348 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1349 Server implementation by Todd Lyons, client by JH.
1350 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1351 config variable "prdr_enable" controls whether the server
1352 advertises the facility. If the client requests PRDR a new
1353 acl_data_smtp_prdr ACL is called once for each recipient, after
1354 the body content is received and before the acl_smtp_data ACL.
1355 The client is controlled by both of: a hosts_try_prdr option
1356 on the smtp transport, and the server advertisement.
1357 Default client logging of deliveries and rejections involving
1358 PRDR are flagged with the string "PRDR".
1360 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1361 fclose(). Diagnosis by Todd Lyons.
1363 PP/17 Update configure.default to handle IPv6 localhost better.
1364 Patch by Alain Williams (plus minor tweaks).
1367 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1368 This is now consistent with GnuTLS, and is now documented: the
1369 previous undocumented portable approach to treating the option as
1370 unset was to force an expansion failure. That still works, and
1371 an empty string is now equivalent.
1373 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1374 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1375 not performing validation itself.
1377 PP/20 Added force_command boolean option to pipe transport.
1378 Patch from Nick Koston, of cPanel Inc.
1380 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1383 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1385 PP/21 Fix eximon continuous updating with timestamped log-files.
1386 Broken in a format-string cleanup in 4.80, missed when I repaired the
1387 other false fix of the same issue.
1388 Report and fix from Heiko Schlichting.
1391 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1392 Report from Prashanth Katuri.
1394 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1395 It's SecureTransport, so affects any MacOS clients which use the
1396 system-integrated TLS libraries, including email clients.
1398 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1399 using a MIME ACL for non-SMTP local injection.
1400 Report and assistance in diagnosis by Warren Baker.
1402 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1404 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1406 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1407 last id that failed so it may be referenced in subsequent ACL's.
1409 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1412 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1413 ldap library initialization, allowing self-signed CA's to be
1414 used. Also properly sets require_cert option later in code by
1415 using NULL (global ldap config) instead of ldap handle (per
1416 session). Bug diagnosis and testing by alxgomz.
1418 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1419 the src/util/ subdirectory.
1421 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1422 renamed to Transport Post Delivery Action by Jeremy Harris, as
1425 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1426 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1427 redis_servers = needs to be configured which will be used by the redis
1428 lookup. Patch from Warren Baker, of The Packet Hub.
1430 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1432 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1433 hostname or reverse DNS when processing a host list. Used suggestions
1434 from multiple comments on this bug.
1436 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1438 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1439 Missed a few lines, added it to make the runtest require no keyboard
1442 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1443 contains upper case chars. Make router use caseful_local_part.
1445 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1446 support when GnuTLS has been built with p11-kit.
1452 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1454 This, or similar/improved, will also be change PP/11 of 4.82.
1460 PP/01 Handle short writes when writing local log-files.
1461 In practice, only affects FreeBSD (8 onwards).
1462 Bugzilla 1053, with thanks to Dmitry Isaikin.
1464 NM/01 Bugzilla 949 - Documentation tweak
1466 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1469 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1471 PP/02 Implemented gsasl authenticator.
1473 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1475 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1476 `pkg-config foo` for cflags/libs.
1478 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1479 with rest of GSASL and with heimdal_gssapi.
1481 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1482 `pkg-config foo` for cflags/libs for the TLS implementation.
1484 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1485 properties get this fed in as external SSF. A number of robustness
1486 and debugging improvements to the cyrus_sasl authenticator.
1488 PP/08 cyrus_sasl server now expands the server_realm option.
1490 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1491 Patch by Jeremy Harris.
1493 PP/10 Added dbmjz lookup type.
1495 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1497 PP/12 MAIL args handles TAB as well as SP, for better interop with
1498 non-compliant senders.
1499 Analysis and variant patch by Todd Lyons.
1501 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1502 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1503 Patch from Dirk Mueller <dmueller@suse.com>
1505 PP/13 tls_peerdn now print-escaped for spool files.
1506 Observed some $tls_peerdn in wild which contained \n, which resulted
1507 in spool file corruption.
1509 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1510 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1511 or write after TLS renegotiation, which otherwise led to messages
1514 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1515 as a tracking header (ie: a signed header comes before the signature).
1516 Patch from Wolfgang Breyha.
1518 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1519 comma-sep list; embedded commas doubled.
1521 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1523 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1525 Report and patch from Dmitry Banschikov.
1527 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1528 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1529 failures. We appear to now support TLS1.1+ with Exim.
1531 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1532 lets Exim select keys and certificates based upon TLS SNI from client.
1533 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1534 before an outbound SMTP session. New log_selector, +tls_sni.
1536 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1537 NULL dereference. Report and patch from Alun Jones.
1539 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1540 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1541 Not seeing resolver debug output on NetBSD, but suspect this is a
1542 resolver implementation change.
1544 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1545 Left warnings. Added "eximon gdb" invocation mode.
1547 PP/21 Defaulting "accept_8bitmime" to true, not false.
1549 PP/22 Added -bw for inetd wait mode support.
1551 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1552 locate the relevant includes and libraries. Made this the default.
1554 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1555 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1557 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1558 This may cause build issues on older platforms.
1560 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1561 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1562 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1563 Added SNI support via GnuTLS too.
1564 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1566 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1568 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1569 Applied second patch from Janne, implementing suggestion to default
1570 multiple-strings-in-record handling to match SPF spec.
1572 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1574 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1575 read-only, out of scope).
1576 Patch from Wolfgang Breyha, report from Stuart Northfield.
1578 PP/29 Fix three issues highlighted by clang analyser static analysis.
1579 Only crash-plausible issue would require the Cambridge-specific
1580 iplookup router and a misconfiguration.
1581 Report from Marcin Mirosław.
1583 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1585 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1586 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1587 As part of this, removing so much warning spew let me fix some minor
1588 real issues in debug logging.
1590 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1591 assignment on my part. Fixed.
1593 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1594 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1595 Janne Snabb (who went above and beyond: thank you).
1597 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1598 string otherwise requires a connection and a bunch more work and it's
1599 relatively easy to get wrong. Should also expose TLS library linkage
1602 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1603 64-bit ${eval} (JH/03).
1605 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1606 GNU libc to support some of the 64-bit stuff, should not lead to
1607 conflicts. Defined before os.h is pulled in, so if a given platform
1608 needs to override this, it can.
1610 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1611 protection layer was required, which is not implemented.
1612 Bugzilla 1254, patch from Wolfgang Breyha.
1614 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1615 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1616 tls_dhparam take prime identifiers. Also unbreak combination of
1617 OpenSSL+DH_params+TLSSNI.
1619 PP/39 Disable SSLv2 by default in OpenSSL support.
1625 PP/01 Solaris build fix for Oracle's LDAP libraries.
1626 Bugzilla 1109, patch from Stephen Usher.
1628 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1630 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1633 TF/02 Fix a couple more cases where we did not log the error message
1634 when unlink() failed. See also change 4.74-TF/03.
1636 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1637 lock up or crash if it happened to be inside a call to libc when it
1638 got a SIGUSR1 from exiwhat.
1640 The SIGUSR1 handler appends the current process status to the process
1641 log which is later printed by exiwhat. It used to use the general
1642 purpose logging code to do this, but several functions it calls are
1643 not safe for signals.
1645 The new output code in the SIGUSR1 handler is specific to the process
1646 log, and simple enough that it's easy to inspect for signal safety.
1647 Removing some special cases also simplifies the general logging code.
1648 Removing the spurious timestamps from the process log simplifies
1651 TF/04 Improved ratelimit ACL condition.
1653 The /noupdate option has been deprecated in favour of /readonly which
1654 has clearer semantics. The /leaky, /strict, and /readonly update modes
1655 are mutually exclusive. The update mode is no longer included in the
1656 database key; it just determines when the database is updated. (This
1657 means that when you upgrade Exim will forget old rate measurements.)
1659 Exim now checks that the per_* options are used with an update mode that
1660 makes sense for the current ACL. For example, when Exim is processing a
1661 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1662 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1663 must specify per_mail/readonly. If you omit the update mode it defaults to
1664 /leaky where that makes sense (as before) or /readonly where required.
1666 The /noupdate option is now undocumented but still supported for
1667 backwards compatibility. It is equivalent to /readonly except that in
1668 ACLs where /readonly is required you may specify /leaky/noupdate or
1669 /strict/noupdate which are treated the same as /readonly.
1671 A useful new feature is the /count= option. This is a generalization
1672 of the per_byte option, so that you can measure the throughput of other
1673 aggregate values. For example, the per_byte option is now equivalent
1674 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1676 The per_rcpt option has been generalized using the /count= mechanism
1677 (though it's more complicated than the per_byte equivalence). When it is
1678 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1679 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1680 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1681 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1682 Note that using per_rcpt with a non-readonly update mode in more than
1683 one ACL will cause the recipients to be double-counted. (The per_mail
1684 and per_byte options don't have this problem.)
1686 The handling of very low rates has changed slightly. If the computed rate
1687 is less than the event's count (usually one) then this event is the first
1688 after a long gap. In this case the rate is set to the same as this event's
1689 count, so that the first message of a spam run is counted properly.
1691 The major new feature is a mechanism for counting the rate of unique
1692 events. The new per_addr option counts the number of different
1693 recipients that someone has sent messages to in the last time period. It
1694 behaves like per_rcpt if all the recipient addresses are different, but
1695 duplicate recipient addresses do not increase the measured rate. Like
1696 the /count= option this is a general mechanism, so the per_addr option
1697 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1698 example, measure the rate that a client uses different sender addresses
1699 with the options per_mail/unique=$sender_address. There are further
1700 details in the main documentation.
1702 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1704 TF/06 Removed a few PCRE remnants.
1706 TF/07 Automatically extract Exim's version number from tags in the git
1707 repository when doing development or release builds.
1709 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1710 Bugzilla 879. Patch from Paul Fisher.
1712 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1713 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1716 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1718 PP/05 Variable $av_failed, true if the AV scanner deferred.
1719 Bugzilla 1078. Patch from John Horne.
1721 PP/06 Stop make process more reliably on build failure.
1722 Bugzilla 1087. Patch from Heiko Schlittermann.
1724 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1725 Bugzilla 1089. Patch from Heiko Schlittermann.
1727 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1728 Bugzilla 1131. Patch from Holger Weiß.
1730 PP/09 Handle IPv6 addresses with SPF.
1731 Bugzilla 860. Patch from Wolfgang Breyha.
1733 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1735 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1738 PP/11 match_* no longer expand right-hand-side by default.
1739 New compile-time build option, EXPAND_LISTMATCH_RHS.
1740 New expansion conditions, "inlist", "inlisti".
1742 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1744 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1746 PP/14 fix log_write() format string regression from TF/03.
1747 Bugzilla 1152. Patch from Dmitry Isaikin.
1753 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1755 PP/02 Harmonised TLS library version reporting; only show if debugging.
1756 Layout now matches that introduced for other libraries in 4.74 PP/03.
1758 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1760 PP/04 New "dns_use_edns0" global option.
1762 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1765 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1766 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1768 TK/01 Updated PolarSSL code to 0.14.2.
1769 Bugzilla 1097. Patch from Andreas Metzler.
1771 PP/07 Catch divide-by-zero in ${eval:...}.
1772 Fixes bugzilla 1102.
1774 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1777 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1778 format-string attack -- SECURITY: remote arbitrary code execution.
1780 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1781 time unintentionally subject to list matching rules, letting the header
1782 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1783 arbitrary string expansion). This allowed for information disclosure.
1785 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1786 INT_MIN/-1 -- value coerced to INT_MAX.
1792 NM/01 Workaround for PCRE version dependency in version reporting
1795 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1796 This fixes portability to compilers other than gcc, notably
1797 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1799 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1800 makefiles for portability to HP-UX and POSIX correctness.
1802 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1803 Also via indented variable definition in the Makefile.
1804 (Debugging by Oliver Heesakkers).
1806 PP/02 Restore caching of spamd results with expanded spamd_address.
1807 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1809 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1810 Improves build reliability. Fix from: Frank Elsner
1812 NM/02 Fix wide character breakage in the rfc2047 coding
1813 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1815 NM/03 Allow underscore in dnslist lookups
1816 Fixes bug 1026. Patch from Graeme Fowler
1818 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1819 Code patches from Adam Ciarcinski of NetBSD.
1821 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1824 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1825 is logged, to avoid truncation. Patch from John Horne.
1827 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1828 Patch from Jakob Hirsch.
1830 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1831 SQL string expansion failure details.
1832 Patch from Andrey Oktyabrski.
1834 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1835 Patch from Simon Arlott.
1837 PP/09 New lookups functionality failed to compile on old gcc which rejects
1838 extern declarations in function scope.
1839 Patch from Oliver Fleischmann
1841 PP/10 Use sig_atomic_t for flags set from signal handlers.
1842 Check getgroups() return and improve debugging.
1843 Fixed developed for diagnosis in bug 927 (which turned out to be
1846 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1847 Patch from Mark Zealey.
1849 PP/12 Bugzilla 1056: Improved spamd server selection.
1850 Patch from Mark Zealey.
1852 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1853 Based on patch from Heiko Schlittermann.
1855 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1856 Patch from Uwe Doering, sign-off by Michael Haardt.
1858 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1859 variable declaration deep within a block. Bug and patch from
1862 PP/15 lookups-Makefile IRIX compatibility coercion.
1864 PP/16 Make DISABLE_DKIM build knob functional.
1866 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1867 Patch by Simon Arlott
1869 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1870 variable argument macros. Our copy now differs from upstream.
1876 TF/01 Failure to get a lock on a hints database can have serious
1877 consequences so log it to the panic log.
1879 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1880 controlled using the smtp_confirmation log selector.
1882 TF/03 Include the error message when we fail to unlink a spool file.
1884 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1885 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1886 for maintaining out-of-tree patches for some time.
1888 PP/01 Bugzilla 139: Documentation and portability issues.
1889 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1890 Handle per-OS dynamic-module compilation flags.
1892 PP/02 Let /dev/null have normal permissions.
1893 The 4.73 fixes were a little too stringent and complained about the
1894 permissions on /dev/null. Exempt it from some checks.
1895 Reported by Andreas M. Kirchwitz.
1897 PP/03 Report version information for many libraries, including
1898 Exim version information for dynamically loaded libraries. Created
1899 version.h, now support a version extension string for distributors
1900 who patch heavily. Dynamic module ABI change.
1902 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1903 privilege escalation vulnerability whereby the Exim run-time user
1904 can cause root to append content of the attacker's choosing to
1907 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1910 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1911 If dropping privileges for untrusted macros, we disabled normal logging
1912 on the basis that it would fail; for the Exim run-time user, this is not
1913 the case, and it resulted in successful deliveries going unlogged.
1914 Fixed. Reported by Andreas Metzler.
1920 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1921 only prepend for the Resent-* case. Fixes regression introduced in
1922 Exim 4.70 by NM/22 for Bugzilla 607.
1924 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1925 increasing numbers of administrators be bitten by this.
1927 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1929 PP/03 Bugzilla 994: added openssl_options main configuration option.
1931 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1933 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1935 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1937 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1938 without a peer certificate, leading to a segfault because of an
1939 assumption that peers always have certificates. Be a little more
1940 paranoid. Problem reported by Martin Tscholak.
1942 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1943 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1944 NB: ClamAV planning to remove STREAM in "middle of 2010".
1945 CL also introduces -bmalware, various -d+acl logging additions and
1946 more caution in buffer sizes.
1948 PP/09 Implemented reverse_ip expansion operator.
1950 PP/10 Bugzilla 937: provide a "debug" ACL control.
1952 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1954 PP/12 Bugzilla 973: Implement --version.
1956 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1958 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1960 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1962 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1963 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1964 ignore trailing whitespace.
1966 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1968 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1971 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1972 Notification from Dr Andrew Aitchison.
1974 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1975 ExtendedDetectionInfo response format.
1976 Notification from John Horne.
1978 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1981 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1982 XSL and documented dependency on system catalogs, with examples of how
1985 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1988 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1989 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1990 configuration file which is writeable by the Exim user or group.
1992 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1993 of configuration files to cover files specified with the -C option if
1994 they are going to be used with root privileges, not just the default
1997 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1998 option (effectively making it always true).
2000 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2001 files to be used while preserving root privileges.
2003 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2004 that rogue child processes cannot use them.
2006 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2007 run-time user, instead of root.
2009 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2010 Exim run-time user without dropping privileges.
2012 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2013 result string, instead of calling string_vformat() twice with the same
2016 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2017 for other users. Others should always drop root privileges if they use
2018 -C on the command line, even for a whitelisted configure file.
2020 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2022 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2028 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2029 $data_path, and $header_path variables; fixed documentation bugs and
2032 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2033 exipick to access non-standard spools, including the "frozen" queue
2036 NM/01 Bugzilla 965: Support mysql stored procedures.
2037 Patch from Alain Williams
2039 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2041 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2042 Patch from Andreas Metzler
2044 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2045 Patch from Kirill Miazine
2047 NM/05 Bugzilla 671: Added umask to procmail example.
2049 JJ/03 installed exipick 20100323.0, fixing doc bug
2051 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2052 directory. Notification and patch from Dan Rosenberg.
2054 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2056 TK/02 Improve log output when DKIM signing operation fails.
2058 MH/01 Treat the transport option dkim_domain as a colon separated
2059 list, not as a single string, and sign the message with each element,
2060 omitting multiple occurences of the same signer.
2062 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2063 Bugzilla 985, 986. Patch by Simon Arlott
2065 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2066 Patch by Simon Arlott
2068 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2069 MBX locking. Notification from Dan Rosenberg.
2075 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2077 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2079 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2081 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2083 NM/04 Bugzilla 915: Flag broken perl installation during build.
2089 TK/01 Added patch by Johannes Berg that expands the main option
2090 "spamd_address" if it starts with a dollar sign.
2092 TK/02 Write list of recipients to X-Envelope-Sender header when building
2093 the mbox-format spool file for content scanning (suggested by Jakob
2096 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2097 (http://www.dcc-servers.net/) support via dccifd. Activated by
2098 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2100 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2101 by Mark Daniel Reidel <mr@df.eu>.
2103 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2104 When building exim an external PCRE library is now needed -
2105 PCRE is a system library on the majority of modern systems.
2106 See entry on PCRE_LIBS in EDITME file.
2108 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2109 conversation. Added nologin parameter to request.
2110 Patch contributed by Kirill Miazine.
2112 TF/01 Do not log submission mode rewrites if they do not change the address.
2114 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2116 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2117 log files in place. Contributed by Roberto Lima.
2119 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2121 TF/03 Bugzilla 615: When checking the local_parts router precondition
2122 after a local_part_suffix or local_part_prefix option, Exim now
2123 does not use the address's named list lookup cache, since this
2124 contains cached lookups for the whole local part.
2126 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2127 Robert Millan. Documentation is in experimental-spec.txt.
2129 TF/04 Bugzilla 668: Fix parallel build (make -j).
2131 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2133 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2134 Patch provided by Jan Srzednicki.
2136 TF/05 Leading white space used to be stripped from $spam_report which
2137 wrecked the formatting. Now it is preserved.
2139 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2140 that they are available at delivery time.
2142 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2144 TF/08 TLS error reporting now respects the incoming_interface and
2145 incoming_port log selectors.
2147 TF/09 Produce a more useful error message if an SMTP transport's hosts
2148 setting expands to an empty string.
2150 NM/06 Bugzilla 744: EXPN did not work under TLS.
2151 Patch provided by Phil Pennock.
2153 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2154 Patch provided by Richard Godbee.
2156 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2157 acl_smtp_notquit, added index entry.
2159 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2160 Patch provided by Eugene Bujak.
2162 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2163 accept(). Patch provided by Maxim Dounin.
2165 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2166 Patch provided by Phil Pennock.
2168 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2170 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2171 Patch provided by Brad "anomie" Jorsch.
2173 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2174 Patch provided by Dean Brooks.
2176 TK/05 Add native DKIM support (does not depend on external libraries).
2178 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2179 Patch provided by Graeme Fowler.
2181 NM/16 Bugzilla 851: Documentation example syntax fix.
2183 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2185 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2188 NM/19 Bugzilla 745: TLS version reporting.
2189 Patch provided by Phil Pennock.
2191 NM/20 Bugzilla 167: bool: condition support.
2192 Patch provided by Phil Pennock.
2194 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2195 clients. Patch provided by Phil Pennock.
2197 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2198 Patch provided by Brad "anomie" Jorsch.
2200 NM/23 Bugzilla 687: Fix misparses in eximstats.
2201 Patch provided by Heiko Schlittermann.
2203 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2204 Patch provided by Heiko Schlittermann.
2206 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2207 plus update to original patch.
2209 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2211 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2212 Patch provided by David Brownlee.
2214 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2216 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2218 NM/30 Bugzilla 888: TLS documentation bugfixes.
2220 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2222 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2223 Unlike the original bugzilla I have changed all shell scripts in src tree.
2225 NM/33 Bugzilla 898: Transport filter timeout fix.
2226 Patch by Todd Rinaldo.
2228 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2229 Patch by Serge Demonchaux.
2231 NM/35 Bugzilla 39: Base64 decode bug fixes.
2232 Patch by Jakob Hirsch.
2234 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2236 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2238 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2240 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2246 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2247 ALT-N's libdkim that I have put here:
2248 http://duncanthrax.net/exim-experimental/
2250 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2251 were called 'true' and it seems that C99 defines that as a reserved
2252 keyword to be used with 'bool' variable types. That means you could
2253 not include C99-style headers which use bools without triggering
2254 build errors in sieve.c.
2256 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2257 as mailq or other aliases. Changed the --help handling significantly
2258 to do whats expected. exim_usage() emits usage/help information.
2260 SC/01 Added the -bylocaldomain option to eximstats.
2262 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2264 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2266 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2272 PH/01 Another patch from the Sieve maintainer.
2274 PH/02 When an IPv6 address is converted to a string for single-key lookup
2275 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2276 dots are used instead of colons so that keys in lsearch files need not
2277 contain colons. This was done some time before quoting was made available
2278 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2279 (notated using the quote facility) so as to distinguish them from IPv4
2280 keys. This meant that lookups for IP addresses in host lists did not work
2281 for iplsearch lookups.
2283 This has been fixed by arranging for IPv6 addresses to be expressed with
2284 colons if the lookup type is iplsearch. This is not incompatible, because
2285 previously such lookups could never work.
2287 The situation is now rather anomalous, since one *can* have colons in
2288 ordinary lsearch keys. However, making the change in all cases is
2289 incompatible and would probably break a number of configurations.
2291 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2294 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2295 conversion specification without a maximum field width, thereby enabling
2296 a rogue spamd server to cause a buffer overflow. While nobody in their
2297 right mind would setup Exim to query an untrusted spamd server, an
2298 attacker that gains access to a server running spamd could potentially
2299 exploit this vulnerability to run arbitrary code as the Exim user.
2301 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2302 $primary_hostname instead of what libspf2 thinks the hosts name is.
2304 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2305 a directory entry by the name of the lookup key. Previously, if a
2306 symlink pointed to a non-existing file or a file in a directory that
2307 Exim lacked permissions to read, a lookup for a key matching that
2308 symlink would fail. Now it is enough that a matching directory entry
2309 exists, symlink or not. (Bugzilla 503.)
2311 PH/03 The body_linecount and body_zerocount variables are now exported in the
2314 PH/04 Added the $dnslist_matched variable.
2316 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2317 This means they are set thereafter only if the connection becomes
2320 PH/06 Added the client_condition to authenticators so that some can be skipped
2321 by clients under certain conditions.
2323 PH/07 The error message for a badly-placed control=no_multiline_responses left
2324 "_responses" off the end of the name.
2326 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2328 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2329 (without spaces) instead of just copying the configuration text.
2331 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2333 PH/11 Added $max_received_linelength.
2335 PH/12 Added +ignore_defer and +include_defer to host lists.
2337 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2338 way in which PCRE > 7.0 is built.
2340 PH/14 Implemented queue_only_load_latch.
2342 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2343 MAIL command. The effect was to mangle the value on 64-bit systems.
2345 PH/16 Another patch from the Sieve maintainer.
2347 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2349 PH/18 If a system quota error occurred while trying to create the file for
2350 a maildir delivery, the message "Mailbox is full" was not appended to the
2351 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2352 only to a quota excession during the actual writing of the file.
2354 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2355 characters?) which causes problems in log lines. The DN values are now
2356 passed through string_printing() before being added to log lines.
2358 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2359 and InterBase are left for another time.)
2361 PH/21 Added message_body_newlines option.
2363 PH/22 Guard against possible overflow in moan_check_errorcopy().
2365 PH/23 POSIX allows open() to be a macro; guard against that.
2367 PH/24 If the recipient of an error message contained an @ in the local part
2368 (suitably quoted, of course), incorrect values were put in $domain and
2369 $local_part during the evaluation of errors_copy.
2375 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2376 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2379 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2380 issue a MAIL command.
2382 PH/02 In an ACL statement such as
2384 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2386 if a client was not listed at all, or was listed with a value other than
2387 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2388 the condition was not true (as it should be), so access was not denied.
2389 The bug was that the ! inversion was incorrectly passed on to the second
2390 item. This has been fixed.
2392 PH/03 Added additional dnslists conditions == and =& which are different from
2393 = and & when the dns lookup returns more than one IP address.
2395 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2396 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2398 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2399 FSYNC, which compiles an option called disable_fsync that allows for
2400 bypassing fsync(). The documentation is heavily laced with warnings.
2402 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2404 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2405 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2406 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2407 including adding "make clean"; (3) Added -fPIC when compiling the test
2408 dynamically loaded module, to get rid of a warning.
2410 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2411 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2412 = 0s. The bug is otherwise harmless.
2414 PH/07 There was a bug in the dovecot authenticator such that the value of
2415 $auth1 could be overwritten, and so not correctly preserved, after a
2416 successful authentication. This usually meant that the value preserved by
2417 the server_setid option was incorrect.
2419 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2421 PH/09 Installed PCRE release 7.0.
2423 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2424 run for batched SMTP input. It is now run at the start of every message
2425 in the batch. While fixing this I discovered that the process information
2426 (output by running exiwhat) was not always getting set for -bs and -bS
2427 input. This is fixed, and it now also says "batched" for BSMTP.
2429 PH/11 Added control=no_pipelining.
2431 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2432 patch, slightly modified), and move the expansion of helo_data till after
2433 the connection is made in the smtp transport (so it can use these
2436 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2438 PH/14 Added log_selector = +pid.
2440 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2442 PH/16 Add ${if forany and ${if forall.
2444 PH/17 Added dsn_from option to vary the From: line in DSNs.
2446 PH/18 Flush SMTP output before performing a callout, unless control =
2447 no_callout_flush is set.
2449 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2450 was true (the default) a successful delivery failed to delete the retry
2451 item, thus causing premature timeout of the address. The bug is now
2454 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2456 PH/21 Long custom messages for fakedefer and fakereject are now split up
2457 into multiline responses in the same way that messages for "deny" and
2458 other ACL rejections are.
2460 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2461 with slight modification.
2463 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2464 draft, changing the syntax and factoring some duplicate code".
2466 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2467 for deliveries of the second and subsequent messages over the same SMTP
2470 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2471 ${reduce, with only minor "tidies".
2473 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2475 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2476 expansion side effects.
2478 PH/27 When a message times out after an over-quota error from an Exim-imposed
2479 quota, the bounce message says "mailbox is full". This message was not
2480 being given when it was a system quota that was exceeded. It now should
2483 MH/03 Made $recipients available in local_scan(). local_scan() already has
2484 better access to the recipient list through recipients_list[], but
2485 $recipients can be useful in postmaster-provided expansion strings.
2487 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2488 in the case of a MAIL command with additional options following the
2489 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2490 were accidentally chopped off.
2492 PH/29 SMTP synchronization checks are implemented when a command is read -
2493 there is a check that no more input is waiting when there shouldn't be
2494 any. However, for some commands, a delay in an ACL can mean that it is
2495 some time before the response is written. In this time, more input might
2496 arrive, invalidly. So now there are extra checks after an ACL has run for
2497 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2498 pipelining has not been advertised.
2500 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2502 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2503 returned in the SMTP response when smtp_return_error_details was false.
2504 This has been fixed.
2506 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2507 instead of the C I/O that was originally supplied, because problems were
2508 reported on Solaris.
2510 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2511 Exim which did not show up earlier: it was assuming that a call to
2512 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2513 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2514 was a macro that became an assignment, so it seemed to work. This has
2515 changed to a proper function call with a void return, hence the compile
2516 error. Exim's code has been fixed.
2518 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2521 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2523 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2525 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2526 because it turns out to be too big or there is a timeout) there is no
2527 "Completed" line in the log. When some messages of this type were
2528 selected by exigrep, they were listed as "not completed". Others were
2529 picked up by some special patterns. I have improved the selection
2530 criteria to be more general.
2532 PH/38 The host_find_failed option in the manualroute router can now be set
2533 to "ignore", to completely ignore a host whose IP address cannot be
2534 found. If all hosts are ignored, the behaviour is controlled by the new
2535 host_all_ignored option.
2537 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2538 homing or because of multiple MX records with /mx) generated more than
2539 one IP address, and the following item turned out to be the local host,
2540 all the secondary addresses of the first item were incorrectly removed
2541 from the list, along with the local host and any following hosts (which
2542 is what is supposed to happen).
2544 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2545 whoever called Exim into the -H file. In the case of the daemon it was
2546 behaving confusingly. When first started, it used values for whoever
2547 started the daemon, but after a SIGHUP it used the Exim user (because it
2548 calls itself on a restart). I have changed the code so that it now always
2551 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2552 message are rejected with the same error (e.g. no authentication or bad
2553 sender address), and a DATA command is nevertheless sent (as can happen
2554 with PIPELINING or a stupid MUA), the error message that was given to the
2555 RCPT commands is included in the rejection of the DATA command. This is
2556 intended to be helpful for MUAs that show only the final error to their
2559 PH/42 Another patch from the Sieve maintainer.
2561 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2562 Eximstats - Fixed some broken HTML links and added missing column headers
2564 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2565 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2567 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2569 PH/43 Yet another patch from the Sieve maintainer.
2571 PH/44 I found a way to check for a TCP/IP connection going away before sending
2572 the response to the final '.' that terminates a message, but only in the
2573 case where the client has not sent further data following the '.'
2574 (unfortunately, this is allowed). However, in many cases there won't be
2575 any further data because there won't be any more messages to send. A call
2576 to select() can be used: if it shows that the input is "ready", there is
2577 either input waiting, or the socket has been closed. An attempt to read
2578 the next input character can distinguish the two cases. Previously, Exim
2579 would have sent an OK response which the client would never have see.
2580 This could lead to message repetition. This fix should cure that, at
2581 least in a lot of common cases.
2583 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2584 advertised in response to EHLO.
2590 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2591 fixed by 4.65/MH/01 (is this a record?) are fixed:
2593 (i) An empty string was always treated as zero by the numeric comparison
2594 operators. This behaviour has been restored.
2596 (ii) It is documented that the numeric comparison operators always treat
2597 their arguments as decimal numbers. This was broken in that numbers
2598 starting with 0 were being interpreted as octal.
2600 While fixing these problems I realized that there was another issue that
2601 hadn't been noticed. Values of message_size_limit (both the global option
2602 and the transport option) were treated as octal if they started with 0.
2603 The documentation was vague. These values are now always treated as
2604 decimal, and I will make that clear in the documentation.
2610 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2611 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2614 MH/01 Don't check that the operands of numeric comparison operators are
2615 integers when their expansion is in "skipping" mode (fixes bug
2616 introduced by 4.64-PH/07).
2618 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2619 child addresses, Exim now panics and dies. Previously, because the count
2620 is held in a short int, deliveries were likely to be lost. As such a
2621 large number of recipients for a single message is ridiculous
2622 (performance will be very, very poor), I have chosen to impose a limit
2623 rather than extend the field.
2629 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2630 leftover -K file (the existence of which was triggered by #402).
2631 While we were at it, introduced process PID as part of the -K
2632 filename. This should rule out race conditions when creating
2635 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2636 processing considerably. Previous code took too long for large mails,
2637 triggering a timeout which in turn triggers #401.
2639 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2640 in the DK code in transports.c. sendfile() is not really portable,
2641 hence the _LINUX specificness.
2643 TF/01 In the add_headers option to the mail command in an Exim filter,
2644 there was a bug that Exim would claim a syntax error in any
2645 header after the first one which had an odd number of characters
2648 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2649 callout verification, Exim cached a "reject" for the entire domain. This
2650 is correct for most verifications, but it is not correct for a recipient
2651 verification with use_sender or use_postmaster set, because in that case
2652 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2653 case of MAIL FROM:<> rejection from other early rejections (e.g.
2654 rejection of HELO). When verifying a recipient using a non-null MAIL
2655 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2656 Whatever the result of the callout, the value of the domain cache is
2657 left unchanged (for any other kind of callout, getting as far as trying
2658 RCPT means that the domain itself is ok).
2660 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2663 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2664 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2665 session, the daemon ignores ECONNECT errors and logs others; it now
2666 ignores EPIPE as well.
2668 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2669 (quoted-printable decoding).
2671 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2672 later the small subsequent patch to fix an introduced bug.
2674 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2676 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2678 PH/08 An error is now given if message_size_limit is specified negative.
2680 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2681 to be given (somewhat) arbitrary names.
2683 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2686 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2687 miscellaneous code fixes
2689 PH/10 Added the log_reject_target ACL modifier to specify where to log
2692 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2693 hostname. This is wrong, because it relates to the incoming message (and
2694 probably the interface on which it is arriving) and not to the outgoing
2695 callout (which could be using a different interface). This has been
2696 changed to use the value of the helo_data option from the smtp transport
2697 instead - this is what is used when a message is actually being sent. If
2698 there is no remote transport (possible with a router that sets up host
2699 addresses), $smtp_active_hostname is used.
2701 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2702 tweaks were necessary in order to get it to work (see also 21 below):
2703 (a) The code assumed that strncpy() returns a negative number on buffer
2704 overflow, which isn't the case. Replaced with Exim's string_format()
2706 (b) There were several signed/unsigned issues. I just did the minimum
2707 hacking in of casts. There is scope for a larger refactoring.
2708 (c) The code used strcasecmp() which is not a standard C function.
2709 Replaced with Exim's strcmpic() function.
2710 (d) The code set only $1; it now sets $auth1 as well.
2711 (e) A simple test gave the error "authentication client didn't specify
2712 service in request". It would seem that Dovecot has changed its
2713 interface. Fortunately there's a specification; I followed it and
2714 changed what the client sends and it appears to be working now.
2716 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2719 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2720 address A is aliased to B and C, where B exists and C does not. Without
2721 -v the output is "A verified" because verification stops after a
2722 successful redirection if more than one address is generated. However,
2723 with -v the child addresses are also verified. Exim was outputting "A
2724 failed to verify" and then showing the successful verification for C,
2725 with its parentage. It now outputs "B failed to verify", showing B's
2726 parentage before showing the successful verification of C.
2728 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2729 look up a TXT record in a specific list after matching in a combined
2732 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2733 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2734 they consult the DNS. I had assumed they would set it the way they
2735 wanted; and indeed my experiments on Linux seem to show that in some
2736 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2737 To be on the safe side, however, I have now made the interface to
2738 host_find_byname() similar to host_find_bydns(), with an argument
2739 containing the DNS resolver options. The host_find_byname() function now
2740 sets these options at its start, just as host_find_bydns() does. The smtp
2741 transport options dns_qualify_single and dns_search_parents are passed to
2742 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2743 of host_find_byname() use the default settings of RES_DEFNAMES
2744 (qualify_single) but not RES_DNSRCH (search_parents).
2746 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2747 spool_read_header() do less string testing, by means of a preliminary
2748 switch on the second character of optional "-foo" lines. (This is
2749 overdue, caused by the large number of possibilities that now exist.
2750 Originally there were few.) While I was there, I also converted the
2751 str(n)cmp tests so they don't re-test the leading "-" and the first
2752 character, in the hope this might squeeze out yet more improvement.
2754 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2755 flag allowing group syntax was set by the header_syntax check but not
2756 turned off, possible causing trouble later; (2) The flag was not being
2757 set at all for the header_verify test, causing "group"-style headers to
2758 be rejected. I have now set it in this case, and also caused header_
2759 verify to ignore an empty address taken from a group. While doing this, I
2760 came across some other cases where the code for allowing group syntax
2761 while scanning a header line wasn't quite right (mostly, not resetting
2762 the flag correctly in the right place). These bugs could have caused
2763 trouble for malformed header lines. I hope it is now all correct.
2765 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2766 with the "reply" argument non-NULL. The code, however (which originally
2767 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2768 but it didn't always do it. This confused somebody who was copying the
2769 code for some other use. I have removed all the tests.
2771 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2772 feature that was used to support insecure browsers during the U.S. crypto
2773 embargo. It requires special client support, and Exim is probably the
2774 only MTA that supported it -- and would never use it because real RSA is
2775 always available. This code has been removed, because it had the bad
2776 effect of slowing Exim down by computing (never used) parameters for the
2777 RSA_EXPORT functionality.
2779 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2780 authenticator to fail if there's a tab character in the incoming data
2781 (there should never be unless someone is messing about, as it's supposed
2782 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2783 if the connection is using TLS or if the remote IP is the same as the
2784 local IP, and the "valid-client-cert option" if a client certificate has
2787 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2788 authenticators. This can be used for authorization after authentication
2789 succeeds. (In the case of plaintext, it servers for both authentication
2792 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2793 if any retry times were supplied.
2795 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2796 connection, where there is no client IP address to check. In this
2797 situation, the verify now always succeeds.
2799 PH/25 Applied John Jetmore's -Mset patch.
2801 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2803 PH/27 In a string expansion for a processed (not raw) header when multiple
2804 headers of the same name were present, leading whitespace was being
2805 removed from all of them, but trailing whitespace was being removed only
2806 from the last one. Now trailing whitespace is removed from each header
2807 before concatenation. Completely empty headers in a concatenation (as
2808 before) are ignored.
2810 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2811 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2813 PH/29 [Removed. This was a change that I later backed out, and forgot to
2814 correct the ChangeLog entry (that I had efficiently created) before
2815 committing the later change.]
2817 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2818 address errors (4xx response to RCPT) over the same connection as other
2819 messages routed to the same hosts. Such deliveries are always "forced",
2820 so retry times are not inspected. This resulted in far too many retries
2821 for the affected addresses. The effect occurred only when there were more
2822 hosts than the hosts_max_try setting in the smtp transport when it had
2823 the 4xx errors. Those hosts that it had tried were not added to the list
2824 of hosts for which the message was waiting, so if all were tried, there
2825 was no problem. Two fixes have been applied:
2827 (i) If there are any address or message errors in an SMTP delivery, none
2828 of the hosts (tried or untried) are now added to the list of hosts
2829 for which the message is waiting, so the message should not be a
2830 candidate for sending over the same connection that was used for a
2831 successful delivery of some other message. This seems entirely
2832 reasonable: after all the message is NOT "waiting for some host".
2833 This is so "obvious" that I'm not sure why it wasn't done
2834 previously. Hope I haven't missed anything, but it can't do any
2835 harm, as the worst effect is to miss an optimization.
2837 (ii) If, despite (i), such a delivery is accidentally attempted, the
2838 routing retry time is respected, so at least it doesn't keep
2839 hammering the server.
2841 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2842 in ${readsocket because some servers need this prod.
2844 PH/32 Added some extra debug output when updating a wait-xxx database.
2846 PH/33 The hint "could be header name not terminated by colon", which has been
2847 given for certain expansion errors for a long time, was not being given
2848 for the ${if def:h_colon_omitted{... case.
2850 PH/34 The spec says: "With one important exception, whenever a domain list is
2851 being scanned, $domain contains the subject domain." There was at least
2852 one case where this was not true.
2854 PH/35 The error "getsockname() failed: connection reset by peer" was being
2855 written to the panic log as well as the main log, but it isn't really
2856 panic-worthy as it just means the connection died rather early on. I have
2857 removed the panic log writing for the ECONNRESET error when getsockname()
2860 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2861 runs only) independently of the message's sender address. This meant
2862 that, if the 4xx error was in fact related to the sender, a different
2863 message to the same recipient with a different sender could confuse
2864 things. In particular, this can happen when sending to a greylisting
2865 server, but other circumstances could also provoke similar problems.
2866 I have changed the default so that the retry time for these errors is now
2867 based a combination of the sender and recipient addresses. This change
2868 can be overridden by setting address_retry_include_sender=false in the
2871 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2872 remote server are returned as part of bounce messages. This was not
2873 happening for LMTP over a pipe (the lmtp transport), but now it is the
2874 same for both kinds of LMTP.
2876 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2877 in header lines that were in fact CNAMEs. This is no longer the case.
2879 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2880 and queue runs started by the daemon processed all messages. This has
2881 been fixed so that -R and -S can now usefully be given with -q<time>.
2883 PH/40 Import PCRE release 6.7 (fixes some bugs).
2885 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2887 PH/42 Give an error if -q is specified more than once.
2889 PH/43 Renamed the variables $interface_address and $interface_port as
2890 $received_ip_address and $received_port, to make it clear that these
2891 values apply to message reception, and not to the outgoing interface when
2892 a message is delivered. (The old names remain recognized, of course.)
2894 PH/44 There was no timeout on the connect() call when using a Unix domain
2895 socket in the ${readsocket expansion. There now is.
2897 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2898 be meaningful with "accept".
2900 SC/01 Eximstats V1.43
2901 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2903 SC/02 Eximstats V1.44
2904 Use a glob alias rather than an array ref in the generated
2905 parser. This improves both readability and performance.
2907 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2908 Collect SpamAssassin and rejection statistics.
2909 Don't display local sender or destination tables unless
2910 there is data to show.
2911 Added average volumes into the top table text output.
2913 SC/04 Eximstats V1.46
2914 Collect data on the number of addresses (recipients)
2915 as well as the number of messages.
2917 SC/05 Eximstats V1.47
2918 Added 'Message too big' to the list of mail rejection
2919 reasons (thanks to Marco Gaiarin).
2921 SC/06 Eximstats V1.48
2922 Mainlog lines which have GMT offsets and are too short to
2923 have a flag are now skipped.
2925 SC/07 Eximstats V1.49 (Alain Williams)
2926 Added the -emptyok flag.
2928 SC/08 Eximstats V1.50
2929 Fixes for obtaining the IP address from reject messages.
2931 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2932 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2933 whitespace changes from 4.64-PH/27
2935 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2938 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2939 are found, allow negative numbers in numeric criteria)
2941 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2943 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2944 to match changes made in 4.64-PH/43
2946 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2948 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2949 "domain" is actually an IP address, and force a failure. This locks out
2950 those revolvers/nameservers that support "A-for-A" lookups, in
2951 contravention of the specifications.
2953 PH/48 When a host name was looked up from an IP address, and the subsequent
2954 forward lookup of the name timed out, the host name was left in
2955 $sender_host_name, contrary to the specification.
2957 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2958 restricted to single-key lookups, Exim was not diagnosing an error if
2959 * or *@ was used with a query-style lookup.
2961 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2963 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2964 long ago, but noone interested enough thought of it. Let's just say that
2965 the "1.1" means that there are some new functions that weren't there at
2966 some point in the past.
2968 PH/51 Error processing for expansion failure of helo_data from an smtp
2969 transport during callout processing was broken.
2971 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2972 tested/used via the -bh/-bhc/-bs options.
2974 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2975 bug, fixed in subsequent PCRE releases).
2977 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2978 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2980 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2986 SC/01 Use a glob alias rather than an array ref in eximstats generated
2987 parser. This improves both readability and performance.
2989 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2990 Don't display local sender or destination tables in eximstats unless
2991 there is data to show.
2992 Added average volumes into the eximstats top table text output.
2994 SC/03 Collect data on the number of addresses (recipients) as well
2995 as the number of messages in eximstats.
2997 TF/01 Correct an error in the documentation for the redirect router. Exim
2998 does (usually) call initgroups() when daemonizing.
3000 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3001 with consistent privilege compared to when running as a daemon.
3003 TF/03 Note in the spec that $authenticated_id is not set for local
3004 submissions from trusted users.
3006 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3007 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3009 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3010 by adding some example configuration directives to the default
3011 configuration file. A little bit of work is required to uncomment the
3012 directives and define how usernames and passwords are checked, but
3013 there is now a framework to start from.
3015 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3016 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3017 without this. I don't know how relevant this is to other LDAP libraries.
3019 PH/02 Add the verb name to the "unknown ACL verb" error.
3021 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3023 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3025 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3026 directory not expanded when it should be if an expanded home directory
3027 was set for the address (which is overridden by the transport).
3029 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3032 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3033 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3034 because it is too late at that time, and has no effect.
3036 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3037 security issue with \' (bugzilla #107). I could not use the
3038 PQescapeStringConn() function, because it needs a PGconn value as one of
3041 PH/08 When testing addresses using -bt, indicate those final addresses that
3042 are duplicates that would not cause an additional delivery. At least one
3043 person was confused, thinking that -bt output corresponded to deliveries.
3044 (Suppressing duplicates isn't a good idea as you lose the information
3045 about possibly different redirections that led to the duplicates.)
3047 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3048 systems where poll() doesn't work, in particular OS X.
3050 PH/10 Added more information to debugging output for retry time not reached.
3052 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3053 operations in malware.c.
3055 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3058 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3059 syslog_duplication set false, log lines that would normally be written
3060 both the the main log and to the reject log were not written to syslog at
3063 PH/14 In the default configuration, change the use of "message" in ACL warn
3064 statements to "add_header".
3066 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3067 not followed by a command (e.g. "seen endif").
3069 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3070 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3073 PH/17 Added extra conditions to the default value of delay_warning_condition
3077 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3078 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3079 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3082 The Auto-Submitted: and various List- headers are standardised, whereas I
3083 don't think Precedence: ever was.
3085 PH/18 Refactored debugging code in route_finduser() to show more information,
3086 in particular, the error code if getpwnam() issues one.
3088 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3089 This is apparently needed in addition to the PH/07 change above to avoid
3090 any possible encoding problems.
3092 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3093 but not after initializing Perl.
3095 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3096 output them only if debugging. By default they are written stderr,
3097 apparently, which is not desirable.
3099 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3102 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3105 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3107 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3108 authenticated or an ident call has been made. Suppress the default
3109 values for $authenticated_id and $authenticated_sender (but permit -oMai
3110 and -oMas) when testing with -bh.
3112 PH/24 Re-jigged the order of the tests in the default configuration so that the
3113 tests for valid domains and recipients precede the DNS black list and CSA
3114 tests, on the grounds that those ones are more expensive.
3116 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3117 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3118 This bug exists in every version of Exim that I still have, right back to
3121 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3122 However, an attempt to turn on case-sensitivity in a regex key by
3123 including (?-i) didn't work because the subject string was already
3124 lowercased, and the effects were non-intuitive. It turns out that a
3125 one-line patch can be used to allow (?-i) to work as expected.
3131 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3132 other effects) broke the use of negated acl sub-conditions.
3134 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3137 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3138 "Deny" causes Exim to reject the incoming connection with a 554 error.
3139 Unfortunately, if there is a major crisis, such as a disk failure,
3140 tcp-wrappers gives "deny", whereas what one would like would be some
3141 kind of temporary error. A kludge has been added to help with this.
3142 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3143 554 error is used if errno is still zero or contains ENOENT (which occurs
3144 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3147 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3149 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3150 errors. Otherwise a message that provokes a temporary error (when other
3151 messages do not) can cause a whole host to time out.
3153 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3154 addresses were routed directly to files or pipes from a redirect router.
3155 File deliveries just didn't batch; pipe deliveries might have suffered
3158 PH/06 A failure to get a lock for a hints database would erroneously always say
3159 "Failed to get write lock", even when it was really a read lock.
3161 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3162 of 0600. This has been changed to use the value of the lockfile_mode
3163 option (which defaults to 0600).
3165 PH/08 Applied small patch from the Sieve maintainer.
3167 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3168 folder from quota calculations, a direct delivery into this folder messed
3169 up the contents of the maildirsize file. This was because the regex was
3170 used only to exclude .Trash (or whatever) when the size of the mailbox
3171 was calculated. There was no check that a delivery was happening into an
3172 excluded directory. This bug has been fixed by ignoring all quota
3173 processing for deliveries into excluded directories.
3175 PH/10 Added the maildirfolder_create_regex option to appendfile.
3181 PH/01 The code for finding all the local interface addresses on a FreeBSD
3182 system running IPv6 was broken. This may well have applied to all BSD
3183 systems, as well as to others that have similar system calls. The broken
3184 code found IPv4 interfaces correctly, but gave incorrect values for the
3185 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3186 that it would not match correctly against @[] and not recognize the IPv6
3189 PH/02 The ipliteral router was not recognizing addresses of the form user@
3190 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3192 PH/03 Added disable_ipv6.
3194 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3195 decoded form, because it is most often used to construct To: headers
3196 lines in autoreplies, and the decoded form may well be syntactically
3197 invalid. However, $reply_address has leading white space removed, and all
3198 newlines turned into spaces so that the autoreply transport does not
3201 PH/05 If group was specified without a user on a router, and no group or user
3202 was specified on a transport, the group from the router was ignored.
3204 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3205 for visible compile-time settings that can be used to change these
3206 numbers, for those that want even more. Backwards compatibility with old
3207 spool files has been maintained. However, going back to a previous Exim
3208 release will lost any variables that are in spool files.
3210 PH/07 Two small changes when running in the test harness: increase delay when
3211 passing a TCP/IP connection to a new process, in case the original
3212 process has to generate a bounce, and remove special handling of
3213 127.0.0.2 (sic), which is no longer necessary.
3215 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3216 be the same on different OS.
3218 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3221 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3222 whether --show-vars was specified or not
3224 JJ/02 exipick: Added support for new ACL variable spool format introduced
3227 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3228 syntactically invalid From: or Reply-to: line, and a filter used this to
3229 generate an autoreply, and therefore failed to obtain an address for the
3230 autoreply, Exim could try to deliver to a non-existent relative file
3231 name, causing unrelated and misleading errors. What now happens is that
3232 it logs this as a hard delivery error, but does not attempt to create a
3235 PH/11 The exinext utility has a -C option for testing purposes, but although
3236 the given file was scanned by exinext itself; it wasn't being passed on
3237 when Exim was called.
3239 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3240 an end-of-file indication when reading a command response.
3242 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3243 compiled. In many other places in Exim, IPv6 addresses are always
3244 recognized, so I have changed this. It also means that IPv4 domain
3245 literals of the form [IPV4:n.n.n.n] are now always recognized.
3247 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3248 used if the router is not running as root, for example, when verifying at
3249 ACL time, or when using -bh. The debugging output from this situation was
3250 non-existent - all you got was a failure to exec. I have made two
3253 (a) Failures to set uid/gid, the current directory, or a process leader
3254 in a subprocess such as that created by queryprogram now generate
3255 suitable debugging output when -d is set.
3257 (b) The queryprogram router detects when it is not running as root,
3258 outputs suitable debugging information if -d is set, and then runs
3259 the subprocess without attempting to change uid/gid.
3261 PH/15 Minor change to Makefile for building test_host (undocumented testing
3264 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3265 additional section of a DNS packet that returns MX or SRV records.
3266 Instead, it always explicitly searches for A/AAAA records. This avoids
3267 major problems that occur when a DNS server includes only records of one
3268 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3269 fixed another bug: if SRV records were looked up and the corresponding
3270 address records were *not* found in the additional section, the port
3271 values from the SRV records were lost.
3273 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3274 using the correct key (the original address) when searching the retry
3275 rules in order to find which one to use for generating the retry hint.
3277 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3278 adding the default one. Similarly, if it contains a Reply-To: header, the
3279 errors_reply_to option, if set, is not used.
3281 PH/19 When calculating a retry time, Exim used to measure the "time since
3282 failure" by looking at the "first failed" field in the retry record. Now
3283 it does not use this if it is later than than the arrival time of the
3284 message. Instead it uses the arrival time. This makes for better
3285 behaviour in cases where some deliveries succeed, thus re-setting the
3286 "first failed" field. An example is a quota failure for a huge message
3287 when small messages continue to be delivered. Without this change, the
3288 "time since failure" will always be short, possible causing more frequent
3289 delivery attempts for the huge message than are intended.
3290 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3292 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3293 $1, $2, $3) because the numerical variables can be reset during some
3294 expansion items (e.g. "match"), thereby losing the authentication data.
3296 PH/21 Make -bV show the size of off_t variables so that the test suite can
3297 decide whether to run tests for quotas > 2G.
3299 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3300 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3301 filecount value is greater than 2G or if a quota value is greater than 2G
3302 on a system where the size of off_t is not greater than 4, a panic error
3305 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3306 never match. The debug and -bh output now contains an explicit error
3307 message indicating a malformed IPv4 address or mask.
3309 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3310 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3311 PH/23 above applies.
3313 PH/25 Do not write to syslog when running in the test harness. The only
3314 occasion when this arises is a failure to open the main or panic logs
3315 (for which there is an explicit test).
3317 PH/26 Added the /no_tell option to "control=freeze".
3319 PH/27 If a host name lookup failed very early in a connection, for example, if
3320 the IP address matched host_lookup and the reverse lookup yielded a name
3321 that did not have a forward lookup, an error message of the form "no IP
3322 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3323 could be logged. Now it outputs the IP address instead of "NULL".
3325 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3326 allows the sender and the authenticated sender to be set when
3327 submitting a message from within Exim. Since child_open_exim() is
3328 documented for local_scan(), the new function should be too.
3330 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3331 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3332 results in an empty string is now treated as unset.
3334 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3336 PH/31 Added sender_verify_fail logging option.
3338 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3339 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3340 tidied the source and removed it altogether.
3342 PH/33 When a queue run was abandoned because the load average was too high, a
3343 log line was always written; now it is written only if the queue_run log
3344 selector is set. In addition, the log line for abandonment now contains
3345 information about the queue run such as the pid. This is always present
3346 in "start" and "stop" lines but was omitted from the "abandon" line.
3348 PH/34 Omit spaces between a header name and the colon in the error message that
3349 is given when verify = headers_syntax fails (if there are lots of them,
3350 the message gets confusing).
3352 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3353 names, as there are now some PTR records that contain slashes. This check
3354 is only to protect against broken name servers that fall over on strange
3355 characters, so the fact that it applies to all lookups doesn't matter.
3357 PH/36 Now that the new test suite is complete, we can remove some of the
3358 special code in Exim that was needed for the old test suite. For example,
3359 sorting DNS records because real resolvers return them in an arbitrary
3360 order. The new test suite's fake resolver always returns records in the
3363 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3364 bounces) except when queue_only is set, to avoid logging races between
3365 the different processes.
3367 PH/38 Panic-die if .include specifies a non-absolute path.
3369 PH/39 A tweak to the "H" retry rule from its user.
3371 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3372 a label. They prevented compilation on older perls.
3374 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3375 a warning to be raised on newish perls.
3377 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3378 on queue. Changes to match documented behaviour of showing count of
3379 messages matching specified criteria.
3381 PH/40 Changed the default ident timeout from 30s to 5s.
3383 PH/41 Added support for the use of login_cap features, on those BSD systems
3384 that have them, for controlling the resources used by pipe deliveries.
3386 PH/42 The content-scanning code uses fopen() to create files in which to put
3387 message data. Previously it was not paying any attention to the mode of
3388 the files. Exim runs with umask(0) because the rest of the code creates
3389 files with open(), and sets the required mode explicitly. Thus, these
3390 files were ending up world-writeable. This was not a big issue, because,
3391 being within the spool directory, they were not world-accessible. I have
3392 created a function called modefopen, which takes an additional mode
3393 argument. It sets umask(777), creates the file, chmods it to the required
3394 mode, then resets the umask. All the relevant calls to fopen() in the
3395 content scanning code have been changed to use this function.
3397 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3398 to 24 hours. This avoids potential overflow problems when processing G
3399 and H retry rules. I suspect nobody ever tinkers with this value.
3401 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3403 PH/45 When the plaintext authenticator is running as a client, the server's
3404 challenges are checked to ensure they are valid base64 strings. By
3405 default, the authentication attempt is cancelled if an invalid string is
3406 received. Setting client_ignore_invalid_base64 true ignores these errors.
3407 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3408 they are received. Thus, the responses can be made to depend on the
3409 challenges. If an invalid string is ignored, an empty string is placed in
3412 PH/46 Messages that are created by the autoreply transport now contains a
3413 References: header, in accordance with RFCs 2822 and 3834.
3415 PH/47 Added authenticated_sender_force to the smtp transport.
3417 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3419 PH/49 Installed latest patch from the Sieve maintainer.
3421 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3422 was also set, the appendfile transport was unnecessarily scanning a
3423 directory of message files (e.g. for maildir delivery) to find the count
3424 of files (along with the size), even though it did not need this
3425 information. It now does the scan only if it needs to find either the
3426 size of the count of files.
3428 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3430 PH/52 Two bugs concerned with error handling when the smtp transport is
3433 (i) Exim was not creating retry information for temporary errors given
3434 for individual recipients after the DATA command when the smtp transport
3435 was used in LMTP mode. This meant that they could be retried too
3436 frequently, and not timed out correctly.
3438 (ii) Exim was setting the flag that allows error details to be returned
3439 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3440 recipients that were returned after the DATA command.
3442 PH/53 This is related to PH/52, but is more general: for any failing address,
3443 when detailed error information was permitted to be returned to the
3444 sender, but the error was temporary, then after the final timeout, only
3445 "retry timeout exceeded" was returned. Now it returns the full error as
3446 well as "retry timeout exceeded".
3448 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3449 do this, and (what is worse) MTAs that accept it.
3451 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3452 will now be deprecated.
3454 PH/56 New os.c-cygwin from the Cygwin maintainer.
3456 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3457 formats (previously only available in exim formats via -bpr, -bpru,
3458 and -bpra. Now also available in native and exiqgrep formats)
3460 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3461 with very large, slow to parse queues
3463 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3465 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3467 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3468 responses to authentication challenges, though it was showing the
3469 challenges; (ii) I've removed the CR characters from the debug output for
3472 PH/58 Allow for the insertion of a newline as well as a space when a string
3473 is turned into more than one encoded-word during RFC 2047 encoding. The
3474 Sieve code now uses this.
3476 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3477 data_4xx, lost_connection, tls_required.
3479 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3480 message was being sent as an SMTP response.
3482 PH/61 Add -l and -k options to exicyclog.
3484 PH/62 When verifying, if an address was redirected to one new address, so that
3485 verification continued, and the new address failed or deferred after
3486 having set something in $address_data, the value of $address_data was not
3487 passed back to the ACL. This was different to the case when no
3488 redirection occurred. The value is now passed back in both cases.
3490 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3491 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3492 use login_cap.h, so on its own it isn't the distinguishing feature. The
3493 new name refers directly to the setclassresources() function.
3495 PH/65 Added configuration files for NetBSD3.
3497 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3499 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3500 is preferred over IPv4.
3502 PH/68 The bounce_return_message and bounce_return_body options were not being
3503 honoured for bounces generated during the reception of non-SMTP messages.
3504 In particular, this applied to messages rejected by the ACL. This bug has
3505 been fixed. However, if bounce_return_message is true and bounce_return_
3506 body is false, the headers that are returned for a non-SMTP message
3507 include only those that have been read before the error was detected.
3508 (In the case of an ACL rejection, they have all been read.)
3510 PH/69 The HTML version of the specification is now built in a directory called
3511 spec_html instead of spec.html, because the latter looks like a path with
3512 a MIME-type, and this confuses some software.
3514 PH/70 Catch two compiler warnings in sieve.c.
3516 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3517 function verify_get_ident() calls ip_connect() to connect a socket, but
3518 if the "connect()" function timed out, ip_connect() used to close the
3519 socket. However, verify_get_ident() also closes the socket later, and in
3520 between Exim writes to the log, which may get opened at this point. When
3521 the socket was closed in ip_connect(), the log could get the same file
3522 descriptor number as the socket. This naturally causes chaos. The fix is
3523 not to close the socket in ip_connect(); the socket should be closed by
3524 the function that creates it. There was only one place in the code where
3525 this was missing, in the iplookup router, which I don't think anybody now
3526 uses, but I've fixed it anyway.
3528 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3529 well as to direct DNS lookups. Otherwise the handling of names in host
3530 lists is inconsistent and therefore confusing.
3536 PH/01 Two changes to the default runtime configuration:
3538 (1) Move the checks for relay_from_hosts and authenticated clients from
3539 after to before the (commented out) DNS black list checks.
3541 (2) Add control=submission to the relay_from_hosts and authenticated
3542 clients checks, on the grounds that messages accepted by these
3543 statements are most likely to be submissions.
3545 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3547 (1) Generate an error if the third argument for the ${prvs expansion is
3550 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3553 (3) Reset the variables that are obtained from the first argument of
3554 ${prvscheck and used in the second argument before leaving the code,
3555 because their memory is reclaimed, so using them afterwards may do
3558 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3559 one (it's much easier than Tom thought :-).
3561 (5) Because of (4), we can now allow for the use of $prvscheck_result
3562 inside the third argument.
3564 PH/03 For some reason, the default setting of PATH when running a command from
3565 a pipe transport was just "/usr/bin". I have changed it to
3568 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3569 anything to be listed in the output from -bV.
3571 PH/05 When a filter generated an autoreply, the entire To: header line was
3572 quoted in the delivery log line, like this:
3574 => >A.N.Other <ano@some.domain> <original@ddress> ...
3576 This has been changed so that it extracts the operative address. There
3577 may be more than one such address. If so, they are comma-separated, like
3580 => >ano@some.domain,ona@other.domain <original@ddress> ...
3582 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3583 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3584 not being looked up in the rDNS to get a host name, Exim was showing the
3585 IP address twice in Received: lines, even though the IP addresses were
3586 identical. For example:
3588 Received: from [1.2.3.4] (helo=[1.2.3.4])
3590 However, if the real host name was known, it was omitting the HELO data
3591 if it matched the actual IP address. This has been tidied up so that it
3592 doesn't show the same IP address twice.
3594 PH/07 When both +timestamp and +memory debugging was on, the value given by
3595 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3596 called by the debug printing, thereby overwriting the timestamp buffer.
3597 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3599 PH/08 When the original message was included in an autoreply transport, it
3600 always said "this is a copy of the message, including all the headers",
3601 even if body_only or headers_only was set. It now gives an appropriate
3604 PH/09 Applied a patch from the Sieve maintainer which:
3606 o fixes some comments
3607 o adds the (disabled) notify extension core
3608 o adds some debug output for the result of if/elsif tests
3609 o points to the current vacation draft in the documentation
3610 and documents the missing references header update
3614 o fixes a bug in processing the envelope test (when testing
3615 multiple envelope elements, the last element determined the
3618 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3619 Electronic Mail") by including:
3621 Auto-submitted: auto-generated
3623 in the messages that it generates (bounce messages and others, such as
3624 warnings). In the case of bounce messages for non-SMTP messages, there was
3625 also a typo: it was using "Auto_submitted" (underscore instead of
3626 hyphen). Since every message generated by Exim is necessarily in response
3627 to another message, thes have all been changed to:
3629 Auto-Submitted: auto-replied
3631 in accordance with these statements in the RFC:
3633 The auto-replied keyword:
3635 - SHOULD be used on messages sent in direct response to another
3636 message by an automatic process,
3638 - MUST NOT be used on manually-generated messages,
3640 - MAY be used on Delivery Status Notifications (DSNs) and Message
3641 Disposition Notifications (MDNs),
3643 - MUST NOT be used on messages generated by automatic or periodic
3644 processes, except for messages which are automatic responses to
3647 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3648 to the default Received: header definition.
3650 PH/12 Added log selector acl_warn_skipped (default on).
3652 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3653 variables because (a) they are in the wrong storage pool and (b) even if
3654 they were copied, it wouldn't work properly because of the caching.
3656 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3657 checking when decoding. Apparently there are clients that generate
3658 overlong encoded strings. Why am I not surprised?
3660 PH/15 If the first argument of "${if match_address" was not empty, but did not
3661 contain an "@" character, Exim crashed. Now it writes a panic log message
3662 and treats the condition as false.
3664 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3666 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3667 extension "envelope-auth". The code is finished and in agreement with
3668 other implementations, but there is no documentation so far and in fact,
3669 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3670 not changing the active code.
3672 Print executed "if" and "elsif" statements when debugging is used. This
3673 helps a great deal to understand what a filter does.
3675 Document more things not specified clearly in RFC3028. I had all this
3676 sorted out, when out of a sudden new issues came to my mind. Oops."
3678 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3681 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3682 canonical form (as documented). However, after a host name lookup from
3683 the IP address, check_host() was doing a simple string comparison with
3684 addresses acquired from the DNS when checking that the found name did
3685 have the original IP as one of its addresses. Since any found IPv6
3686 addresses are likely to be in abbreviated form, the comparison could
3687 fail. Luckily, there already exists a function for doing the comparison
3688 by converting both addresses to binary, so now that is used instead of
3689 the text comparison.
3691 PH/20 There was another similar case to PH/19, when a complete host name was
3692 given in a host list; looking up its IP address could give an abbreviated
3693 form, whereas the current host's name might or might not be abbreviated.
3694 The same fix has been applied.
3700 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3701 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3704 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3705 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3707 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3709 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3710 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3711 Also, for "Auto-Submitted", treat anything other than "no" as
3712 non-personal, in accordance with RFC 3834. (Previously it treated
3713 anything starting "auto-" as non-personal.)
3715 TF/01 The control=submission/name=... option had a problem with syntax
3716 errors if the name included a slash character. The /name= option
3717 now slurps the rest of the string, so it can include any characters
3718 but it must come last in the list of options (after /sender_retain
3721 PH/05 Some modifications to the interface to the fake nameserver for the new
3729 TK/01 Added the "success_on_redirect" address verification option. See
3730 NewStuff for rationale and an example.
3732 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3734 PH/02 Patch to exigrep to allow it to work on syslog lines.
3736 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3737 fread() to skip over the body file's header line, because in Cygwin the
3738 header line is locked and is inaccessible.
3740 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3741 co-exist for some time) to make it clear that it is the Exim ID that is
3742 referenced, not the Message-ID: header line.
3744 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3745 string_format() function, because snprintf() does not exist on all
3748 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3749 ${stat: expansion item.
3751 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3752 protocol synchronization error", to keep the pedants happy.
3754 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3755 well as for IRIX systems, when gcc is being used. See the host.c source
3758 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3760 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3763 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3764 transport and to the smtp transport in LMTP mode.
3766 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3768 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3769 run a filter in a subprocess. This could lead to confusion in subsequent
3770 lookups in the parent process. There should also be a search_tidyup() at
3771 the end of the subprocess.
3773 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3774 only if the host matched helo_try_verify_hosts, which caused the
3775 verification to occur when the EHLO/HELO command was issued. The ACL just
3776 tested the remembered result. Now, if a previous verification attempt has
3777 not happened, "verify = helo" does it there and then.
3779 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3781 TK/03 Fix log output including CR from clamd.
3783 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3784 exist provoked a memory error which could cause a segfault.
3786 PH/15 Installed PCRE 6.2
3788 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3790 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3791 of the problem. Specifically, suggested +O2 rather than +O1 for the
3794 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3796 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3797 record, the port was not correctly logged when the outgoing_port log
3798 selector was set (it logged the transort's default port).
3800 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3801 fallback_hosts, and "hosts" in the smtp transport.
3803 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3804 host errors such as "Connection refused".
3806 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3807 authentication with radiusclient 0.4.9:
3809 - Error returned from rc_read_config was caught wrongly
3810 - Username/password not passed on to radius server due to wrong length.
3812 The presumption is that some radiusclient API changes for 4.51/PH/17
3813 were not taken care of correctly. The code is still untested by me (my
3814 Linux distribution still has 0.3.2 of radiusclient), but it was
3815 contributed by a Radius user.
3817 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3818 expanding the "port" option of the smtp transport.
3820 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3821 while reading a MIME header. Thanks to Tom Hughes for a patch.
3823 PH/24 Include config.h inside local_scan.h so that configuration settings are
3826 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3827 that in an ACL for RCPT (for example), you can examine exactly what was
3830 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3831 commands, but it was not correctly comparing the address with the actual
3832 client host address. Thus, it would show the EHLO address in Received:
3833 header lines when this was not necessary.
3835 PH/27 Added the % operator to ${eval:}.
3837 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3838 it should be ignoring failures (because with -C, for example, it has lost
3839 privilege). It wasn't ignoring creation failures other than "already
3842 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3845 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3846 because an input line was too long, either on its own, or by virtue of
3847 too many continuations, the temporary file was not being removed, and the
3848 return code was incorrect.
3850 PH/31 Missing "BOOL" in function definition in filtertest.c.
3852 PH/32 Applied Sieve patches from the maintainer.
3854 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3856 PH/33 Added "verify = not_blind".
3858 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3859 Local/Makefile (with some defaults set). These are used in built scripts
3860 such as exicyclog, but they have never been used in the exim_install
3861 script (though there are many overriding facilities there). I have
3862 arranged that the exim_install script now takes note of these two
3865 PH/35 Installed configuration files for Dragonfly.
3867 PH/36 When a locally submitted message by a trusted user did not contain a
3868 From: header, and the sender address was obtained from -f or from an SMTP
3869 MAIL command, and the trusted user did not use -F to supply a sender
3870 name, $originator_name was incorrectly used when constructing a From:
3871 header. Furthermore, $originator_name was used for submission mode
3872 messages from external hosts without From: headers in a similar way,
3873 which is clearly wrong.
3875 PH/37 Added control=suppress_local_fixups.
3877 PH/38 When log_selector = +received_sender was set, and the addition of the
3878 sender made the log line's construction buffer exactly full, or one byte
3879 less than full, an overflow happened when the terminating "\n" was
3882 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3883 when the result of a list match is failure because a DNS lookup failed.
3885 PH/40 RM_COMMAND is now used in the building process.
3887 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3888 the "build-* directories that it finds.
3890 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3891 address in a domain literal was a prefix of an interface address.
3893 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3894 when verifying a sender address, unless rewrite_headers is false.
3896 PH/44 Wrote a long comment about why errors_to addresses are verified as
3897 recipients, not senders.
3899 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3900 the ratelimit ACL was added.
3902 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3904 PH/46 Added extra information about PostgreSQL errors to the error string.
3906 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3907 avoiding the need to install special zones in a real server. This is
3908 backwards compatible; if it can't find the fake resolver, it drops back.
3909 Thus, both old and new test suites can be run.
3911 TF/02 Added util/ratelimit.pl
3913 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3914 clock is set back in time.
3916 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3917 Candler <B.Candler@pobox.com>.
3919 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3920 OK for addresses that are the result of redirections.
3922 PH/48 A number of further additions for the benefit of the new test suite,
3923 including a fake gethostbyname() that interfaces to the fake DNS resolver
3926 TF/06 The fix for widen_domains has also been applied to qualify_single and
3927 search_parents which are the other dnslookup options that can cause
3930 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3933 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3935 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3936 service" if it encountered an MX record with an empty target hostname.
3937 The message is now "an MX or SRV record indicated no SMTP service".
3939 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3940 if the DNS of the sending site is misconfigured. This is quite a
3941 common situation. This change restores the behaviour of treating a
3942 helo verification defer as a failure.
3944 PH/51 If self=fail was set on a router, the bounce message did not include the
3945 actual error message.
3951 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3953 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3954 command itself ended in error, the underlying message about the transport
3955 filter timeout was being overwritten with the pipe command error. Now the
3956 underlying error message should be appended to the second error message.
3958 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3960 PH/02 Reduce the amount of output that "make" produces by default. Full output
3961 can still be requested.
3963 PH/03 The warning log line about a condition test deferring for a "warn" verb
3964 was being output only once per connection, rather than after each
3965 occurrence (because it was using the same function as for successful
3966 "warn" verbs). This seems wrong, so I have changed it.
3968 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3969 it should not have, which might have caused a crash in the right
3970 circumstances, but probably never did.
3972 PH/04 Installed a modified version of Tony Finch's patch to make submission
3973 mode fix the return path as well as the Sender: header line, and to
3974 add a /name= option so that you can make the user's friendly name appear
3977 TF/03 Added the control = fakedefer ACL modifier.
3979 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3980 Mark Lowes for thorough testing.
3982 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3984 TK/03 Merged latest SRS patch from Miles Wilton.
3986 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3987 with the definition in sysexits.h (which is #included earlier).
3988 Fortunately, Exim does not actually use EX_OK. The code used to try to
3989 preserve the sysexits.h value, by assuming that macro definitions were
3990 scanned for macro replacements. I have been disabused of this notion,
3991 so now the code just undefines EX_OK before #including unistd.h.
3993 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3994 in the smtp transport. When a block could not be written in a single
3995 write() function, the timeout was being re-applied to each part-write.
3996 This seems wrong - if the receiver was accepting one byte at a time it
3997 would take for ever. The timeout is now adjusted when this happens. It
3998 doesn't have to be particularly precise.
4000 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4001 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4003 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4004 without a domain if the check to <postmaster@domain> fails.
4006 SC/01 Eximstats: added -xls and the ability to specify output files
4007 (patch written by Frank Heydlauf).
4009 SC/02 Eximstats: use FileHandles for outputting results.
4011 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4013 SC/04 Eximstats: fixed display of large numbers with -nvr option
4015 SC/05 Eximstats: fixed merging of reports with empty tables.
4017 SC/06 Eximstats: added the -include_original_destination flag
4019 SC/07 Eximstats: removed tabs and trailing whitespace.
4021 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4023 TK/06 MBOX spool code: Add real "From " MBOX separator line
4024 so the .eml file is really in mbox format (even though
4025 most programs do not really care). Patch from Alex Miller.
4027 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4028 The latter is generated from $received_to and is only set if the
4029 message has one envelope recipient. SA can use these headers,
4030 obviously out-of-the-box. Patch from Alex Miller.
4032 PH/08 The ${def test on a variable was returning false if the variable's
4033 value was "0", contrary to what the specification has always said!
4034 The result should be true unless the variable is empty.
4036 PH/09 The syntax error of a character other than { following "${if
4037 def:variable_name" (after optional whitespace) was not being diagnosed.
4038 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4039 accidental colon was present, for example, could give incorrect results.
4041 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4042 result is used (not including appendfile, where other changes are about
4045 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4046 This involved changing a lot of size variables from int to off_t. It
4047 should work with maildirs and everything.
4049 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4050 spamd dying while we are connected to it.
4052 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4055 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4056 function for BDB 4.3.
4058 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4060 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4061 the sources for PCRE have been re-arranged and more files are now
4064 PH/15 The code I had for printing potentially long long variables in PH/11
4065 above was not the best (it lost precision). The length of off_t variables
4066 is now inspected at build time, and an appropriate printing format (%ld
4067 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4068 to be "long long int" or "long int". This is needed for the internal
4069 formatting function string_vformat().
4071 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4072 the configuration file to be ":syslog", then the script "guesses" where
4073 the logs files are, rather than using the compiled in default. In our
4074 case the guess is not the same as the compiled default, so the script
4075 suddenly stopped working when I started to use syslog. The patch checks
4076 to see if log_file_path is "". If so, it attempts to read it from exim
4077 with no configuration file to get the compiled in version, before it
4078 falls back to the previous guessing code."
4080 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4081 implementing BATV in an Exim configuration. See NewStuff for the gory
4084 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4085 Makefile that are specific to HP-UX.
4087 PH/18 If the "use_postmaster" option was set for a recipient callout together
4088 with the "random" option, the postmaster address was used as the MAIL
4089 FROM address for the random test, but not for the subsequent recipient
4090 test. It is now used for both.
4092 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4093 patch removes a few documentation additions to RFC 3028, because the
4094 latest draft now contains them. It adds the new en;ascii-case comparator
4095 and a new error check for 8bit text in MIME parts. Comparator and
4096 require names are now matched exactly. I enabled the subaddress
4097 extension, but it is not well tested yet (read: it works for me)."
4099 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4100 rework some of the code of TK/09 above to avoid the hardwired use of
4101 "%lld" and "long long". Replaced the call to snprintf() with a call to
4104 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4105 records point to non-existent hosts", "retry timeout exceeded", and
4106 "retry time not reached for any host after a long failure period".
4108 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4109 experimental DomainKeys support:
4111 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4112 (2) On an error such as an illegally used "control", the wrong name for
4113 the control was given.
4115 These problems did NOT occur unless DomainKeys support was compiled.
4117 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4119 PH/24 Added ${if match_ip condition.
4121 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4122 databases so that it will be absolutely obvious if a crash occurs in the
4123 DB library. This is a regular occurrence (often caused by mis-matched
4126 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4127 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4128 picked up on a user's system that detects such things. There doesn't seem
4129 to be a gcc warning option for this - only an attribute that has to be
4130 put on the function's prototype. It seems that in Fedora Core 4 they have
4131 set this on a number of new functions. No doubt there will be more in due
4134 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4135 specify a transport. However, if an address that was verified by such a
4136 router was the subject of a callout, Exim crashed because it tried to
4137 read the rcpt_include_affixes from the non-existent transport. Now it
4138 just assumes that the setting of that option is false. This bug was
4139 introduced by 4.51/PH/31.
4141 PH/28 Changed -d+all to exclude +memory, because that information is very
4142 rarely of interest, but it makes the output a lot bigger. People tend to
4143 do -d+all out of habit.
4145 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4146 code in os-type was giving problems when libc.so lives in lib64, like on
4149 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4150 aren't the modern standard, and it seems that some systems' include files
4151 don't always have them. Exim was already checking for some of the newer
4152 ones like T_AAAA, and defining it itself. I've added checks for all the
4153 record types that Exim uses.
4155 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4156 not automatically generating a new one, as it is supposed to. This
4157 prevented TLS from working. If the file did exist, but contained invalid
4158 data, a new version was generated, as expected. It was only the case of a
4159 non-existent file that was broken.
4161 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4162 with a change in libdomainkeys > 0.64.
4164 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4165 from DNS. If the selector record carries the flag, it now has
4166 precedence over the domain-wide flag.
4168 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4170 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4171 the use of a port name that isn't defined in /etc/services) occurred, the
4172 message was deferred as in a normal delivery, and thus remained on the
4173 spool, instead of being failed because of the mua_wrapper setting. This
4174 is now fixed, and I tidied up some of the mua_wrapper messages at the
4177 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4178 the messages in a hash of arrays rather than using individual hashes.
4179 This is a bit cleaner and results in dramatic memory savings, albeit
4180 at a slight CPU cost.
4182 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4183 as requested by Marc Sherman.
4185 SC/10 Eximstats: added histograms for user specified patterns as requested
4188 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4190 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4191 fopen() in the content-scanning modules that did not already have it.
4197 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4198 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4200 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4202 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4204 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4205 to test Sieve filters that use "vacation".
4207 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4208 that changes the way the GnuTLS parameters are stored in the cache file.
4209 The new format can be generated externally. For backward compatibility,
4210 if the data in the cache doesn't make sense, Exim assumes it has read an
4211 old-format file, and it generates new data and writes a new file. This
4212 means that you can't go back to an older release without removing the
4215 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4216 work if there are any delivery delays because "one_time" forces the
4217 parent to be marked "delivered", so its unseen clone is never tried
4218 again. For this reason, Exim now forbids the simultaneous setting of
4221 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4222 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4223 entry if you want to know the details. The fix, however, neglected to
4224 consider the case when local delivery batching is involved. The test for
4225 "previously delivered" was not happening when checking to see if an
4226 address could be batched with a previous (undelivered) one; under
4227 certain circumstances this could lead to multiple deliveries to the same
4230 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4231 in its include files, and this causes problems building Exim.
4233 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4234 header_syntax) but Exim was just ignoring anything given after a slash.
4235 In particular, this caused confusion with an attempt to use "verify =
4236 reverse_host_lookup/defer_ok". An error is now given when options are
4237 supplied for verify items that do not have them. (Maybe reverse_host_
4238 lookup should have a defer_ok option, but that's a different point.)
4240 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4241 defined by RFC 821) to 2048, because there were problems with some AUTH
4242 commands, and RFC 1869 says the size should be increased for extended
4243 SMTP commands that take arguments.
4245 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4248 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4249 "unknown" error; now it says that the functionality isn't in the binary.
4251 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4252 an address' error message when a string expansion fails (syntax or
4253 whatever). Otherwise the password may appear in the log. Following change
4254 PH/42 below, there is no longer a chance of it appearing in a bounce
4257 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4259 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4260 of its addresses was ever tried. (Bugzilla bug #2.)
4262 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4263 the result incorrectly in the debug output. (It correctly added a newline
4264 to what was transported.)
4266 TF/01 Added $received_time.
4268 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4269 commented out examples of how to interface to a virus scanner and to
4270 SpamAssassin. Also added commented examples of av_scanner and
4271 spamd_address settings.
4273 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4274 and controls are allowed in which ACLs. There were a couple of minor
4275 errors. Some of the entries in the conditions table (which is a table of
4276 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4277 negation of where the condition IS allowed.
4279 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4281 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4282 header file does not have a version number, so I've had to invent a new
4283 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4284 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4285 radiusclient), but it was contributed by a Radius user.
4287 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4288 files or directories, for OpenSSL.
4290 PH/19 When an Exim process that is running as root has to create an Exim log
4291 file, it does so in a subprocess that runs as exim:exim so as to get the
4292 ownership right at creation (otherwise, other Exim processes might see
4293 the file with the wrong ownership). There was no test for failure of this
4294 fork() call, which would lead to the process getting stuck as it waited
4295 for a non-existent subprocess. Forks do occasionally fail when resources
4296 run out. I reviewed all the other calls to fork(); they all seem to check
4299 PH/20 When checking for unexpected SMTP input at connect time (before writing
4300 the banner), Exim was not dealing correctly with a non-positive return
4301 from the read() function. If the client had disconnected by this time,
4302 the result was a log entry for a synchronization error with an empty
4303 string after "input=" when read() returned zero. If read() returned -1
4304 (an event I could not check), uninitialized data bytes were printed.
4305 There were reports of junk text (parts of files, etc) appearing after
4308 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4310 PH/22 Added support for macro redefinition, and (re)definition in between
4311 driver and ACL definitions.
4313 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4314 forgetting to use the resulting value; it was using the unexpanded value.
4316 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4317 hadn't been configured. The fix is from Juergen Kreileder, who
4318 understands it better than I do:
4320 "Here's what I see happening with three configured cyrus_sasl
4321 authenticators configured (plain, login, cram-md5):
4323 On startup auth_cyrus_sasl_init() gets called for each of these.
4324 This means three calls to sasl_listmech() without a specified mech_list.
4325 => SASL tests which mechs of all available mechs actually work
4326 => three warnings about OTP not working
4327 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4329 With the patch, sasl_listmech() also gets called three times. But now
4330 SASL's mech_list option is set to the server_mech specified in the the
4331 authenticator. Or in other words, the answer from sasl_listmech()
4332 gets limited to just the mech you're testing for (which is different
4334 => the return list contains just 'plain' or 'login', 'cram-md5' or
4335 nothing depending on the value of ob->server_mech.
4337 I've just tested the patch: Authentication still works fine,
4338 unavailable mechs specified in the exim configuration are still
4339 caught, and the auth.log warnings about OTP are gone."
4341 PH/25 When debugging is enabled, the contents of the command line are added
4342 to the debugging output, even when log_selector=+arguments is not
4345 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4346 answer is "GNU", and only if the return is "GNU/something" is the answer
4349 PH/27 $acl_verify_message is now set immediately after the failure of a
4350 verification in an ACL, and so is available in subsequent modifiers. In
4351 particular, the message can be preserved by coding like this:
4353 warn !verify = sender
4354 set acl_m0 = $acl_verify_message
4356 Previously, $acl_verify_message was set only while expanding "message"
4357 and "log_message" when a very denied access.
4359 PH/28 Modified OS/os.c-Linux with
4361 -#ifndef OS_LOAD_AVERAGE
4362 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4364 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4365 nomenclature these days.)
4367 PH/29 Installed patch from the Sieve maintainer that adds the options
4368 sieve_useraddress and sieve_subaddress to the redirect router.
4370 PH/30 In these circumstances:
4371 . Two addresses routed to the same list of hosts;
4372 . First host does not offer TLS;
4373 . First host accepts first address;
4374 . First host gives temporary error to second address;
4375 . Second host offers TLS and a TLS session is established;
4376 . Second host accepts second address.
4377 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4378 and peerdn, if requested) that were in fact used only for the second
4381 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4382 attention to any local part prefix or suffix that was matched by the
4383 router that accepted the address. It now behaves in the same way as it
4384 does for delivery: the affixes are removed from the local part unless
4385 rcpt_include_affixes is set on the transport.
4387 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4388 timeout during the DATA phase of an incoming message.
4390 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4391 applied a patch sanctioned by the Sieve maintainer.
4393 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4394 the uid or gid is negative. A case of a negative gid caused this to be
4395 noticed. The fix allows for either to be negative.
4397 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4398 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4401 PH/36 The change PH/12 above was broken. Fixed it.
4403 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4404 the grounds that routing the same address twice would always produce the
4405 same answer. This might have been true once, but it is certainly no
4406 longer true now. Routing a child address may depend on the previous
4407 routing that produced that child. Some complicated redirection strategies
4408 went wrong when messages had multiple recipients, and made Exim's
4409 behaviour dependent on the order in which the addresses were given.
4411 I have moved the duplicate checking until after the routing is complete.
4412 Exim scans the addresses that are assigned to local and remote
4413 transports, and removes any duplicates. This means that more work will be
4414 done, as duplicates will always all be routed, but duplicates are
4415 presumably rare, so I don't expect this is of any significance.
4417 For deliveries to pipes, files, and autoreplies, the duplicate checking
4418 still happens during the routing process, since they are not going to be
4421 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4422 It corrects a timeout issue with spamd. This is Ian's comment: "The
4423 background is that sometimes spamd either never reads data from a
4424 connection it has accepted, or it never writes response data. The exiscan
4425 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4426 blindly assumes that writes won't block so it may never time out."
4428 PH/39 Allow G after quota size as well as K and M.
4430 PH/40 The value set for $authenticated_id in an authenticator may not contain
4431 binary zeroes or newlines because the value is written to log lines and
4432 to spool files. There was no check on this. Now the value is run through
4433 the string_printing() function so that such characters are converted to
4434 printable escape sequences.
4436 PH/41 $message_linecount is a new variable that contains the total number of
4437 lines in the message. Compare $body_linecount, which is the count for the
4440 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4441 bounce and delay warning messages, except in certain special cases, which
4444 (a) An SMTP error message from a remote host;
4445 (b) A message specified in a :fail: redirection;
4446 (c) A message specified in a "fail" command in a system filter;
4447 (d) A message specified in a FAIL return from the queryprogram router;
4448 (e) A message specified by the cannot_route_message router option.
4450 In these cases only, Exim does include the error details in bounce and
4451 warning messages. There are also a few cases where bland messages such
4452 as "unrouteable address" or "local delivery error" are given.
4454 PH/43 $value is now also set for the "else" part of a ${run expansion.
4456 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4457 being worked on, but at least Exim now implements the latest version to
4460 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4461 process to complete was treated as a delivery failure, a timeout while
4462 writing the message to the pipe was logged, but erroneously treated as a
4463 successful delivery. Such timeouts include transport filter timeouts. For
4464 consistency with the overall process timeout, these timeouts are now
4465 treated as errors, giving rise to delivery failures by default. However,
4466 there is now a new Boolean option for the pipe transport called
4467 timeout_defer, which, if set TRUE, converts the failures into defers for
4468 both kinds of timeout. A transport filter timeout is now identified in
4471 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4472 systems where "make" and "gmake" are different, calling "gmake" at top
4473 level broke things. I've arranged for the value of $(MAKE) to be passed
4474 from the Makefile to this script so that it can call the same version of
4478 A note about Exim versions 4.44 and 4.50
4479 ----------------------------------------
4481 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4482 changes of various kinds. As a consequence, a big documentation update was
4483 needed. This delayed the release for rather longer than seemed good, especially
4484 in the light of a couple of (minor) security issues. Therefore, the changes
4485 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4486 release. So 4.44 and 4.50 are in effect two different branches that both start
4489 I have left the 4.50 change log unchanged; it contains all the changes since
4490 4.43. The change log for 4.44 is below; many of its items are identical to
4491 those for 4.50. This seems to be the most sensible way to preserve the
4492 historical information.
4498 1. Minor wording change to the doc/README.SIEVE file.
4500 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4501 computation of the current number of files was incorrect.
4503 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4504 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4507 4. Give more explanation in the error message when the command for a transport
4508 filter fails to execute.
4510 5. There are several places where Exim runs a non-Exim command in a
4511 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4512 was being done only for the command run by the queryprogram router. It is
4513 now done for all such subprocesses. The other cases are: ${run, transport
4514 filters, and the commands run by the lmtp and pipe transports.
4516 6. Added CONFIGURE_GROUP build-time option.
4518 7. Some older OS have a limit of 256 on the maximum number of file
4519 descriptors. Exim was using setrlimit() to set 1000 as a large value
4520 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4521 systems. I've change it so that if it can't get 1000, it tries for 256.
4523 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4524 was an oversight, and furthermore, ever since the addition of extra
4525 controls (e.g. 4.43/32), the checks on when to allow different forms of
4526 "control" were broken. There should now be diagnostics for all cases when a
4527 control that does not make sense is encountered.
4529 9. Added the /retain_sender option to "control=submission".
4531 10. $recipients is now available in the predata ACL (oversight).
4533 11. Tidy the search cache before the fork to do a delivery from a message
4534 received from the command line. Otherwise the child will trigger a lookup
4535 failure and thereby defer the delivery if it tries to use (for example) a
4536 cached ldap connection that the parent has called unbind on.
4538 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4539 of $address_data from the recipient verification was clobbered by the
4540 sender verification.
4542 13. The value of address_data from a sender verification is now available in
4543 $sender_address_data in subsequent conditions in the ACL statement.
4545 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4547 15. Added a new option "connect=<time>" to callout options, to set a different
4550 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4551 was its contents. (It was OK if the option was not defined at all.)
4553 17. A "Completed" log line is now written for messages that are removed from
4554 the spool by the -Mrm option.
4556 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4557 information about exactly what failed.
4559 19. Added -dd to debug only the daemon process.
4561 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4562 handles timeouts, both on the server side and network timeouts. Renamed the
4563 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4565 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4566 It is now set to "smtps".
4568 22. $host_address is now set to the target address during the checking of
4569 ignore_target_hosts.
4571 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4572 being passed; this would have caused $sender_host_name to have been used if
4573 matching the list had actually called for a host name (not very likely,
4574 since this list is usually IP addresses). A host name is now passed as
4577 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4578 code that specifies a non-restarting handler (typically sigaction() in
4579 modern systems) in an attempt to fix a rare and obscure crash bug.
4581 25. Narrowed the window for a race in the daemon that could cause it to ignore
4582 SIGCHLD signals. This is not a major problem, because they are used only to
4583 wake it up if nothing else does.
4585 26. A malformed maildirsize file could cause Exim to calculate negative values
4586 for the mailbox size or file count. Odd effects could occur as a result.
4587 The maildirsize information is now recalculated if the size or filecount
4590 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4591 support for a long time. Removed HAVE_SYS_VFS_H.
4593 28. Installed the latest version of exipick from John Jetmore.
4595 29. In an address list, if the pattern was not a regular expression, an empty
4596 subject address (from a bounce message) matched only if the pattern was an
4597 empty string. Non-empty patterns were not even tested. This was the wrong
4598 because it is perfectly reasonable to use an empty address as part of a
4599 database query. An empty address is now tested by patterns that are
4600 lookups. However, all the other forms of pattern expect the subject to
4601 contain a local part and a domain, and therefore, for them, an empty
4602 address still always fails if the pattern is not itself empty.
4604 30. Exim went into a mad DNS loop when attempting to do a callout where the
4605 host was specified on an smtp transport, and looking it up yielded more
4606 than one IP address.
4608 31. Re-factored the code for checking spool and log partition space into a
4609 function that finds that data and another that does the check. The former
4610 is then used to implement four new variables: $spool_space, $log_space,
4611 $spool_inodes, and $log_inodes.
4613 32. The RFC2047 encoding function was originally intended for short strings
4614 such as real names; it was not keeping to the 75-character limit for
4615 encoded words that the RFC imposes. It now respects the limit, and
4616 generates multiple encoded words if necessary. To be on the safe side, I
4617 have increased the buffer size for the ${rfc2047: expansion operator from
4620 33. It is now permitted to omit both strings after an "if" condition; if the
4621 condition is true, the result is "true". As before, when the second string
4622 is omitted, a false condition yields an empty string. This makes it less
4623 cumbersome to write custom ACL and router conditions.
4625 34. Failure to deliver a bounce message always caused it to be frozen, even if
4626 there was an errors_to setting on the router. The errors_to setting is now
4629 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4630 canonical form (fully expanded) before being placed in
4631 $sender_host_address.
4633 36. The table in the code that translates DNS record types into text (T_A to
4634 "A" for instance) was missing entries for NS and CNAME. It is just possible
4635 that this could have caused confusion if both these types were looked up
4636 for the same domain, because the text type is used as part of Exim's
4637 per-process caching. But the chance of anyone hitting this buglet seems
4640 37. The dnsdb lookup has been extended in a number of ways.
4642 (1) There is a new type, "zns", which walks up the domain tree until it
4643 finds some nameserver records. It should be used with care.
4645 (2) There is a new type, "mxh", which is like "mx" except that it returns
4646 just the host names, not the priorities.
4648 (3) It is now possible to give a list of domains (or IP addresses) to be
4649 looked up. The behaviour when one of the lookups defers can be
4650 controlled by a keyword.
4652 (4) It is now possible to specify the separator character for use when
4653 multiple records are returned.
4655 38. The dnslists ACL condition has been extended: it is now possible to supply
4656 a list of IP addresses and/or domains to be looked up in a particular DNS
4659 39. Added log_selector=+queue_time_overall.
4661 40. When running the queue in the test harness, wait just a tad after forking a
4662 delivery process, to get repeatability of debugging output.
4664 41. Include certificate and key file names in error message when GnuTLS fails
4665 to set them up, because the GnuTLS error message doesn't include the name
4666 of the failing file when there is a problem reading it.
4668 42. Allow both -bf and -bF in the same test run.
4670 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4672 44. The "Exiscan patch" is now merged into the mainline Exim source.
4674 45. Sometimes the final signoff response after QUIT could fail to get
4675 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4676 < 0 before doing a fflush(). This bug looks as though it goes back to the
4677 introduction of TLS in release 3.20, but "sometimes" must have been rare
4678 because the tests only now provoked it.
4680 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4681 (this can affect the format of dates).
4683 47. exim_tidydb, when checking for the continued existence of a message for
4684 which it has found a message-specific retry record, was not finding
4685 messages that were in split spool directories. Consequently, it was
4686 deleting retry records that should have stayed in existence.
4688 48. Steve fixed some bugs in eximstats.
4690 49. The SPA authentication driver was not abandoning authentication and moving
4691 on to the next authenticator when an expansion was forced to fail,
4692 contradicting the general specification for all authenticators. Instead it
4693 was generating a temporary error. It now behaves as specified.
4695 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4696 (the order specifies the preference for clients). The order is now AES256,
4697 AES128, 3DES, ARCFOUR128.
4699 51. Small patch to Sieve code - explicitly set From: when generating an
4702 52. Exim crashed if a remote delivery caused a very long error message to be
4703 recorded - for instance if somebody sent an entire SpamAssassin report back
4704 as a large number of 550 error lines. This bug was coincidentally fixed by
4705 increasing the size of one of Exim's internal buffers (big_buffer) that
4706 happened as part of the Exiscan merge. However, to be on the safe side, I
4707 have made the code more robust (and fixed the comments that describe what
4710 53. Now that there can be additional text after "Completed" in log lines (if
4711 the queue_time_overall log selector is set), a one-byte patch to exigrep
4712 was needed to allow it to recognize "Completed" as not the last thing in
4715 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4716 patch that reportedly fixes this has been added. I am not expert enough to
4717 create a test for it. This is what the patch creator wrote:
4719 "I found a little strange behaviour of ldap code when working with
4720 Windows 2003 AD Domain, where users was placed in more than one
4721 Organization Units. When I tried to give exim partial DN, the exit code
4722 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4723 But simultaneously result of request was absolutely normal ldap result,
4724 so I produce this patch..."
4726 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4727 REFERENCE, so I have modified the code to exclude the patch when that macro
4730 55. Some experimental protocols are using DNS PTR records for new purposes. The
4731 keys for these records are domain names, not reversed IP addresses. The
4732 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4733 leaves it alone. Component reversal etc. now happens only for IP addresses.
4736 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4738 57. Double the size of the debug message buffer (to 2048) so that more of very
4739 long debug lines gets shown.
4741 58. The exicyclog utility now does better if the number of log files to keep
4742 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4744 59. Two changes related to the smtp_active_hostname option:
4746 (1) $smtp_active_hostname is now available as a variable.
4747 (2) The default for smtp_banner uses $smtp_active_hostname instead
4748 of $primary_hostname.
4750 60. The host_aton() function is supposed to be passed a string that is known
4751 to be a valid IP address. However, in the case of IPv6 addresses, it was
4752 not checking this. This is a hostage to fortune. Exim now panics and dies
4753 if the condition is not met. A case was found where this could be provoked
4754 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4755 components; fortuitously, this particular loophole had already been fixed
4756 by change 4.50/55 above.
4758 If there are any other similar loopholes, the new check in host_aton()
4759 itself should stop them being exploited. The report I received stated that
4760 data on the command line could provoke the exploit when Exim was running as
4761 exim, but did not say which command line option was involved. All I could
4762 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4763 running as the user.
4766 61. There was a buffer overflow vulnerability in the SPA authentication code
4767 (which came originally from the Samba project). I have added a test to the
4768 spa_base64_to_bits() function which I hope fixes it.
4771 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4772 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4774 63. The daemon start-up calls getloadavg() while still root for those OS that
4775 need the first call to be done as root, but it missed one case: when
4776 deliver_queue_load_max is set with deliver_drop_privilege. This is
4777 necessary for the benefit of the queue runner, because there is no re-exec
4778 when deliver_drop_privilege is set.
4780 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4781 This has been fixed.
4783 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4784 was in use, was not putting the data itself into the right store pool;
4785 consequently, it could be overwritten for a subsequent message in the same
4786 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4789 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4791 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4792 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4793 the function were treating the return as a boolean value, which happened to
4794 work because 0=false and not-0=true, but is not correct code.
4796 68. The host_aton() function was not handling scoped IPv6 addresses (those
4797 with, for example, "%eth0" on the end) correctly.
4799 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4800 negated items (that is, ~something) in unsigned ints. Some compilers
4801 apparently mutter when there is no cast.
4803 70. If an address verification called from an ACL failed, and did not produce a
4804 user-specific message (i.e. there was only a "system" message), nothing was
4805 put in $acl_verify_message. In this situation, it now puts the system
4808 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4809 session; change 4.31/43 added the unwanted input to the log line - except
4810 that it did not do this in the start of session case. It now does.
4812 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4813 This is wrong and can cause the other end to generate a synchronization
4814 error if it is another Exim or anything else that does the synchronization
4815 check. A QUIT command is no longer sent after a timeout.
4817 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4818 during host lookups.
4820 74. The defer_ok option of callout verification was not working if it was used
4821 when verifying addresses in header lines, that is, for this case:
4823 verify = header_sender/callout=defer_ok
4825 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4826 those file descriptors could be used for SMTP connections. If anything
4827 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4828 could be sent to the SMTP client, causing chaos. The daemon now opens
4829 stdin, stdout, and stderr to /dev/null when it puts itself into the
4832 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4833 log by default. The user can override this with suitable Perl magic.
4835 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4836 the log message when discard triggers, was not working for the DATA ACL or
4837 for the non-SMTP ACL.
4839 78. Error message wording change in sieve.c.
4841 79. If smtp_accept_max_per_host was set, the number of connections could be
4842 restricted to fewer than expected, because the daemon was trying to set up
4843 a new connection before checking whether the processes handling previous
4844 connections had finished. The check for completed processes is now done
4845 earlier. On busy systems, this bug wouldn't be noticed because something
4846 else would have woken the daemon, and it would have reaped the completed
4849 80. If a message was submitted locally by a user whose login name contained one
4850 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4851 It caused a spool format error. I have fixed the spool reading code. A
4852 related problem was that the "from" clause in the Received: line became
4853 illegal because of the space(s). It is now covered by ${quote_local_part.
4855 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4858 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4860 83. Patch from Sieve maintainer to fix unterminated string problem in
4861 "vacation" handling.
4863 84. Some minor changes to the Linux configuration files to help with other
4864 OS variants using glibc.
4866 85. One more patch for Sieve to update vacation handling to latest spec.
4869 ----------------------------------------------------
4870 See the note above about the 4.44 and 4.50 releases.
4871 ----------------------------------------------------
4877 1. Change 4.43/35 introduced a bug that caused file counts to be
4878 incorrectly computed when quota_filecount was set in an appendfile
4881 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4882 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4885 3. Give more explanation in the error message when the command for a transport
4886 filter fails to execute.
4888 4. There are several places where Exim runs a non-Exim command in a
4889 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4890 was being done only for the command run by the queryprogram router. It is
4891 now done for all such subprocesses. The other cases are: ${run, transport
4892 filters, and the commands run by the lmtp and pipe transports.
4894 5. Some older OS have a limit of 256 on the maximum number of file
4895 descriptors. Exim was using setrlimit() to set 1000 as a large value
4896 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4897 systems. I've change it so that if it can't get 1000, it tries for 256.
4899 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4900 was an oversight, and furthermore, ever since the addition of extra
4901 controls (e.g. 4.43/32), the checks on when to allow different forms of
4902 "control" were broken. There should now be diagnostics for all cases when a
4903 control that does not make sense is encountered.
4905 7. $recipients is now available in the predata ACL (oversight).
4907 8. Tidy the search cache before the fork to do a delivery from a message
4908 received from the command line. Otherwise the child will trigger a lookup
4909 failure and thereby defer the delivery if it tries to use (for example) a
4910 cached ldap connection that the parent has called unbind on.
4912 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4913 of $address_data from the recipient verification was clobbered by the
4914 sender verification.
4916 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4917 was its contents. (It was OK if the option was not defined at all.)
4919 11. A "Completed" log line is now written for messages that are removed from
4920 the spool by the -Mrm option.
4922 12. $host_address is now set to the target address during the checking of
4923 ignore_target_hosts.
4925 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4926 being passed; this would have caused $sender_host_name to have been used if
4927 matching the list had actually called for a host name (not very likely,
4928 since this list is usually IP addresses). A host name is now passed as
4931 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4932 code that specifies a non-restarting handler (typically sigaction() in
4933 modern systems) in an attempt to fix a rare and obscure crash bug.
4935 15. Narrowed the window for a race in the daemon that could cause it to ignore
4936 SIGCHLD signals. This is not a major problem, because they are used only to
4937 wake it up if nothing else does.
4939 16. A malformed maildirsize file could cause Exim to calculate negative values
4940 for the mailbox size or file count. Odd effects could occur as a result.
4941 The maildirsize information is now recalculated if the size or filecount
4944 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4945 support for a long time. Removed HAVE_SYS_VFS_H.
4947 18. Updated exipick to current release from John Jetmore.
4949 19. Allow an empty sender to be matched against a lookup in an address list.
4950 Previously the only cases considered were a regular expression, or an
4953 20. Exim went into a mad DNS lookup loop when doing a callout where the
4954 host was specified on the transport, if the DNS lookup yielded more than
4957 21. The RFC2047 encoding function was originally intended for short strings
4958 such as real names; it was not keeping to the 75-character limit for
4959 encoded words that the RFC imposes. It now respects the limit, and
4960 generates multiple encoded words if necessary. To be on the safe side, I
4961 have increased the buffer size for the ${rfc2047: expansion operator from
4964 22. Failure to deliver a bounce message always caused it to be frozen, even if
4965 there was an errors_to setting on the router. The errors_to setting is now
4968 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4969 canonical form (fully expanded) before being placed in
4970 $sender_host_address.
4972 24. Updated eximstats to version 1.33
4974 25. Include certificate and key file names in error message when GnuTLS fails
4975 to set them up, because the GnuTLS error message doesn't include the name
4976 of the failing file when there is a problem reading it.
4978 26. Expand error message when OpenSSL has problems setting up cert/key files.
4981 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4982 (this can affect the format of dates).
4984 28. exim_tidydb, when checking for the continued existence of a message for
4985 which it has found a message-specific retry record, was not finding
4986 messages that were in split spool directories. Consequently, it was
4987 deleting retry records that should have stayed in existence.
4989 29. eximstats updated to version 1.35
4990 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4991 1.35 - bugfix such that pie charts by volume are generated correctly
4993 30. The SPA authentication driver was not abandoning authentication and moving
4994 on to the next authenticator when an expansion was forced to fail,
4995 contradicting the general specification for all authenticators. Instead it
4996 was generating a temporary error. It now behaves as specified.
4998 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4999 (the order specifies the preference for clients). The order is now AES256,
5000 AES128, 3DES, ARCFOUR128.
5002 31. Small patch to Sieve code - explicitly set From: when generating an
5005 32. Exim crashed if a remote delivery caused a very long error message to be
5006 recorded - for instance if somebody sent an entire SpamAssassin report back
5007 as a large number of 550 error lines. This bug was coincidentally fixed by
5008 increasing the size of one of Exim's internal buffers (big_buffer) that
5009 happened as part of the Exiscan merge. However, to be on the safe side, I
5010 have made the code more robust (and fixed the comments that describe what
5013 33. Some experimental protocols are using DNS PTR records for new purposes. The
5014 keys for these records are domain names, not reversed IP addresses. The
5015 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5016 leaves it alone. Component reversal etc. now happens only for IP addresses.
5019 34. The host_aton() function is supposed to be passed a string that is known
5020 to be a valid IP address. However, in the case of IPv6 addresses, it was
5021 not checking this. This is a hostage to fortune. Exim now panics and dies
5022 if the condition is not met. A case was found where this could be provoked
5023 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5024 components; fortuitously, this particular loophole had already been fixed
5025 by change 4.50/55 or 4.44/33 above.
5027 If there are any other similar loopholes, the new check in host_aton()
5028 itself should stop them being exploited. The report I received stated that
5029 data on the command line could provoke the exploit when Exim was running as
5030 exim, but did not say which command line option was involved. All I could
5031 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5032 running as the user.
5035 35. There was a buffer overflow vulnerability in the SPA authentication code
5036 (which came originally from the Samba project). I have added a test to the
5037 spa_base64_to_bits() function which I hope fixes it.
5040 36. The daemon start-up calls getloadavg() while still root for those OS that
5041 need the first call to be done as root, but it missed one case: when
5042 deliver_queue_load_max is set with deliver_drop_privilege. This is
5043 necessary for the benefit of the queue runner, because there is no re-exec
5044 when deliver_drop_privilege is set.
5046 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5047 was in use, was not putting the data itself into the right store pool;
5048 consequently, it could be overwritten for a subsequent message in the same
5049 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5052 38. Sometimes the final signoff response after QUIT could fail to get
5053 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5054 < 0 before doing a fflush(). This bug looks as though it goes back to the
5055 introduction of TLS in release 3.20, but "sometimes" must have been rare
5056 because the tests only now provoked it.
5062 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5063 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5064 Now it returns an integer. A number of calls were still expecting a T/F
5065 return. Fortuitously, in all cases, the tests worked in OK situations,
5066 which is the norm. However, things would have gone wrong on any write
5067 failures on the smtp file descriptor. This function is used when sending
5068 messages over SMTP and also when doing verify callouts.
5070 2. When Exim is called to do synchronous delivery of a locally submitted
5071 message (the -odf or -odi options), it no longer closes stderr before doing
5074 3. Implemented the mua_wrapper option.
5076 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5078 5. Implemented the functions header_remove(), header_testname(),
5079 header_add_at_position(), and receive_remove_recipient(), and exported them
5082 6. If an ACL "warn" statement specified the addition of headers, Exim already
5083 inserted X-ACL-Warn: at the start if there was no header name. However, it
5084 was not making this test for the second and subsequent header lines if
5085 there were newlines in the string. This meant that an invalid header could
5086 be inserted if Exim was badly configured.
5088 7. Allow an ACL "warn" statement to add header lines at the start or after all
5089 the Received: headers, as well as at the end.
5091 8. Added the rcpt_4xx retry error code.
5093 9. Added postmaster_mailfrom=xxx to callout verification option.
5095 10. Added mailfrom=xxxx to the callout verification option, for verify=
5098 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5099 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5101 12. Inserted some casts to stop certain compilers warning when using pointer
5102 differences as field lengths or precisions in printf-type calls (mostly
5103 affecting debugging statements).
5105 13. Added optional readline() support for -be (dynamically loaded).
5107 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5108 same clock tick as a message's arrival, so that its received time was the
5109 same as the "first fail" time on the retry record, and that message
5110 remained on the queue past the ultimate address timeout, every queue runner
5111 would try a delivery (because it was past the ultimate address timeout) but
5112 after another failure, the ultimate address timeout, which should have then
5113 bounced the address, did not kick in. This was a "< instead of <=" error;
5114 in most cases the first failure would have been in the next clock tick
5115 after the received time, and all would be well.
5117 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5118 being recognized when the domain list was tested by the match_domain
5119 condition in an expansion string.
5121 16. Added the ${str2b64: operator.
5123 17. Exim was always calling setrlimit() to set a large limit for the number of
5124 processes, without checking whether the existing limit was already
5125 adequate. (It did check for the limit on file descriptors.) Furthermore,
5126 errors from getrlimit() and setrlimit() were being ignored. Now they are
5127 logged to the main and panic logs, but Exim does carry on, to try to do its
5128 job under whatever limits there are.
5130 18. Imported PCRE 5.0.
5132 19. Trivial typo in log message " temporarily refused connection" (the leading
5135 20. If the log selector return_path_on_delivery was set and an address was
5136 redirected to /dev/null, the delivery process crashed because it assumed
5137 that a return path would always be set for a "successful" delivery. In this
5138 case, the whole delivery is bypassed as an optimization, and therefore no
5141 21. Internal re-arrangement: the function for sending a challenge and reading
5142 a response while authentication was assuming a zero-terminated challenge
5143 string. It's now changed to take a pointer and a length, to allow for
5144 binary data in such strings.
5146 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5148 23. Exim was not respecting finduser_retries when seeking the login of the
5149 uid under which it was called; it was always trying 10 times. (The default
5150 setting of finduser_retries is zero.) Also, it was sleeping after the final
5151 failure, which is pointless.
5153 24. Implemented tls_on_connect_ports.
5155 25. Implemented acl_smtp_predata.
5157 26. If the domain in control=submission is set empty, Exim assumes that the
5158 authenticated id is a complete email address when it generates From: or
5159 Sender: header lines.
5161 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5162 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5163 chown and chgrp in /bin and hostname in /usr/bin.
5165 28. Exim was keeping the "process log" file open after each use, just as it
5166 does for the main log. This opens the possibility of it remaining open for
5167 long periods when the USR1 signal hits a daemon. Occasional processlog
5168 errors were reported, that could have been caused by this. Anyway, it seems
5169 much more sensible not to leave this file open at all, so that is what now
5172 29. The long-running daemon process does not normally write to the log once it
5173 has entered its main loop, and it closes the log before doing so. This is
5174 so that log files can straightforwardly be renamed and moved. However,
5175 there are a couple of unusual error situations where the daemon does write
5176 log entries, and I had neglected to close the log afterwards.
5178 30. The text of an SMTP error response that was received during a remote
5179 delivery was being truncated at 512 bytes. This is too short for some of
5180 the long messages that one sometimes sees. I've increased the limit to
5183 31. It is now possible to make retry rules that apply only when a message has a
5184 specific sender, in particular, an empty sender.
5186 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5187 it possible to be selective about when SMTP synchronization is enforced.
5189 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5191 32. Implemented hosts_connection_nolog.
5193 33. Added an ACL for QUIT.
5195 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5198 35. Added mailbox_size and mailbox_filecount to appendfile.
5200 36. Added control = no_multiline_responses to ACLs.
5202 37. There was a bug in the logic of the code that waits for the clock to tick
5203 in the case where the clock went backwards by a substantial amount such
5204 that the microsecond fraction of "now" was more than the microsecond
5205 fraction of "then" (but the whole seconds number was less).
5207 38. Added support for the libradius Radius client library this is found on
5208 FreeBSD (previously only the radiusclient library was supported).
5214 1. When certain lookups returned multiple values in the form name=value, the
5215 quoting of the values was not always being done properly. Specifically:
5216 (a) If the value started with a double quote, but contained no whitespace,
5218 (b) If the value contained whitespace other than a space character (i.e.
5219 tabs or newlines or carriage returns) it was not quoted.
5220 This fix has been applied to the mysql and pgsql lookups by writing a
5221 separate quoting function and calling it from the lookup code. The fix
5222 should probably also be applied to nisplus, ibase and oracle lookups, but
5223 since I cannot test any of those, I have not disturbed their existing code.
5225 2. A hit in the callout cache for a specific address caused a log line with no
5226 reason for rejecting RCPT. Now it says "Previous (cached) callout
5227 verification failure".
5229 3. There was an off-by-one bug in the queryprogram router. An over-long
5230 return line was truncated at 256 instead of 255 characters, thereby
5231 overflowing its buffer with the terminating zero. As well as fixing this, I
5232 have increased the buffer size to 1024 (and made a note to document this).
5234 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5235 when Exim is waiting for an SMTP response from a remote server, Exim
5236 restarts its select() call on the socket, thereby resetting its timeout.
5237 This is not a problem when such interrupts are rare. Somebody set up a cron
5238 job to run exiwhat every 2 minutes, which is less than the normal select()
5239 timeout (5 or 10 minutes). This meant that the select() timeout never
5240 kicked in because it was always reset. I have fixed this by comparing the
5241 time when an interrupt arrives with the time at the start of the first call
5242 to select(). If more time than the timeout has elapsed, the interrupt is
5243 treated as a timeout.
5245 5. Some internal re-factoring in preparation for the addition of Sieve
5246 extensions (by MH). In particular, the "personal" test is moved to a
5247 separate function, and given an option for scanning Cc: and Bcc: (which is
5248 not set for Exim filters).
5250 6. When Exim created an email address using the login of the caller as the
5251 local part (e.g. when creating a From: or Sender: header line), it was not
5252 quoting the local part when it contained special characters such as @.
5254 7. Installed new OpenBSD configuration files.
5256 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5257 try to make them clearer.
5259 9. Callout options, other than the timeout value, were being ignored when
5260 verifying sender addresses in header lines. For example, when using
5262 verify = header_sender/callout=no_cache
5264 the cache was (incorrectly) being used.
5266 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5267 only the Cygwin environment.
5269 11. When return_path_on_delivery was set as a log selector, if different remote
5270 addresses in the same message used different return paths and parallel
5271 remote delivery occurred, the wrong values would sometimes be logged.
5272 (Whenever a remote delivery process finished, the return path value from
5273 the most recently started remote delivery process was logged.)
5275 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5276 lines when AUTH and/or TLS are in use. This is the "received protocol"
5277 field. Exim used to use "asmtp" for authenticated SMTP, without any
5278 indication (in the protocol name) for TLS use. Now it follows the RFC and
5279 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5280 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5281 names appear in log lines as well as in Received: header lines.
5283 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5284 extensions, and comparison tests, and to fix some bugs.
5286 14. Changes to the "personal" filter test:
5288 (1) The test was buggy in that it was just doing the equivalent of
5289 "contains" tests on header lines. For example, if a user's address was
5290 anne@some.where, the "personal" test would incorrectly be true for
5292 To: susanne@some.where
5294 This test is now done by extracting each address from the header in turn,
5295 and checking the entire address. Other tests that are part of "personal"
5296 are now done using regular expressions (for example, to check local parts
5297 of addresses in From: header lines).
5299 (2) The list of non-personal local parts in From: addresses has been
5300 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5301 taken from the Sieve specification recommendations.
5303 (3) If the message contains any header line starting with "List-" it is
5304 treated as non-personal.
5306 (4) The test for "circular" in the Subject: header line has been removed
5307 because it now seems ill-conceived.
5309 15. Minor typos in src/EDITME comments corrected.
5311 16. Installed latest exipick from John Jetmore.
5313 17. If headers_add on a router specified a text string that was too long for
5314 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5315 of string_sprintf() is now avoided.
5317 18. $message_body_size was not set (it was always zero) when running the DATA
5318 ACL and the local_scan() function.
5320 19. For the "mail" command in an Exim filter, no default was being set for
5321 the once_repeat time, causing a random time value to be used if "once" was
5322 specified. (If the value happened to be <= 0, no repeat happened.) The
5323 default is now 0s, meaning "never repeat". The "vacation" command was OK
5324 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5325 (I found it when inspecting the code).
5327 20. There is now an overall timeout for performing a callout verification. It
5328 defaults to 4 times the callout timeout, which applies to individual SMTP
5329 commands during the callout. The overall timeout applies when there is more
5330 than one host that can be tried. The timeout is checked before trying the
5331 next host. This prevents very long delays if there are a large number of
5332 hosts and all are timing out (e.g. when the network connections are timing
5333 out). The value of the overall timeout can be changed by specifying an
5334 additional sub-option for "callout", called "maxwait". For example:
5336 verify = sender/callout=5s,maxwait=20s
5338 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5339 to the end before writing, but this should make it even safer).
5341 22. Exim was forgetting that it had advertised PIPELINING for the second and
5342 subsequent messages on an SMTP connection. It was also not resetting its
5343 memory on STARTTLS and an internal HELO.
5345 23. When Exim logs an SMTP synchronization error within a session, it now
5346 records whether PIPELINING has been advertised or not.
5348 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5349 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5350 rather than long int.
5352 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5354 26. Added the never_mail option to autoreply.
5360 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5361 crash if the getsockname() call failed; this can happen if a connection is
5362 closed very soon after it is established. The problem was simply in the
5363 order in which certain operations were done, causing Exim to try to write
5364 to the SMTP stream before it had set up the file descriptor. The bug has
5365 been fixed by making things happen in the correct order.
5371 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5372 before the connection was closed, thus losing the rejection response.
5374 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5375 some early Solaris releases, but causes trouble in current releases where
5376 socklen_t is defined.
5378 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5381 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5384 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5385 that releasing the top of it at the end releases what was used for sub-
5386 expansions (unless the block got too big). However, discard this block if
5387 the first thing is a variable or header, so that we can use its block when
5388 it is dynamic (useful for very large $message_headers, for example).
5390 6. Lookups now cache *every* query, not just the most recent. A new, separate
5391 store pool is used for this. It can be recovered when all lookup caches are
5392 flushed. Lookups now release memory at the end of their result strings.
5393 This has involved some general refactoring of the lookup sources.
5395 7. Some code has been added to the store_xxx() functions to reduce the amount
5396 of flapping under certain conditions.
5398 8. log_incoming_interface used to affect only the <= reception log lines. Now
5399 it causes the local interface and port to be added to several more SMTP log
5400 lines, for example "SMTP connection from", and rejection lines.
5402 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5404 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5406 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5407 could be overwritten at the end of the current message (or the start of a
5408 new message if it was set in a HELO ACL). The value is now preserved for
5409 the duration of the SMTP connection.
5411 12. If a transport had a headers_rewrite setting, and a matching header line
5412 contained an unqualified address, that address was qualified, even if it
5413 did not match any rewriting rules. The underlying bug was that the values
5414 of the flags that permit the existence of unqualified sender and recipient
5415 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5416 non-local messages, and by -bnq for local messages) were not being
5417 preserved with the message after it was received.
5419 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5420 "next input=" as part of the text comprising the host identity instead of
5421 the correct text. The code was using the same buffer for two different
5422 strings. However, depending on which order the printing function evaluated
5423 its arguments, the bug did not always show up. Under Linux, for example, my
5424 test suite worked just fine.
5426 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5427 doesn't work with some older versions of Perl. It has been changed to "my",
5428 which in any case is probably the better facility to use.
5430 15. A really picky compiler found some instances of statements for creating
5431 error messages that either had too many or two few arguments for the format
5434 16. The size of the buffer for calls to the DNS resolver has been increased
5435 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5436 for addresses that have a lot of PTR records. This alleviates a problem; it
5437 does not fully solve it.
5439 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5440 buffer now truncates the list and logs the incident, which is the same
5441 action as happens when Exim is looking up a host name and its aliases.
5442 Previously in this situation something unpredictable would happen;
5443 sometimes it was "internal error: store_reset failed".
5445 18. If a server dropped the connection unexpectedly when an Exim client was
5446 using GnuTLS and trying to read a response, the client delivery process
5447 crashed while trying to generate an error log message.
5449 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5450 string, for example:
5452 warn message = H1: something\nH2: something
5454 the text was added as a single header line from Exim's point of view
5455 though it ended up OK in the delivered message. However, searching for the
5456 second and subsequent header lines using $h_h2: did not work. This has been
5457 fixed. Similarly, if a system filter added multiple headers in this way,
5458 the routers could not see them.
5460 20. Expanded the error message when iplsearch is called with an invalid key to
5461 suggest using net-iplsearch in a host list.
5463 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5464 ACLs are no longer actually imposed (and a message to that effect is
5467 22. If a "gecos" field in a passwd entry contained escaped characters, in
5468 particular, if it contained a \" sequence, Exim got it wrong when building
5469 a From: or a Sender: header from that name. A second bug also caused
5470 incorrect handling when an unquoted " was present following a character
5471 that needed quoting.
5473 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5474 was not being matched caselessly.
5476 24. Arranged for all hyphens in the exim.8 source to be escaped with
5479 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5480 properly. Recipient callout cache records were still being keyed to include
5481 the sender, even when use_sender was set false. This led to far more
5482 callouts that were necessary. The sender is no longer included in the key
5483 when use_sender is false.
5485 26. Added "control = submission" modifier to ACLs.
5487 27. Added the ${base62d: operator to decode base 62 numbers.
5489 28. dnsdb lookups can now access SRV records.
5491 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5492 the configuration file.
5494 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5495 (-v) mode. This makes the output for a verbose queue run more intelligible.
5497 31. Added a use_postmaster feature to recipient callouts.
5499 32. Added the $body_zerocount variable, containing the number of binary zero
5500 bytes in the message body.
5502 33. The time of last modification of the "new" subdirectory is now used as the
5503 "mailbox time last read" when there is a quota error for a maildir
5506 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5508 35. Added +ignore_unknown as a special item in host lists.
5510 36. Code for decoding IPv6 addresses in host lists is now included, even if
5511 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5512 address was recognized as an IP address, but was then not correctly decoded
5513 into binary, causing unexpected and incorrect effects when compared with
5520 1. Very minor rewording of debugging text in manualroute to say "list of
5521 hosts" instead of "hostlist".
5523 2. If verify=header_syntax was set, and a header line with an unqualified
5524 address (no domain) and a large number of spaces between the end of the
5525 name and the colon was received, the reception process suffered a buffer
5526 overflow, and (when I tested it) crashed. This was caused by some obsolete
5527 code that should have been removed. The fix is to remove it!
5529 3. When running in the test harness, delay a bit after writing a bounce
5530 message to get a bit more predictability in the log output.
5532 4. Added a call to search_tidyup() just before forking a reception process. In
5533 theory, someone could use a lookup in the expansion of smtp_accept_max_
5534 per_host which, without the tidyup, could leave open a database connection.
5536 5. Added the variables $recipient_data and $sender_data which get set from a
5537 lookup success in an ACL "recipients" or "senders" condition, or a router
5538 "senders" option, similar to $domain_data and $local_part_data.
5540 6. Moved the writing of debug_print from before to after the "senders" test
5543 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5544 problems for message scanning, either using a data ACL, or using
5545 local_scan() because the Received: header was not generated till after they
5546 were called (in order to set the time as the time of reception completion).
5547 I have revised the way this works. The header is now generated after the
5548 body is received, but before the ACL or local_scan() are called. After they
5549 are run, the timestamp in the header is updated.
5555 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5556 before starting a queue runner without re-exec. This happened only when
5557 deliver_drop_privilege was set or when the Exim user was set to root. The
5558 effect of the bug was that timeouts during subsequent deliveries caused
5559 crashes instead of being properly handled. The handler is now left at its
5560 default (and expected) setting.
5562 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5563 message, again when deliver_drop_privilege is set or Exim is run as root.
5564 The bug described in (1) was not present in this case, but the tidying up
5565 of the other signals was missing. I have made the two cases consistent.
5567 3. The ignore_target_hosts setting on a manualroute router was being ignored
5568 for hosts that were looked up using the /MX notation.
5570 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5573 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5574 operated on the sender address. After changing the $sender_address to <>
5575 for the sender address verify, Exim was re-instated it as the original
5576 (before rewriting) address, but remembering that it had rewritten it, so it
5577 wasn't rewriting it again. This bug also had the effect of breaking the
5578 sender address verification caching when the sender address was rewritten.
5580 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5581 This has been changed so that if the ip literal address matches
5582 ignore_target_hosts, the router declines.
5584 7. Added expansion conditions match_domain, match_address, and match_local_
5585 part (NOT match_host).
5587 8. The placeholder for the Received: header didn't have a length field set.
5589 9. Added code to Exim itself and to exim_lock to test for a specific race
5590 condition that could lead to file corruption when using MBX delivery. The
5591 issue is with the lockfile that is created in /tmp. If this file is removed
5592 after a process has opened it but before that process has acquired a lock,
5593 there is the potential for a second process to recreate the file and also
5594 acquire a lock. This could lead to two Exim processes writing to the file
5595 at the same time. The added code performs the same test as UW imapd; it
5596 checks after acquiring the lock that its file descriptor still refers to
5597 the same named file.
5599 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5600 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5603 11. Added the smtp_active_hostname option. If used, this will typically be made
5604 to depend on the incoming interface address. Because $interface_address is
5605 not set up until the daemon has forked a reception process, error responses
5606 that can happen earlier (such as "too many connections") no longer contain
5609 12. If an expansion in a condition on a "warn" statement fails because a lookup
5610 defers, the "warn" statement is abandoned, and the next ACL statement is
5611 processed. Previously this caused the whole ACL to be aborted.
5613 13. Added the iplsearch lookup type.
5615 14. Added ident_timeout as a log selector.
5617 15. Added tls_certificate_verified as a log selector.
5619 16. Added a global option tls_require_ciphers (compare the smtp transport
5620 option of the same name). This controls incoming TLS connections.
5622 17. I finally figured out how to make tls_require_ciphers do a similar thing
5623 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5624 before starting the TLS session.
5626 18. Tabs are now shown as \t in -bP output.
5628 19. If the log selector return_path_on_delivery was set, Exim crashed when
5629 bouncing a message because it had too many Received: header lines.
5631 20. If two routers both had headers_remove settings, and the first one included
5632 a superfluous trailing colon, the final name in the first list and the
5633 first name in the second list were incorrectly joined into one item (with a
5634 colon in the middle).
5640 1. Added -C and -D options to the exinext utility, mainly to make it easier
5641 to include in the automated testing, but these could be helpful when
5642 multiple configurations are in use.
5644 2. The exinext utility was not formatting the output nicely when there was
5645 an alternate port involved in the retry record key, nor when there was a
5646 message id as well (for retries that were specific to a specific message
5647 and a specific host). It was also confused by IPv6 addresses, because of
5648 the additional colons they contain. I have fixed the IPv4 problem, and
5649 patched it up to do a reasonable job for IPv6.
5651 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5652 delivery, the log line now contains "pipelined" if PIPELINING was used.
5654 4. An SMTP transport process used to panic and die if the bind() call to set
5655 an explicit outgoing interface failed. This has been changed; it is now
5656 treated in the same way as a connect() failure.
5658 5. A reference to $sender_host_name in the part of a conditional expansion
5659 that was being skipped was still causing a DNS lookup. This no longer
5662 6. The def: expansion condition was not recognizing references to header lines
5663 that used bh_ and bheader_.
5665 7. Added the _cache feature to named lists.
5667 8. The code for checking quota_filecount in the appendfile transport was
5668 allowing one more file than it should have been.
5670 9. For compatibility with Sendmail, the command line option
5678 and sets the incoming protocol and host name (for trusted callers). The
5679 host name and its colon can be omitted when only the protocol is to be set.
5680 Note the Exim already has two private options, -pd and -ps, that refer to
5681 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5682 "s", but I don't think that's a major issue.
5684 10. A number of refactoring changes to the code, none of which should affect
5687 (a) The number of logging options was getting close to filling up the
5688 32-bit word that was used as a bit map. I have split them into two classes:
5689 those that are passed in the argument to log_write(), and those that are
5690 only ever tested independently outside of that function. These are now in
5691 separate 32-bit words, so there is plenty of room for expansion again.
5692 There is no change in the user interface or the logging behaviour.
5694 (b) When building, for example, log lines, the code previously used a
5695 macro that called string_cat() twice, in order to add two strings. This is
5696 not really sufficiently general. Furthermore, there was one instance where
5697 it was actually wrong because one of the argument was used twice, and in
5698 one call a function was used. (As it happened, calling the function twice
5699 did not affect the overall behaviour.) The macro has been replaced by a
5700 function that can join an arbitrary number of extra strings onto a growing
5703 (c) The code for expansion conditions now uses a table and a binary chop
5704 instead of a serial search (which was left over from when there were very
5705 few conditions). Also, it now recognizes conditions like "pam" even when
5706 the relevant support is not compiled in: a suitably worded error message is
5707 given if an attempt is made to use such a condition.
5709 11. Added ${time_interval:xxxxx}.
5711 12. A bug was causing one of the ddress fields not to be passed back correctly
5712 from remote delivery subprocesses. The field in question was not being
5713 subsequently used, so this caused to problems in practice.
5715 13. Added new log selectors queue_time and deliver_time.
5717 14. Might have fixed a bug in maildirsizefile handling that threw up
5718 "unexpected character" debug warnings, and recalculated the data
5719 unnecessarily. In any case, I expanded the warning message to give more
5722 15. Added the message "Restricted characters in address" to the statements in
5723 the default ACL that block characters like @ and % in local parts.
5725 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5726 Three changes have been made:
5728 (a) There was a serious bug; a negative response to MAIL caused the whole
5729 recipient domain to be cached as invalid, thereby blocking all messages
5730 to all local parts at the same domain, from all senders. This bug has
5731 been fixed. The domain is no longer cached after a negative response to
5732 MAIL if the sender used is not empty.
5734 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5737 (c) A new callout option, "use_sender" has been added for people who want
5738 the modified behaviour.
5744 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5747 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5748 indeed breaks things for older releases.
5750 3. Added additional logging to the case where there is a problem reading data
5751 from a filter that is running in a subprocess using a pipe, in order to
5752 try to track down a specific problem.
5754 4. Testing facility fudge: when running in the test harness and attempting
5755 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5756 getting "No route to host". Convert this to a timeout.
5758 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5761 6. Some OS don't have socklen_t but use size_t instead. This affects the
5762 fifth argument of getsockopt() amongst other things. This is now
5763 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5764 can be set for individual OS. I have set it for SunOS5, OSF1, and
5765 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5766 some earlier ones do not.
5768 7. Change 4.30/15 was not doing the test caselessly.
5770 8. The standard form for an IPv6 address literal was being rejected by address
5771 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5772 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5773 this, as well as the form without the "IPv6" on the front (but only when
5774 address literals are enabled, of course).
5776 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5778 10. Exim crashed if a message with an empty sender address specified by -f
5779 encountered a router with an errors_to setting. This could be provoked only
5780 by a command such as
5784 where an empty string was supplied; "<>" did not hit this bug.
5786 11. Installed PCRE release 4.5.
5788 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5789 remained set. It is now erased.
5791 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5792 times from message ids (which are base 36 rather than the normal 62).
5794 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5795 were being counted as actual protocol errors, and logged if the log
5796 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5797 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5798 and DATA following a set of rejected RCPTs do not count as protocol errors.
5799 In other words, Exim assumes they were pipelined, though this may not
5800 actually be the case. Of course, in all cases the client gets an
5801 appropriate error code.
5803 15. If a lookup fails in an ACL condition, a message about the failure may
5804 be available; it is used if testing the ACL cannot continue, because most
5805 such messages specify what the cause of the deferral is. However, some
5806 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5807 that caused an old message to be retained and used if a later statement
5808 caused a defer, replacing the real cause of the deferral.
5810 16. If an IP address had so many PTR records that the DNS lookup buffer
5811 was not large enough to hold them, Exim could crash while trying to process
5812 the truncated data. It now detects and logs this case.
5814 17. Further to 4.21/58, another change has been made: if (and only if) the
5815 first line of a message (the first header line) ends with CRLF, a bare LF
5816 in a subsequent header line has a space inserted after it, so as not to
5817 terminate the header.
5819 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5820 unnecessarily, used atoi() instead of strtol(), and didn't check the
5821 termination when getting file sizes from file names by regex.
5823 19. Completely re-implemented the support for maildirsize files, in the light
5824 of a number of problems with the previous contributed implementation
5825 (4.30/29). In particular:
5827 . If the quota is zero, the maildirsize file is maintained, but no quota is
5830 . If the maildir directory does not exist, it is created before any attempt
5831 to write a maildirsize file.
5833 . The quota value in the file is just a cache; if the quota is changed in
5834 the transport, the new value overrides.
5836 . A regular expression is available for excluding directories from the
5839 20. The autoreply transport checks the characters in options that define the
5840 message's headers; it allows continued headers, but it was checking with
5841 isspace() after an embedded newline instead of explicitly looking for a
5844 21. If all the "regular" hosts to which an address was routed had passed their
5845 expiry times, and had not reached their retry times, the address was
5846 bounced, even if fallback hosts were defined. Now Exim should go on to try
5849 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5850 equivalent code in the SMTP transport. Some hosts send humungous responses
5851 to HELO/EHLO, more than 1024 it seems.
5853 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5854 gives compiler warnings in some environments. I've now done it "properly",
5857 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5858 (because of problems with the built-in one) was declared to return uschar *
5859 instead of char *, causing compiler failure.
5861 25. Fixed a file descriptor leak when processing alias/forward files.
5863 26. Fixed a minor format string issue in dbfn.c.
5865 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5867 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5868 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5871 29. When a sender address is verified, it is cached, to save repeating the test
5872 when there is more than one recipient in a message. However, when the
5873 verification involves a callout, it is possible for different callout
5874 options to be set for different recipients. It is too complicated to keep
5875 track of this in the cache, so now Exim always runs a verification when a
5876 callout is required, relying on the callout cache for the optimization.
5877 The overhead is duplication of the address routing, but this should not be
5880 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5881 to be verified with callout=postmaster, and the main callout worked but the
5882 postmaster check failed, the verification correctly failed. However, if a
5883 subsequent RCPT command asked for sender verification *without* the
5884 postmaster check, incorrect caching caused this verification also to fail,
5887 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5888 it was not caching the DNS options (qualify_single, search_parents) that
5889 were used when the lookup failed. A subsequent lookup with different
5890 options therefore always gave the same answer, though there were cases
5891 where it should not have. (Example: a "domains = !$mx_any" option on a
5892 dnslookup router: the "domains" option is always processed without any
5893 widening, but the router might have qualify_single set.) Now Exim uses the
5894 cached value only when the same options are set.
5896 32. Added John Jetmore's "exipick" utility to the distribution.
5898 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5899 than a timeout (e.g. a certificate is required, and is not provided), an
5900 Exim server now closes the connection immediately. Previously it waited for
5901 the client to close - but if the client is SSL, it seems that they each
5902 wait for each other, leading to a delay before one of them times out.
5904 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5905 maintained 0.8.x compatibility because I don't think many are using it, and
5906 it is clearly obsolete.
5908 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5911 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5912 client certificate was expired. A simple patch fixes this, though I don't
5913 understand the full logic of why the verify callback is called multiple
5916 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5917 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5918 which causes problems with some clients (such as the Certicom SSL Plus
5919 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5920 disables the coutermeasure allowing Eudora to connect."
5922 38. Exim was not checking that a write() to a log file succeeded. This could
5923 lead to Bad Things if a log got too big, in particular if it hit a file
5924 size limit. Exim now panics and dies if it cannot write to a log file, just
5925 as it does if it cannot open a log file.
5927 39. Modified OS/Makefile-Linux so that it now contains
5929 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5931 The two -D definitions ensure that Exim is compiled with large file
5932 support, which makes it possible to handle log files that are bigger than
5935 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5936 instance) a domain was checked against a named list that involved a lookup,
5937 causing $domain_data to be set, then another domain was checked against the
5938 same list, then the first domain was re-checked, the value of $domain_data
5939 after the final check could be wrong. In particular, if the second check
5940 failed, it could be set empty. This bug probably also applied to
5943 41. The strip_trailing_dot option was not being applied to the address given
5944 with the -f command-line option.
5946 42. The code for reading a message's header from the spool was incrementing
5947 $received_count, but never initializing it. This meant that the value was
5948 incorrect (doubled) while delivering a message in the same process in which
5949 it was received. In the most common configuration of Exim, this never
5950 happens - a fresh exec is done - but it can happen when
5951 deliver_drop_privilege is set.
5953 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5954 it now includes up to 150 characters of the unexpected data in the log
5957 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5958 and building data strings. The size of both of these buffers was 10 000
5959 bytes - far larger than anybody would *ever* want, thought I. Needless to
5960 say, somebody hit the limit. I have increased the maximum line length to
5961 20 000 and the maximum data length of concatenated lines to 100 000. I have
5962 also fixed two bugs, because there was no checking on these buffers. Tsk,
5963 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5964 buffer is too small.
5966 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5967 lsearch lookups. Now it does.
5969 46. When parsing a route_list item in a manualroute router, a fixed-length
5970 buffer was used for the list of hosts. I made this 1024 bytes long,
5971 thinking that nobody would ever have a list of hosts that long. Wrong.
5972 Somebody had a whole pile of complicated expansion conditions, and the
5973 string was silently truncated, leading to an expansion error. It turns out
5974 that it is easier to change to an unlimited length (owing to other changes
5975 that have happened since this code was originally written) than to build
5976 structure for giving a limitation error. The length of the item that
5977 expands into the list of hosts is now unlimited.
5979 47. The lsearch lookup could not handle data where the length of text line was
5980 more than 4095 characters. Such lines were truncated, leading to shortened
5981 data being returned. It should now handle lines of any length.
5983 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5984 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5987 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5988 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5989 no longer happens in comment lines. A list of replacements is now placed
5990 at the head of all of the source files, except those whose only change is
5991 to replace PERL_COMMAND in the very first #! line.
5993 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5994 messages on the queue, with a bottom-up merge sort, using code contributed
5995 by Michael Haardt. This should make operations like -bp somewhat faster on
5996 large queues. It won't affect queue runners, except when queue_run_in_order
5999 51. Installed eximstats 1.31 in the distribution.
6001 52. Added support for SRV lookups to the dnslookup router.
6003 53. If an ACL referred to $message_body or $message_body_end, the value was not
6004 reset for any messages that followed in the same SMTP session.
6006 54. The store-handling optimization for building very long strings was not
6007 differentiating between the different store pools. I don't think this
6008 actually made any difference in practice, but I've tidied it.
6010 55. While running the routers to verify a sender address, $sender_address
6011 was still set to the sender address. This is wrong, because when routing to
6012 send a bounce to the sender, it would be empty. Therefore, I have changed
6013 it so that, while verifying a sender address, $sender_address is set to <>.
6014 (There is no change to what happens when verifying a recipient address.)
6016 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6017 target A or AAAA records (if not already returned) without resetting the
6018 qualify_single or search_parents options of the DNS resolver. These are
6019 inappropriate in this case because the targets of MX and SRV records must
6020 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6021 target that, when qualified, matched something in the local domain. These
6022 two options are now turned off when doing these lookups.
6024 57. It seems that at least some releases of Reiserfs (which does not have the
6025 concept of a fixed number of inodes) returns zero and not -1 for the
6026 number of available inodes. This interacted badly with check_spool_inodes,
6027 which assumed that -1 was the "no such thing" setting. What I have done is
6028 to check that the total number of inodes is greater than zero before doing
6029 the test of how many are available.
6031 58. When a "warn" ACL statement has a log_message modifier, the message is
6032 remembered, and not repeated. This is to avoid a lot of repetition when a
6033 message has many recipients that cause the same warning to be written.
6034 However, Exim was preserving the list of already written lines for an
6035 entire SMTP session, which doesn't seem right. The memory is now reset if a
6036 new message is started.
6038 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6039 debugging output unless log_rewrite was also set.
6041 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6042 of (int)(handle) when we know that handle contains (void *)(-1).
6044 61. The Exim daemon panic-logs an error return when it closes the incoming
6045 connection. However "connection reset by peer" seems to be common, and
6046 isn't really an error worthy of noting specially, so that particular error
6049 62. When Exim is trying to find all the local interfaces, it used to panic and
6050 die if the ioctl to get the interface flags failed. However, it seems that
6051 on at least one OS (Solaris 9) it is possible to have an interface that is
6052 included in the list of interfaces, but for which you get a failure error
6053 for this call. This happens when the interface is not "plumbed" into a
6054 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6055 failure of the "get flags" call assumes that the interface is down.
6057 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6058 makes life easier for people who are doing arithmetic on fields extracted
6059 from dates, where you often get leading zeros that should not be
6060 interpreted as octal.
6062 64. Added qualify_domain to the redirect router, to override the global
6065 65. If a pathologically long header line contained very many addresses (the
6066 report of this problem mentioned 10 000) and each of them was rewritten,
6067 Exim could use up a very large amount of memory. (It kept on making new
6068 copies of the header line as it rewrote, and never released the old ones.)
6069 At the expense of a bit more processing, the header rewriting function has
6070 been changed so that it no longer eats memory in this way.
6072 66. The generation of the Received: header has been moved from the time that a
6073 message starts to be received, to the time that it finishes. The timestamp
6074 in the Received: header should now be very close to that of the <= log
6075 line. There are two side-effects of this change:
6077 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6078 logged header lines no longer include the local Received: line, because
6079 it has not yet been created. The same applies to a copy of the message
6080 that is returned to a non-SMTP sender when a message is rejected.
6082 (b) When a filter file is tested using -bf, no additional Received: header
6083 is added to the test message. After some thought, I decided that this
6086 This change does not affect the value of $received_for. It is still set
6087 after address rewriting, but before local_scan() is called.
6089 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6091 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6092 gave an unhelpful panic error message, and a defer error. I have managed to
6093 change this behaviour so that it now rejects any supplied certificate,
6094 which seems right, as the list of acceptable certificates is empty.
6096 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6097 gave an unhelpful defer error. I have not managed to make this reject any
6098 supplied certificates, but the error message it gives is "no certificate
6099 supplied", which is not helpful.
6101 70. exigrep's output now also includes lines that are not associated with any
6102 message, but which match the given pattern. Implemented by a patch from
6103 Martin Sluka, which also tidied up the Perl a bit.
6105 71. Recipient callout verification, like sender verification, was using <> in
6106 the MAIL FROM command. This isn't really the right thing, since the actual
6107 sender may affect whether the remote host accepts the recipient or not. I
6108 have changed it to use the actual sender in the callout; this means that
6109 the cache record is now keyed on a recipient/sender pair, not just the
6110 recipient address. There doesn't seem to be a real danger of callout loops,
6111 since a callout by the remote host to check the sender would use <>.
6112 [SEE ABOVE: changed after hitting problems.]
6114 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6115 temporary errors. However, in the case of such a code being given after
6116 the end of a data transmission (i.e. after ".") Exim was failing to write
6117 a retry record for the message. (Yes, there was some broken host that was
6118 actually sending 8xx at this point.)
6120 73. An unknown lookup type in a host list could cause Exim to panic-die when
6121 the list was checked. (An example that provoked this was putting <; in the
6122 middle of a list instead of at the start.) If this happened during a DATA
6123 ACL check, a -D file could be left lying around. This kind of configuration
6124 error no longer causes Exim to die; instead it causes a defer error. The
6125 incident is still logged to the main and panic logs.
6127 74. Buglet left over from Exim 3 conversion. The message "too many messages
6128 in one connection" was written to the rejectlog but not the mainlog, except
6129 when address rewriting (yes!) was being logged.
6131 75. Added write_rejectlog option.
6133 76. When a system filter was run not as root (that is, when system_filter_user
6134 was set), the values of the $n variables were not being returned to the
6135 main process; thus, they were not subsequently available in the $sn
6138 77. Added +return_path_on_delivery log selector.
6140 78. A connection timeout was being treated differently from recipients deferred
6141 when testing hosts_max_try with a message that was older than the host's
6142 retry timeout. (The host should not be counted, thus allowing all hosts to
6143 be tried at least once before bouncing.) This may have been the cause of an
6144 occasionally reported bug whereby a message would remain on the queue
6145 longer than the retry timeout, but would be bounced if a delivery was
6146 forced. I say "may" because I never totally pinned down the problem;
6147 setting up timeout/retry tests is difficult. See also the next item.
6149 79. The ultimate address timeout was not being applied to errors that involved
6150 a combination of host plus message (for example, a timeout on a MAIL
6151 command). When an address resolved to a number of possible hosts, and they
6152 were not all tried for each delivery (e.g. because of hosts_max_try), a
6153 message could remain on the queue longer than the retry timeout.
6155 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6158 81. Fixed an obscure SMTP outgoing bug which required at least the following
6159 conditions: (a) there was another message waiting for the same server;
6160 (b) the server returned 5xx to all RCPT commands in the first message so
6161 that the message was not completed; (c) the server dropped the connection
6162 or gave a negative response to the RSET that Exim sends to abort the
6163 transaction. The observed case was a dropped connection after DATA that had
6164 been sent in pipelining mode. That is, the server had advertised PIPELINING
6165 but was not implementing it correctly. The effect of the bug was incorrect
6166 behaviour, such as trying another host, and this could lead to a crash.
6172 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6173 and daemon.c were passed as pointers to ints; they should have been
6174 pointers to socklen_t variables (which are typically unsigned ints).
6176 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6179 3. Fixed a really odd bug that affected only the testing scheme; patching a
6180 certain fixed string in the binary changed the value of another string that
6181 happened to be identical to the end of the original first string.
6183 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6184 name", it returns that address as the IP address. On some operating
6185 systems (e.g. Solaris), it also passes back the IP address string as the
6186 "host name". However, on others (e.g. Linux), it passes back an empty
6187 string. Exim wasn't checking for this, and was changing the host name to an
6188 empty string, assuming it had been canonicalized.
6190 5. Although rare, it is permitted to have more than one PTR record for a given
6191 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6192 all the names associated with an address, because they do in Solaris.
6193 However, it seems that they do not in Linux for data that comes from the
6194 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6195 I found this out when I moved to a new Linux workstation and tried to run
6196 the Exim test suite.
6198 To get round this problem I have changed the code so that it now does its
6199 own call to the DNS to look up PTR records when searching for a host name.
6200 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6201 addresses that are only in /etc/hosts are still found.
6203 This behaviour is, however, controlled by an option called host_lookup_
6204 order, which defaults to "bydns:byaddr". If people want to use the other
6205 order, or indeed, just use one or the other means of lookup, they can
6206 specify it in this variable.
6208 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6209 some operating systems, this comes back from gethostbyaddr() as an empty
6210 string, and this is what Exim used to test for. However, it seems that in
6211 other systems, "." is yielded. Exim now tests for this case too.
6213 7. The values of check_spool_space and check_log_space are now held internally
6214 as a number of kilobytes instead of an absolute number of bytes. If a
6215 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6216 kilobyte. This means that much larger values can be stored.
6218 8. Exim monitor: an attempt to get the action menu when not actually pointing
6219 at a message produces an empty menu entitled "No message selected". This
6220 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6221 no entries in it ("Shell widget menu has zero width and/or height"). So I
6222 have added a single, blank menu entry in this case.
6224 9. Added ${quote_local_part.
6226 10. MIME decoding is now applied to the contents of Subject: header lines when
6229 11. Now that a reference to $sender_host_address automatically causes a reverse
6230 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6231 host lookup before query-style lookups in lists that might use this
6232 variable. This has therefore been abolished, and the "net-" prefix is no
6233 longer necessary for query-style lookups.
6235 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6236 have been a typo for LFLAGS, so it has been changed.
6238 13. The install script calls Exim with "-C /dev/null" in order to find the
6239 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6240 to be output. However, since Exim outputs its version number before the
6241 error, it didn't break the script. It just looked ugly. I fixed this by
6242 always allowing "-C /dev/null" if the caller is root.
6244 14. Ignore overlarge ACL variable number when reading spool file - insurance
6245 against a later release with more variables having written the file.
6247 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6248 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6249 this, as well as the form without the "IPv6" on the front.
6251 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6252 OS/Makefile-Darwin file.
6254 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6255 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6257 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6258 variable. However, some OS use ints for inodes. I've added cast to long int
6259 to get rid of the compiler warning.
6261 19. I had forgotten to lock out "/../" in configuration file names when
6262 ALT_CONFIG_PREFIX was set.
6264 20. Routers used for verification do not need to specify transports. However,
6265 if such a router generated a host list, and callout was configured, Exim
6266 crashed, because it could not find a port number from the (non-existent)
6267 transport. It now assumes port 25 in this circumstance.
6269 21. Added the -t option to exigrep.
6271 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6272 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6273 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6274 them is set, LOOKUP_LSEARCH is forced.
6276 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6278 24. Added sender and host information to the "rejected by local_scan()" log
6279 line; previously there was no indication of these.
6281 25. Added .include_if_exists.
6283 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6284 Linux. It turns out that not all file systems support this. Apparently some
6285 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6286 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6287 not supported on the file descriptor, is now ignored when Exim is trying to
6288 sync a directory. This applies only to Linux.
6290 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6292 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6293 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6296 29. Added support for maildirsize files from supplied patch (modified a bit).
6298 30. The use of :fail: followed by an empty string could lead Exim to respond to
6299 sender verification failures with (e.g.):
6301 550 Verification failed for <xxx>
6302 550 Sender verify failed
6304 where the first response line was missing the '-' that indicates it is not
6305 the final line of the response.
6307 31. The loop for finding the name of the user that called Exim had a hardwired
6308 limit of 10; it now uses the value of finduser_retries, which is used for
6309 all other user lookups.
6311 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6314 33. Exim was neglecting to zero errno before one call of strtol() when
6315 expanding a string and expecting an integer value. On some systems this
6316 resulted in spurious "integer overflow" errors. Also, it was casting the
6317 result into an int without checking.
6319 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6320 did not work. The code looks as if it has *never* worked, though it appears
6321 to have been documented since at least release 1.62. I have made it work.
6323 35. The "timeout_DNS" error in retry rules, also documented since at least
6324 1.62, also never worked. As it isn't clear exactly what this means, and
6325 clearly it isn't a major issue, I have abolished the feature by treating it
6326 as "timeout", and writing a warning to the main and panic logs.
6328 36. The display of retry rules for -brt wasn't always showing the error code
6331 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6332 timeout_connect_A, timeout_connect_MX.
6334 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6335 to the empty sender.
6337 39. The daemon was not analysing the content of -oX till after it had closed
6338 stderr and disconnected from the controlling terminal. This meant that any
6339 syntax errors were only noted on the panic log, and the return code from
6340 the command was 0. By re-arranging the code a little, I've made the
6341 decoding happen first, so such errors now appear on stderr, and the return
6342 code is 1. However, the actual setting up of the sockets still happens in
6343 the disconnected process, so errors there are still only recorded on the
6346 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6347 connections (as happens on some IP stacks) was logged at start up time as
6348 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6349 it from "IPv6 and IPv4", which means that two separate sockets are being
6352 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6353 says whether AF_INET or AF_INET6 was passed as an argument.
6355 42. Exiwhat output was messed up when time zones were included in log
6358 43. Exiwhat now gives more information about the daemon's listening ports,
6359 and whether -tls-on-connect was used.
6361 44. The "port" option of the smtp transport is now expanded.
6363 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6364 silently ignored. Now an error message is written to the main and panic
6367 46. There's a new ACL modifier called "logwrite" which writes to a log file
6368 as soon as it is encountered.
6370 47. Added $local_user_uid and $local_user_gid at routing time.
6372 48. Exim crashed when trying to verify a sender address that was being
6375 49. Exim was recognizing only a space character after ".include". It now also
6376 recognizes a tab character.
6378 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6379 extracting the relevant information from the specification. The man page no
6380 longer contains scrambled data for the -d option, and I've added a section
6381 at the front about calling Exim under different names.
6383 51. Added "extra_headers" argument to the "mail" command in filter files.
6385 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6388 53. Installed eximstats 1.29.
6390 54. Added transport_filter_timeout as a generic transport option.
6392 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6393 Cc: header lines. This was required by RFC 822, but it not required by RFC
6396 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6397 incoming messages that did not have them. Now it does so only if the
6398 message originates locally, that is, if there is no associated remote host
6399 address. When Resent- header lines are present, this applies to the Resent-
6400 lines rather than the non-Resent- lines.
6402 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6403 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6405 58. Messages for configuration errors now include the name of the main
6406 configuration file - useful now that there may be more than one file in a
6407 list (.included file names were always shown).
6409 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6410 for those rare installations that do not start the daemon as root or run it
6411 setuid root. I've cut out the call to initgroups() if the daemon is not
6414 60. The Exim user and group can now be bound into the binary as text strings
6415 that are looked up at the start of Exim's processing.
6417 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6419 62. Added $mailstore_basename variable.
6421 63. Installed patch to sieve.c from Michael Haardt.
6423 64. When Exim failed to open the panic log after failing to open the main log,
6424 the original message it was trying to log was written to stderr and debug
6425 output, but if they were not available (the usual case in production), it
6426 was lost. Now it is written to syslog before the two lines that record the
6427 failures to open the logs.
6429 65. Users' Exim filters run in subprocesses under the user's uid. It is
6430 possible for a "deliver" command or an alias in a "personal" command to
6431 provoke an address rewrite. If logging of address rewriting is configured,
6432 this fails because the process is not running as root or exim. There may be
6433 a better way of dealing with this, but for the moment (because 4.30 needs
6434 to be released), I have disabled address rewrite logging when running a
6435 filter in a non-root, non-exim process.
6441 1. The buildconfig auxiliary program wasn't quoting the value set for
6442 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6443 not defined. This bug was masked in 4.22 by the effect that was fixed in
6446 2. Some messages that were rejected after a message id was allocated were
6447 shown as "incomplete" by exigrep. It no longer does this for messages that
6448 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6450 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6451 have allow_domain_literals set, the ID did not get logged in the <= line.
6452 Domain literals are now always recognized in Message-ID: header lines.
6454 4. The first argument for a ${extract expansion item is the key name or field
6455 number. Leading and trailing spaces in this item were not being ignored,
6456 causing some misleading effects.
6458 5. When deliver_drop_privilege was set, single queue runner processes started
6459 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6460 same command in the process it spins off) were not dropping privilege.
6462 6. When the daemon running as "exim" started a queue runner, it always
6463 re-executed Exim in the spun-off process. This is a waste of effort when
6464 deliver_drop_privilege is set. The new process now just calls the
6465 queue-runner function directly.
6471 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6474 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6475 ignored. Though the use of -oP was forcing the writing of a pid file, it
6476 was always written to the default place.
6478 3. If the message "no IP address found for host xxxx" is generated during
6479 incoming verification, it is now followed by identification of the incoming
6480 connection (so you can more easily find what provoked it).
6482 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6484 5. Added some features to "harden" Exim a bit more against certain attacks:
6486 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6487 be put in Local/Makefile. This is like the never_users runtime option,
6488 but it cannot be overridden. The default setting is "root".
6490 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6491 prefix string with which any file named in a -C command line option
6494 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6495 is retained for -C and -D only if the caller of Exim is root. Without
6496 it, the exim user may also use -C and -D and retain privilege.
6498 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6499 command line option is disabled.
6501 6. Macro names set by the -D option must start with an upper case letter, just
6502 like macro names defined in the configuration file.
6504 7. Added "dereference=" facility to LDAP.
6506 8. Two instances of the typo "uknown" in the source files are fixed.
6508 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6509 the Configure-Makefile script screwed up while processing it.
6511 10. Incorporated PCRE 4.4.
6513 11. The SMTP synchronization check was not operating right at the start of an
6514 SMTP session. For example, it could not catch a HELO sent before the client
6515 waited for the greeting. There is now a check for outstanding input at the
6516 point when the greeting is written. Because of the duplex, asynchronous
6517 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6518 way, but not yet received, when the check is performed.
6520 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6521 on TCP/IP sockets, because this apparently causes some broken clients to
6524 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6525 unchanged) from the Cygwin maintainer.
6527 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6528 was supported instead of testing for mbx. Effectively a typo.
6530 15. The spa authenticator server code was not checking that the input it
6531 received was valid base64.
6533 16. The debug output line for the "set" modifier in ACLs was not showing the
6534 name of the variable that was being set.
6536 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6538 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6539 lookup on its own. Something else was needed to trigger the lookup. For
6540 example, a match in host_lookup or the need for a host name in a host list.
6541 Now, if $sender_host_name is referenced and the host name has not yet been
6542 looked up, a lookup is performed. If the lookup fails, the variable remains
6543 empty, and $host_lookup_failed is set to "1".
6545 19. Added "eqi" as a case-independent comparison operator.
6547 20. The saslauthd authentication condition could segfault if neither service
6548 nor realm was specified.
6550 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6551 error message that was logged was misleading, and incoming SMTP
6552 connections were dropped. The message is now more accurate, and temporary
6553 errors are given to SMTP connections.
6555 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6556 (except RSET) with a 421 error, until QUIT is received. However, it was
6557 failing to send a response to QUIT.
6559 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6560 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6561 ACL is now run after the helo_try_verify_hosts code.
6563 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6564 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6565 case-independent, so other case variants are also recognized). Apparently
6566 some systems use these upper case variants.
6568 25. If more than two messages were waiting for the same host, and a transport
6569 filter was specified for the transport, Exim sent two messages over the
6570 same TCP/IP connection, and then failed with "socket operation on non-
6571 socket" when it tried to send the third.
6573 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6575 27. The extern definition of crypt16() in expand.c was not being excluded when
6576 the OS had its own crypt16() function.
6578 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6579 as a preferred synonym for return_size_limit, both as an option and as an
6582 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6584 30. Changed the default configuration ACL to relax the local part checking rule
6585 for addresses that are not in any local domains. For these addresses,
6586 slashes and pipe symbols are allowed within local parts, but the sequence
6587 /../ is explicitly forbidden.
6589 31. SPA server authentication was not clearing the challenge buffer before
6592 32. log_message in a "warn" ACL statement was writing to the reject log as
6593 well as to the main log, which contradicts the documentation and doesn't
6594 seem right (because no rejection is happening). So I have stopped it.
6596 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6597 However, I am unable to do any testing of this.
6599 34. Fixed an infelicity in the appendfile transport. When checking directories
6600 for a mailbox, to see if any needed to be created, it was accidentally
6601 using path names with one or more superfluous leading slashes; tracing
6602 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6604 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6605 contents are added to the log line that is written for every discarded
6606 recipient. (Previously a log_message setting was ignored.)
6608 36. The ${quote: operator now quotes the string if it is empty.
6610 37. The install script runs exim in order to find its version number. If for
6611 some reason other than non-existence or emptiness, which it checks, it
6612 could not run './exim', it was installing it with an empty version number,
6613 i.e. as "exim-". This error state is now caught, and the installation is
6616 38. An argument was missing from the function that creates an error message
6617 when Exim fails to connect to the socket for saslauthd authentication.
6618 This could cause Exim to crash, or give a corrupted message.
6620 39. Added isip, isip4, and isip6 to ${if conditions.
6622 40. The ACL variables $acl_xx are now saved with the message, and can be
6623 accessed later in routers, transports, and filters.
6625 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6626 strings in the file are not string-expanded.
6628 42. If a MAIL command specified a SIZE value that was too large to fit into an
6629 int variable, the check against message_size_limit failed. Such values are
6630 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6631 day this will have to be increased, but I don't think I want to be around
6632 when emails are that large.
6639 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6640 iconv() is not standard in FreeBSD.
6642 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6643 IPv6 enabled. The observed symptom was a segmentation fault on return from
6644 the function os_common_find_running_interfaces() in src/os.c.
6646 3. In the check_special_case() function in daemon.c I had used "errno" as an
6647 argument name, which causes warnings on some systems. This was basically a
6648 typo, since it was named "eno" in the comments!
6650 4. The code that waits for the clock to tick (at a resolution of some fraction
6651 of a second) so as to ensure message-id uniqueness was always waiting for
6652 at least one whole tick, when it could have waited for less. [This is
6653 almost certainly not relevant at current processor speeds, where it is
6654 unlikely to ever wait at all. But we try to future-proof.]
6656 5. The function that sleeps for a time interval that includes fractions of a
6657 second contained a race. It did not block SIGALRM between setting the
6658 timer, and suspending (a couple of lines later). If the interval was short
6659 and the sigsuspend() was delayed until after it had expired, the suspension
6660 never ended. On busy systems this could lead to processes getting stuck for
6663 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6664 process, before it forks any delivery processes. The open lookup caching
6665 mechanism meant that the open file or database connection was passed into
6666 the delivery process. The problem was that delivery processes always tidy
6667 up cached lookup data. This could cause a problem for the next delivery
6668 process started by the queue runner, because the external queue runner
6669 process does not know about the closure. So the next delivery process
6670 still has data in the lookup cache. In the case of a file lookup, there was
6671 no problem because closing a file descriptor in a subprocess doesn't affect
6672 the parent. However, if the lookup was caching a connection to a database,
6673 the connection was closed, and the second delivery process was likely to
6674 see errors such as "PGSQL: query failed: server closed the connection
6675 unexpectedly". The problem has been fixed by closing all cached lookups
6676 in a queue runner before running a delivery process.
6678 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6679 seem to have the "const" qualifier which it has on other OS. I've
6682 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6683 *of the same type* (client or server) with the same public name that an
6684 error should be diagnosed.
6686 9. When Exim looked up a host name for an IP address, but failed to find the
6687 original IP address when looking up the host name (a safety check), it
6688 output the message "<ip address> does not match any IP for NULL", which was
6689 confusing, to say the least. The bug was that the host name should have
6690 appeared instead of "NULL".
6692 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6693 user that is built into the binary, and the -C or -D options is used, root
6694 privilege is dropped before the configuration file is read. In addition,
6695 logging is switched to stderr instead of the normal log files. If the
6696 configuration then re-defines the Exim user, the unprivileged environment
6697 is probably not what is expected, so Exim logs a panic warning message (but
6700 However, if deliver_drop_privilege is set, the unprivileged state may well
6701 be exactly what is intended, so the warning has been cut out in that case,
6702 and Exim is allowed to try to write to its normal log files.
6708 1. smtp_return_error_details was not giving details for temporary sender
6709 or receiver verification errors.
6711 2. Diagnose a configuration error if two authenticators have the same public
6714 3. Exim used not to create the message log file for a message until the first
6715 delivery attempt. This could be confusing when incoming messages were held
6716 for policy or load reasons. The message log file is now created at the time
6717 the message is received, and an initial "Received" line is written to it.
6719 4. The automatically generated man page for command line options had a minor
6720 bug that caused no ill effects; however, a more serious problem was that
6721 the procedure for building the man page automatically didn't always
6722 operate. Consequently, release 4.20 contains an out-of-date version. This
6723 shouldn't happen again.
6725 5. When building Exim with embedded Perl support, the script that builds the
6726 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6727 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6729 6. The freeze_tell option was not being used for messages that were frozen on
6730 arrival, either by an ACL or by local_scan().
6732 7. Added the smtp_incomplete_transaction log selector.
6734 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6735 was accepting AUTH without a new EHLO.
6737 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6738 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6741 10. Logging of TCP/IP connections (when configured) now happens in the main
6742 daemon process instead of the child process, so that the TCP/IP connection
6743 count is more accurate (but it can never be perfect).
6745 11. The use of "drop" in a nested ACL was not being handled correctly in the
6746 outer ACL. Now, if condition failure induced by the nested "drop" causes
6747 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6748 or "require"), the connection is dropped.
6750 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6751 that yield "discard" can now be used with an "accept" or a "discard" verb,
6752 but an error is generated for any others (because I can't see a useful way
6753 to define what should happen).
6755 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6756 are now processed in the same way as lines in the Exim configuration file.
6757 In particular, continuation lines are supported.
6759 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6761 15. Added -ti meaning -t -i.
6763 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6764 domains, and warn by logging if others are found.
6766 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6767 returned from GIFCONF when Exim is trying to find the list of interfaces on
6768 a host. The code in os.c has been modified to copy each ifreq to an aligned
6769 structure in all cases.
6771 Also, in some cases, the returned ifreq's were being copied to a 'struct
6772 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6773 means the last couple of bytes of an IPv6 address could be chopped if the
6774 ifreq contained only a normal sockaddr (14 bytes storage).
6776 18. Named domain lists were not supported in the hosts_treat_as_local option.
6777 An entry such as +xxxx was not recognized, and was treated as a literal
6780 19. Ensure that header lines added by a DATA ACL are included in the reject log
6781 if the ACL subsequently rejects the message.
6783 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6784 MD5 (which is deprecated).
6786 21. When testing a filter file using -bf, Exim was writing a message when it
6787 took the sender from a "From " line in the message, but it was not doing so
6788 when it took $return_path from a Return-Path: header line. It now does.
6790 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6791 with a valid header line field name (a series of printing characters
6792 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6794 23. Changed "disc" in the source to "disk" to conform to the documentation and
6795 the book and for uniformity.
6797 24. Ignore Sendmail's -Ooption=value command line item.
6799 25. When execve() failed while trying to run a command in a pipe transport,
6800 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6801 could be confused with a return value of 69 from the command itself. This
6802 has been changed to 127, the value the shell returns if it is asked to run
6803 a non-existent command. The wording for the related log line suggests a
6804 non-existent command as the problem.
6806 26. If received_header_text expands to an empty string, do not add a Received:
6807 header line to the message. (Well, it adds a token one on the spool, but
6808 marks it "old" so that it doesn't get used or transmitted.)
6810 27. Installed eximstats 1.28 (addition of -nt option).
6812 28. There was no check for failure on the call to getsockname() in the daemon
6813 code. This can fail if there is a shortage of resources on the system, with
6814 ENOMEM, for example. A temporary error is now given on failure.
6816 29. Contrary to the C standard, it seems that in some environments, the
6817 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6818 program. Exim now does this explicitly; it affects the formatting of
6819 timestamps using strftime().
6821 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6822 complaints. I've now initialized all the variables, to avoid this.
6824 32. Header lines added by a system filter were not being "seen" during
6825 transport-time rewrites.
6827 33. The info_callback() function passed to OpenSSL is set up with type void
6828 (*)(SSL *, int, int), as described somewhere. However, when calling the
6829 function (actually a macro) that sets it up, the type void(*)() is
6830 expected. I've put in a cast to prevent warnings from picky compilers.
6832 34. If a DNS black list lookup found a CNAME record, but there were no A
6833 records associated with the domain it pointed at, Exim crashed.
6835 35. If a DNS black list lookup returned more than one A record, Exim ignored
6836 all but the first. It now scans all returned addresses if a particular IP
6837 value is being sought. In this situation, the contents of the
6838 $dnslist_value variable are a list of all the addresses, separated by a
6841 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6842 to accept a host name and all its aliases if the forward lookup for any of
6843 them yielded the IP address of the incoming connection. Now it accepts only
6844 those names whose forward lookup yields the correct IP address. Any other
6845 names are discarded. This closes a loophole whereby a rogue DNS
6846 administrator could create reverse DNS records to break through a
6847 wildcarded host restriction in an ACL.
6849 37. If a user filter or a system filter that ran in a subprocess used any of
6850 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6851 the wrong values were passed to the pipe command ($thisaddress had the
6852 value of $0, $0 had the value of $1, etc). This bug was introduced by
6853 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6855 38. Improved the line breaking for long SMTP error messages from ACLs.
6856 Previously, if there was no break point between 40 and 75 characters, Exim
6857 left the rest of the message alone. Two changes have been made: (a) I've
6858 reduced the minimum length to 35 characters; (b) if it can't find a break
6859 point between 35 and 75 characters, it looks ahead and uses the first one
6860 that it finds. This may give the occasional overlong line, but at least the
6861 remaining text gets split now.
6863 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6864 file descriptors might be low, and that setting 1000 would always raise it.
6865 It turns out that in some environments, the limit is already over 1000 and
6866 that lowering it causes trouble. So now Exim takes care not to decrease it.
6868 40. When delivering a message, the value of $return_path is set to $sender_
6869 address at the start of routing (routers may change the value). By an
6870 oversight, this default was not being set up when an address was tested by
6871 -bt or -bv, which affected the outcome if any router or filter referred to
6874 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6875 log only when "message" or "log_message" are set. However, if one of the
6876 conditions was an address verification, or a call to a nested ACL, the
6877 messages generated by the underlying test were being passed through. This
6878 no longer happens. The underlying message is available in $acl_verify_
6879 message for both "message" and "log_message" expansions, so it can be
6880 passed through if needed.
6882 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6883 new expansion $bh_ to give the encoded byte string without charset
6884 translation. Translation happens only if iconv() is available; HAVE_ICONV
6885 indicates this at build time. HEADERS_CHARSET gives the charset to
6886 translate to; headers_charset can change it in the configuration, and
6887 "headers charset" can change it in an individual filter file.
6889 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6890 that creates RFC 2047 encoded "words" labels them as that charset instead
6891 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6892 expansion operator; (ii) when Exim creates a From: line for a local
6893 message; (iii) when a header line is rewritten to include a "phrase" part.
6895 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6896 buggy, causing it to skip the first lines of messages whose message ID
6897 ended in 'D'. This would not have bitten before Exim release 4.14, because
6898 message IDs were unlikely to end in 'D' before then. The effect was to have
6899 incorrect size information for certain domains.
6901 45. #include "config.h" was missing at the start of the crypt16.c module. This
6902 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6905 46. If there was a timeout during a "random" callout check, Exim treated it as
6906 a failure of the random address, and carried on sending RSET and the real
6907 address. If the delay was just some slowness somewhere, the response to the
6908 original RCPT would be taken as a response to RSET and so on, causing
6909 mayhem of various kinds.
6911 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6912 when I implemented it. It didn't allow for the fact that some option values
6913 may legitimately be negative (e.g. size_addition), and it didn't even do
6914 the right test for positive values.
6916 48. Domain names in DNS records are case-independent. Exim always looks them up
6917 in lower case. Some resolvers return domain names in exactly the case they
6918 appear in the zone file, that is, they may contain uppercase letters. Not
6919 all resolvers do this - some return always lower case. Exim was treating a
6920 change of case by a resolver as a change of domain, similar to a widening
6921 of a domain abbreviation. This triggered its re-routing code and so it was
6922 trying to route what was effectively the same domain again. This normally
6923 caused routing to fail (because the router wouldn't handle the domain
6924 twice). Now Exim checks for this case specially, and just changes the
6925 casing of the domain that it ultimately uses when it transmits the message
6928 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6931 50. If a filter generated a file delivery with a non-absolute name (possible if
6932 no home directory exists for the router), the forbid_file option was not
6935 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6936 the existing equality matching.
6938 52. Exim was using ints instead of ino_t variables in some places where it was
6939 dealing with inode numbers.
6941 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6942 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6943 called TMPDIR, and if it finds it is different, it changes its value.
6945 54. The smtp_printf() function is now made available to local_scan() so
6946 additional output lines can be written before returning. There is also an
6947 smtp_fflush() function to enable the detection of a dropped connection.
6948 The variables smtp_input and smtp_batched_input are exported to
6951 55. Changed the default runtime configuration: the message "Unknown user"
6952 has been removed from the ACL, and instead placed on the localuser router,
6953 using the cannot_route_message feature. This means that any verification
6954 failures that generate their own messages won't get overridden. Similarly,
6955 the "Unrouteable address" message that was in the ACL for unverifiable
6956 relay addresses has also been removed.
6958 56. Added hosts_avoid_esmtp to the smtp transport.
6960 57. The exicyclog script was not checking for the esoteric option
6961 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6962 will work only if exicyclog is run under the appropriate euid.
6964 58. Following a discussion on the list, the rules by which Exim recognises line
6965 endings on incoming messages have been changed. The -dropcr and drop_cr
6966 options are now no-ops, retained only for backwards compatibility. The
6967 following line terminators are recognized: LF CRLF CR. However, special
6968 processing applies to CR:
6970 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6971 nor a local message in the state where . is a terminator.
6973 (ii) If a bare CR is encountered in a header line, an extra space is added
6974 after the line terminator so as not to end the header. The reasoning
6975 behind this is that bare CRs in header lines are most likely either
6976 to be mistakes, or people trying to play silly games.
6978 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6979 was being incorrectly given as 18 bytes larger than it should have been.
6980 This is a VOB (very old bug).
6982 60. This may never have affected anything current, but just in case it has:
6983 When the local host is found other than at the start of a list of hosts,
6984 the local host, those with the same MX, and any that follow, are discarded.
6985 When the list in question was part of a longer list of hosts, the following
6986 hosts (not currently being processed) were also being discarded. This no
6987 longer happens. I'm not sure if this situation could ever has previously
6990 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6993 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6994 header. This is something that is recommended in a new Internet Draft, and
6995 is something that is documented as being done by Sendmail. There are two
6996 possible values. For messages generated by the autoreply transport, Exim
6999 Auto-Submitted: auto-replied
7001 whereas for all other generated messages (e.g. bounces) it adds
7003 Auto-Submitted: auto-generated
7005 63. The "personal" condition in filters now includes a test for the
7006 Auto-Submitted: header. If it contains the string "auto-" the message it
7007 not considered personal.
7009 64. Added rcpt_include_affixes as a generic transport option.
7011 65. Added queue_only_override (default true).
7013 66. Added the syslog_duplication option.
7015 67. If what should have been the first header line of a message consisted of
7016 a space followed by a colon, Exim was mis-interpreting it as a header line.
7017 It isn't of course - it is syntactically invalid and should therefore be
7018 treated as the start of the message body. The misbehaviour could have
7019 caused a number of strange effects, including loss of data in subsequent
7020 header lines, and spool format errors.
7022 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7023 client host had authenticated. This control can now be exercised by an ACL
7024 for more flexibility.
7026 69. By default, callouts do not happen when testing with -bh. There is now a
7027 variant, -bhc, which does actually run the callout code, including
7028 consulting and updating the callout cache.
7030 70. Added support for saslauthd authentication, courtesy of Alexander
7033 71. If statvfs() failed on the spool or log directories while checking their
7034 size for availability, Exim confusingly gave the error "space shortage".
7035 Furthermore, in debugging mode it crashed with a floating point exception.
7036 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7037 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7038 serious problem, Exim now writes to the main and panic logs when this
7039 happens, with details of the failure. It then refuses to accept the
7040 incoming message, giving the message "spool directory problem" or "log
7041 directory problem" with a 421 code for SMTP messages.
7043 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7044 0, 1, and 2 exist, because some OS complain for execs without them (see
7045 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7046 descriptors. However, the code omitted to check that the open succeeded,
7047 causing mysterious errors if for some reason the permissions on /dev/null
7048 got screwed. Now Exim writes a message to the main and panic logs, and
7049 bombs out if it can't open /dev/null.
7051 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7052 interact so that it is all more flexible. It is supposed to remain
7053 backwards compatible. Also added extra_local_interfaces.
7055 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7056 to bomb out with an assertion failure - to the client this appears as a
7057 connection drop. This problem occurs in the part of the code that was taken
7058 from the Samba project. Fortunately, the assertion is in a very simple
7059 function, so I have fixed this by reproducing the function inline in the
7060 one place where it is called, and arranging for authentication to fail
7061 instead of killing the process with assert().
7063 75. The SPA client code was not working when the server requested OEM rather
7064 than Unicode encoding.
7066 76. Added code to make require_files with a specific uid setting more usable in
7067 the case where statting the file as root fails - usually a non-root-mounted
7068 NFS file system. When this happens and the failure is EACCES, Exim now
7069 forks a subprocess and does the per-uid checking as the relevant uid.
7071 77. Added process_log_path.
7073 78. If log_file_path was not explicitly set, a setting of check_log_space or
7074 check_log_inodes was ignored.
7076 79. If a space check for the spool or log partitions fails, the incident is now
7077 logged. Of course, in the latter case the data may get lost...
7079 80. Added the %p formatting code to string_format() so that it can be used to
7080 print addresses in debug_print(). Adjusted all the address printing in the
7081 debugging in store.c to use %p rather than %d.
7083 81. There was a concern that a line of code in smtp_in.c could overflow a
7084 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7085 initially expressed, the concern was not well-founded, because trailing
7086 spaces are removed early. However, if the trailing spaces were followed by
7087 a NULL, they did not get removed, so the overflow was possible. Two fixes
7090 (a) I re-wrote the offending code in a cleaner fashion.
7091 (b) If an incoming SMTP command contains a NULL character, it is rejected
7094 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7095 runs initgroups(), so that if the Exim user is in any additional groups,
7096 they will be used during message reception.
7102 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob