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.
144 JH/28 Ensure that variables possibly set during message acceptance are marked
145 dead before release of memory in the daemon loop. This stops complaints
146 about them when the debug_store option is enabled. Discovered specifically
147 for sender_rate_period, but applies to a whole set of variables.
153 JH/01 Rework error string handling in TLS interface so that the caller in
154 more cases is responsible for logging. This permits library-sourced
155 string to be attached to addresses during delivery, and collapses
156 pairs of long lines into single ones.
158 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
159 during configuration. Wildcards are allowed and expanded.
161 JH/02 Rework error string handling in DKIM to pass more info back to callers.
162 This permits better logging.
164 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
165 do not close it down and have the child transport start it up again on
166 the passed-on TCP connection. Instead, proxy the child (and any
167 subsequent ones) for TLS via a unix-domain socket channel. Logging is
168 affected: the continued delivery log lines do not have any DNSSEC, TLS
169 Certificate or OCSP information. TLS cipher information is still logged.
171 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
172 identical IP addresses on different listening ports. Will also affect
175 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
176 add noisy ifdef guards to special-case this sillyness.
177 Patch from Bernd Kuhls.
179 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
180 than 255 are no longer allowed.
182 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
183 Disable the session-cache too, which might reduce our load. Since we
184 currrectly use a new context for every connection, both as server and
185 client, there is no benefit for these.
186 GnuTLS appears to not support tickets server-side by default (we don't
187 call gnutls_session_ticket_enable_server()) but client side is enabled
188 by default on recent versions (3.1.3 +) unless the PFS priority string
191 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
192 <https://reproducible-builds.org/specs/source-date-epoch/>.
194 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
195 the check for any unsuccessful recipients did not notice the limit, and
196 erroneously found still-pending ones.
198 JH/08 Pipeline CHUNKING command and data together, on kernels that support
199 MSG_MORE. Only in-clear (not on TLS connections).
201 JH/09 Avoid using a temporary file during transport using dkim. Unless a
202 transport-filter is involved we can buffer the headers in memory for
203 creating the signature, and read the spool data file once for the
204 signature and again for transmission.
206 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
207 4.77 as the kernel support then wasn't solid, having issues in 64bit
208 mode. Now, it's been long enough. Add support for FreeBSD also.
210 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
211 case where the routing stage had gathered several addresses to send to
212 a host before calling the transport for the first, we previously failed
213 to close down TLS in the old transport process before passing the TCP
214 connection to the new process. The new one sent a STARTTLS command
215 which naturally failed, giving a failed delivery and bloating the retry
216 database. Investigation and fix prototype from Wolfgang Breyha.
218 JH/12 Fix check on SMTP command input synchronisation. Previously there were
219 false-negatives in the check that the sender had not preempted a response
220 or prompt from Exim (running as a server), due to that code's lack of
221 awareness of the SMTP input buffering.
223 PP/04 Add commandline_checks_require_admin option.
224 Exim drops privileges sanely, various checks such as -be aren't a
225 security problem, as long as you trust local users with access to their
226 own account. When invoked by services which pass untrusted data to
227 Exim, this might be an issue. Set this option in main configuration
228 AND make fixes to the calling application, such as using `--` to stop
231 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
232 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
233 responses to those, into a single TLS record each way (this usually means
234 a single packet). As a side issue, smtp_enforce_sync now works on TLS
237 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
238 affects you only if you're dancing at the edge of the param size limits.
239 If you are, and this message makes sense to you, then: raise the
240 configured limit or use OpenSSL 1.1. Nothing we can do for older
243 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
244 cmdline element to get the documented default one. Previously it was
247 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
248 are unsuable for non-message-specific reasons - eg. connection timeout,
249 banner-time rejection.
251 JH/16 Fix logging of delivery remote port, when specified by router, under
254 PP/06 Repair manualroute's ability to take options in any order, even if one
255 is the name of a transport.
258 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
260 JH/17 Change the list-building routines interface to use the expanding-string
261 triplet model, for better allocation and copying behaviour.
263 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
264 Previously it was constructed the first time a possibly-matching string
265 was met in the configuration file input during startup; now it is done
268 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
269 compatible one, to avoid the (poorly documented) possibility of a config
270 file in the working directory redirecting the DB files, possibly correpting
271 some existing file. CVE-2017-10140 assigned for BDB.
273 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
274 cache-hot. Previously, although the result was properly cached, the
275 initial verify call returned a defer.
277 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
278 the main verify for receipient in uncached-mode.
280 JH/22 Retire historical build files to an "unsupported" subdir. These are
281 defined as "ones for which we have no current evidence of testing".
283 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
284 if present. Previously it was ignored.
286 JH/24 Start using specified-initialisers in C structure init coding. This is
287 a C99 feature (it's 2017, so now considered safe).
289 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
290 if was a fixed-sized field and bitmask ops via macros; it is now more
293 PP/07 GitHub PR 56: Apply MariaDB build fix.
294 Patch provided by Jaroslav Škarvada.
296 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
297 during Coverity cleanups [4.87 JH/47]
298 Diagnosis and fix provided by Michael Fischer v. Mollard.
300 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
301 the right size to place the terminating semicolon on its own folded
302 line, the header hash was calculated to an incorrect value thanks to
303 the (relaxed) space the fold became.
305 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
306 and confused the parent.
308 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
309 which could crash as a result. This could lead to undeliverable messages.
311 JH/28 Logging: "next input sent too soon" now shows where input was truncated
314 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
315 matters on fast-turnover and PID-randomising systems, which were getting
316 out-of-order delivery.
318 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
319 a possibly-overlapping copy. The symptom was that "Remote host closed
320 connection in response to HELO" was logged instead of the actual 4xx
323 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
324 Previously only that bufferd was discarded, resulting in SYMTP command
327 JH/32 DKIM: when a message has multiple signatures matching an identity given
328 in dkim_verify_signers, run the dkim acl once for each. Previously only
329 one run was done. Bug 2189.
331 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
332 "panic the current process" to "deliberately defer". The panic log is
333 still written with the problem list name; the mail and reject logs now
334 get a temp-reject line for the message that was being handled, saying
335 something like "domains check lookup or other defer". The SMTP 451
336 message is still "Temporary local problem".
338 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
339 A crafted sequence of BDAT commands could result in in-use memory beeing
340 freed. CVE-2017-16943.
342 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
343 from SMTP input. Previously it was always done; now only done for DATA
344 and not BDAT commands. CVE-2017-16944.
346 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
347 to the message (such as an overlong header line). Previously this was
348 not done and we did not exit BDAT mode. Followon from the previous item
349 though a different problem.
355 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
356 than -2003 did; needs libidn2 in addition to libidn.
358 JH/02 The path option on a pipe transport is now expanded before use.
360 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
361 Patch provided by "Björn", documentation fix added too.
363 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
364 missing a wire-to-host endian conversion.
366 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
367 close after a BDAT command line could be taken as a following command,
368 giving a synch failure. Fix by only checking for synch immediately
369 before acknowledging the chunk.
371 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
372 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
373 macro. Patches provided by Josh Soref.
375 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
376 Previously we did not; the RFC seems ambiguous and VRFY is not listed
377 by IANA as a service extension. However, John Klensin suggests that we
380 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
381 the dkim code may be unix-mode line endings rather than smtp wire-format
382 CRLF, so prepend a CR to any bare LF.
384 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
385 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
387 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
388 the first were themselves being wrongly included in the feed into dkim
389 processing; with most chunk sizes in use this resulted in an incorrect
390 body hash calculated value.
392 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
393 DKIM signature block, for verification. Although advised against by
394 standards it is specifically not ruled illegal.
396 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
398 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
399 missing a body hash (the bh= tag).
401 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
402 It seems that HAProxy sends the Proxy Protocol information in clear and
403 only then does a TLS startup, so do the same.
405 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
406 TCP connections (such as for Spamd) unless the daemon successfully set
407 Fast Open mode on its listening sockets. This fixes breakage seen on
408 too-old kernels or those not configured for Fast Open, at the cost of
409 requiring both directions being enabled for TFO, and TFO never being used
410 by non-daemon-related Exim processes.
412 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
413 endings, at least on the first header line. Try to canonify any that get
414 past that check, despite the cost.
416 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
417 now limited to an arbitrary five deep, while parsing addresses with the
418 strip_excess_angle_brackets option enabled.
420 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
421 instead leave the unprompted TLS handshake in socket buffer for the
422 TLS library to consume.
424 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
426 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
428 JH/16 Drop variables when they go out of scope. Memory management drops a whole
429 region in one operation, for speed, and this leaves assigned pointers
430 dangling. Add checks run only under the testsuite which checks all
431 variables at a store-reset and panics on a dangling pointer; add code
432 explicitly nulling out all the variables discovered. Fixes one known
433 bug: a transport crash, where a dangling pointer for $sending_ip_address
434 originally assigned in a verify callout, is re-used.
436 PP/06 Drop '.' from @INC in various Perl scripts.
438 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
440 PP/08 Reduce a number of compilation warnings under clang; building with
441 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
442 should be warning-free.
444 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
446 HS/01 Fix portability problems introduced by PP/08 for platforms where
447 realloc(NULL) is not equivalent to malloc() [SunOS et al].
449 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
450 chunk. This allows us to accept broken chunked messages. We need a more
451 general solution here.
453 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
454 already-broken messages in the queue.
456 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
458 JH/19 Fix reference counting bug in routing-generated-address tracking.
464 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
465 supports it and a size is available (ie. the sending peer gave us one).
467 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
468 years of being deprecated). The replacements are the ACLs
469 acl_smtp_mime and acl_not_smtp_mime.
471 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
472 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
473 or even an in-clear connection were permitted. Now, if the host lookup
474 was dnssec and dane was requested then the host is only used if the
475 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
476 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
477 if one fails this test.
478 This means that a poorly-configured remote DNS will make it incommunicado;
479 but it protects against a DNS-interception attack on it.
481 JH/04 Bug 1810: make continued-use of an open smtp transport connection
482 non-noisy when a race steals the message being considered.
484 JH/05 If main configuration option tls_certificate is unset, generate a
485 self-signed certificate for inbound TLS connections.
487 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
488 in rewrites and routers.
490 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
491 and logged a warning sing 4.83; now they are a configuration file error.
493 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
494 (lacking @domain). Apply the same qualification processing as RCPT.
496 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
498 JH/10 Support ${sha256:} applied to a string (as well as the previous
501 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
502 a cutthrough deliver is pending, as we always want to make a connection.
503 This also avoids re-routing the message when later placing the cutthrough
504 connection after a verify cache hit.
505 Do not update it with the verify result either.
507 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
508 when routing results in more than one destination address.
510 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
511 signing (which inhibits the cutthrough capability). Previously only
512 the presence of an option was tested; now an expansion evaluating as
513 empty is permissible (obviously it should depend only on data available
514 when the cutthrough connection is made).
516 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
517 the relevant preceding SMTP command did not note the pipelining mode.
519 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
520 Previously they were not counted.
522 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
523 as one having no matching records. Previously we deferred the message
524 that needed the lookup.
526 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
527 distinguished as "(=".
529 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
530 for missing MX records. Previously it only worked for missing A records.
532 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
534 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
535 after the data-go-ahead and data-ack. Patch from Jason Betts.
537 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
538 even for a "none" policy. Patch from Tony Meyer.
540 JH/22 Fix continued use of a connection for further deliveries. If a port was
541 specified by a router, it must also match for the delivery to be
544 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
545 When one of the recipients of a message was unsuitable for the connection
546 (has no matching addresses), we lost track of needing to mark it
547 deferred. As a result mail would be lost.
549 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
551 JH/25 Decoding ACL controls is now done using a binary search; the source code
552 takes up less space and should be simpler to maintain. Merge the ACL
553 condition decode tables also, with similar effect.
555 JH/26 Fix problem with one_time used on a redirect router which returned the
556 parent address unchanged. A retry would see the parent address marked as
557 delivered, so not attempt the (identical) child. As a result mail would
560 JH/27 Fix a possible security hole, wherein a process operating with the Exim
561 UID can gain a root shell. Credit to http://www.halfdog.net/ for
562 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
565 JH/28 Enable {spool,log} filesystem space and inode checks as default.
566 Main config options check_{log,spool}_{inodes,space} are now
567 100 inodes, 10MB unless set otherwise in the configuration.
569 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
570 Previously it only applied to the main-section connection policy
573 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
575 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
576 by me. Added RFC7919 DH primes as an alternative.
578 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
579 are not in the system include path.
581 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
582 GnuTLS, when a session startup failed (eg because the client disconnected)
583 Exim did stdio operations after fclose. This was exposed by a recent
584 change which nulled out the file handle after the fclose.
586 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
587 signed directly by the cert-signing cert, rather than an intermediate
588 OCSP-signing cert. This is the model used by LetsEncrypt.
590 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
592 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
593 an incoming connection.
595 HS/02 Bug 1802: Do not half-close the connection after sending a request
598 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
599 fallback to "prime256v1".
601 JH/34 SECURITY: Use proper copy of DATA command in error message.
602 Could leak key material. Remotely exploitable. CVE-2016-9963.
608 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
609 and 3.4.4 - once the server is enabled to respond to an OCSP request
610 it does even when not requested, resulting in a stapling non-aware
611 client dropping the TLS connection.
613 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
614 support variable-length bit vectors. No functional change.
616 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
617 The I= interface field on outgoing lines is now after the H= remote
618 host field, same as incoming lines. There is a separate
619 outgoing_interface log selector which allows you to disable the
622 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
623 If not running log_selector +smtp_connection the mainlog would be held
624 open indefinitely after a "too many connections" event, including to a
625 deleted file after a log rotate. Leave the per net connection logging
626 leaving it open for efficiency as that will be quickly detected by the
627 check on the next write.
629 HS/01 Bug 1671: Fix post transport crash.
630 Processing the wait-<transport> messages could crash the delivery
631 process if the message IDs didn't exist for some reason. When
632 using 'split_spool_directory=yes' the construction of the spool
633 file name failed already, exposing the same netto behaviour.
635 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
636 mime_regex ACL conditions.
638 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
639 to DSN fail messages (bounces): remote IP, remote greeting, remote response
640 to HELO, local diagnostic string.
642 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
643 log line to debug. Even when configured with a tls authenticator many
644 client connections are expected to not authenticate in this way, so
645 an authenticate fail is not an error.
647 HS/02 Add the Exim version string to the process info. This way exiwhat
648 gives some more detail about the running daemon.
650 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
651 matter for fast-change records such as DNSBLs.
653 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
654 retry record, even if constant. There may be multiple transports with
655 different interface settings and the retry behaviour needs to be kept
658 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
660 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
662 JH/10 Bug 840: fix log_defer_output option of pipe transport
664 JH/11 Bug 830: use same host for all RCPTS of a message, even under
665 hosts_randomize. This matters a lot when combined with mua_wrapper.
667 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
668 ${quote_pgsql:<string>} operator.
670 JH/13 Bug 1708: avoid misaligned access in cached lookup.
672 JH/14 Change header file name for freeradius-client. Relevant if compiling
673 with Radius support; from the Gentoo tree and checked under Fedora.
675 JH/15 Bug 1712: Introduce $prdr_requested flag variable
677 JH/16 Bug 1714: Permit an empty string as expansion result for transport
678 option transport_filter, meaning no filtering.
680 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
682 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
683 defaults to "*" (all hosts). The variable is now available when not built
684 with TLS, default unset, mainly to enable keeping the testsuite sane.
685 If a server certificate is not supplied (via tls_certificate) an error is
686 logged, and clients will find TLS connections fail on startup. Presumably
687 they will retry in-clear.
688 Packagers of Exim are strongly encouraged to create a server certificate
689 at installation time.
691 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
692 with the $config_file variable.
694 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
695 in transport context, after the attempt, and per-recipient. The latter type
696 is per host attempted. The event data is the error message, and the errno
697 information encodes the lookup type (A vs. MX) used for the (first) host,
698 and the trailing two digits of the smtp 4xx response.
700 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
701 to write to mainlog (or rejectlog, paniclog) in the window between file
702 creation and permissions/ownership being changed. Particularly affects
703 installations where exicyclog is run as root, rather than exim user;
704 result is that the running daemon panics and dies.
706 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
708 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
709 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
710 "pri" and "weight". Note that the previous implicit priority given by the
711 list order is no longer honoured.
713 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
716 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
717 by defining SUPPORT_SOCKS.
719 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
720 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
721 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
722 variables are renamed to proxy_{local,external}_{address,port}.
724 JH/27 Move Internationalisation support from Experimental to mainline, enabled
725 for a build by defining SUPPORT_I18N
727 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
728 of the query string, and make ${quote_redis:} do that quoting.
730 JH/29 Move Events support from Experimental to mainline, enabled by default
731 and removable for a build by defining DISABLE_EVENT.
733 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
735 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
736 cached by the daemon.
738 JH/32 Move Redis support from Experimental to mainline, enabled for a build
739 by defining LOOKUP_REDIS. The libhiredis library is required.
741 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
742 keys are given for lookup.
744 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
745 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
746 only supported when built with TLS support. The PolarSSL SHA routines
747 are still used when the TLS library is too old for convenient support.
749 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
750 openssl_options), for security. OpenSSL forces this from version 1.1.0
751 server-side so match that on older versions.
753 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
754 allocation for $value could be released as the expansion processing
755 concluded, but leaving the global pointer active for it.
757 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
758 and to use the domains and local_parts ACL conditions.
760 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
761 incorrectly not doubled on cutthrough transmission, hence seen as a
762 body-termination at the receiving system - resulting in truncated mails.
763 Commonly the sender saw a TCP-level error, and retransmitted the message
764 via the normal store-and-forward channel. This could result in duplicates
765 received - but deduplicating mailstores were liable to retain only the
766 initial truncated version.
768 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
770 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
772 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
773 we're in there, support oversigning also; bug 1309.
775 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
777 HS/04 Add support for keep_environment and add_environment options.
779 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
780 either intentional arithmetic overflow during PRNG, or testing config-
783 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
784 delivery resulted in actual delivery. Cancel cutthrough before DATA
787 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
788 rejects a recipient: pass the reject to the originator.
790 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
791 Many were false-positives and ignorable, but it's worth fixing the
794 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
795 for the new environment-manipulation done at startup. Move the routines
796 from being local to tls.c to being global via the os.c file.
798 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
799 an extract embedded as result-arg for a map, the first arg for extract
800 is unavailable so we cannot tell if this is a numbered or keyed
801 extraction. Accept either.
807 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
810 JH/02 The smtp transport option "multi_domain" is now expanded.
812 JH/03 The smtp transport now requests PRDR by default, if the server offers
815 JH/04 Certificate name checking on server certificates, when exim is a client,
816 is now done by default. The transport option tls_verify_cert_hostnames
817 can be used to disable this per-host. The build option
818 EXPERIMENTAL_CERTNAMES is withdrawn.
820 JH/05 The value of the tls_verify_certificates smtp transport and main options
821 default to the word "system" to access the system default CA bundle.
822 For GnuTLS, only version 3.0.20 or later.
824 JH/06 Verification of the server certificate for a TLS connection is now tried
825 (but not required) by default. The verification status is now logged by
826 default, for both outbound TLS and client-certificate supplying inbound
829 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
832 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
833 Status Notification (bounce) messages are now MIME format per RFC 3464.
834 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
835 under the control of the dsn_advertise_hosts option, and routers may
836 have a dsn_lasthop option.
838 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
839 default, modifiable by a malware= option. The list separator for
840 the options can now be changed in the usual way. Bug 68.
842 JH/10 The smtp_receive_timeout main option is now expanded before use.
844 JH/11 The incoming_interface log option now also enables logging of the
845 local interface on delivery outgoing connections.
847 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
848 if the interface and destination host and port all match.
850 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
853 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
854 Patch from Andrew Lewis.
856 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
857 now supports optional time-restrictions, weighting, and priority
858 modifiers per server. Patch originally by <rommer@active.by>.
860 JH/16 The spamd_address main option now supports a mixed list of local
861 and remote servers. Remote servers can be IPv6 addresses, and
862 specify a port-range.
864 JH/17 Bug 68: The spamd_address main option now supports an optional
865 timeout value per server.
867 JH/18 Bug 1581: Router and transport options headers_add/remove can
868 now have the list separator specified.
870 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
873 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
876 JH/21 Support for the A6 type of dns record is withdrawn.
878 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
879 rather than the verbs used.
881 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
882 from 255 to 1024 chars.
884 JH/24 Verification callouts now attempt to use TLS by default.
886 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
887 are generic router options now. The defaults didn't change.
889 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
890 Original patch from Alexander Shikoff, worked over by JH.
892 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
893 files if ZCAT_COMMAND is not executable.
895 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
897 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
899 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
900 Normally benign, it bites when the pair was led to by a CNAME;
901 modern usage is to not canonicalize the domain to a CNAME target
902 (and we were inconsistent anyway for A-only vs AAAA+A).
904 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
906 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
907 when evaluating $sender_host_dnssec.
909 JH/31 Check the HELO verification lookup for DNSSEC, adding new
910 $sender_helo_dnssec variable.
912 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
914 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
916 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
918 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
919 documented as working, but never had. Support all but $spam_report.
921 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
922 added for tls authenticator.
924 HS/03 Add perl_taintmode main config option
930 TL/01 When running the test suite, the README says that variables such as
931 no_msglog_check are global and can be placed anywhere in a specific
932 test's script, however it was observed that placement needed to be near
933 the beginning for it to behave that way. Changed the runtest perl
934 script to read through the entire script once to detect and set these
935 variables, reset to the beginning of the script, and then run through
936 the script parsing/test process like normal.
938 TL/02 The BSD's have an arc4random API. One of the functions to induce
939 adding randomness was arc4random_stir(), but it has been removed in
940 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
941 function when detected.
943 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
944 cause callback expansion.
946 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
947 syntax errors in an expansion can be treated as a string instead of
948 logging or causing an error, due to the internal use of bool_lax
949 instead of bool when processing it.
951 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
952 server certificates when making smtp deliveries.
954 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
956 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
958 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
960 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
961 Merged patch from Sebastian Wiedenroth.
963 JH/05 Fix results-pipe from transport process. Several recipients, combined
964 with certificate use, exposed issues where response data items split
965 over buffer boundaries were not parsed properly. This eventually
966 resulted in duplicates being sent. This issue only became common enough
967 to notice due to the introduction of connection certificate information,
968 the item size being so much larger. Found and fixed by Wolfgang Breyha.
970 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
971 size buffer was used, resulting in syntax errors when an expansion
974 JH/07 Add support for directories of certificates when compiled with a GnuTLS
975 version 3.3.6 or later.
977 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
978 is EXPERIMENTAL_EVENT, the main-configuration and transport options
979 both become "event_action", the variables become $event_name, $event_data
980 and $event_defer_errno. There is a new variable $verify_mode, usable in
981 routers, transports and related events. The tls:cert event is now also
982 raised for inbound connections, if the main configuration event_action
985 TL/06 In test suite, disable OCSP for old versions of openssl which contained
986 early OCSP support, but no stapling (appears to be less than 1.0.0).
988 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
989 server certificate names available under the smtp transport option
990 "tls_verify_cert_hostname" now do not permit multi-component wildcard
993 JH/10 Time-related extraction expansions from certificates now use the main
994 option "timezone" setting for output formatting, and are consistent
995 between OpenSSL and GnuTLS compilations. Bug 1541.
997 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
998 encoded parameter in the incoming message. Bug 1558.
1000 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1001 include certificate info, eximon was claiming there were spoolfile
1004 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1006 JH/14 Log delivery-related information more consistently, using the sequence
1007 "H=<name> [<ip>]" wherever possible.
1009 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1010 are problematic for Debian distribution, omit them from the release
1013 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1015 JH/16 Fix string representation of time values on 64bit time_t architectures.
1018 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1019 output list separator was used.
1024 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1025 checkers that were complaining about end of non-void function with no
1028 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1029 This was a regression introduced in 4.83 by another bugfix.
1031 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1033 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1034 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1040 TF/01 Correctly close the server side of TLS when forking for delivery.
1042 When a message was received over SMTP with TLS, Exim failed to clear up
1043 the incoming connection properly after forking off the child process to
1044 deliver the message. In some situations the subsequent outgoing
1045 delivery connection happened to have the same fd number as the incoming
1046 connection previously had. Exim would try to use TLS and fail, logging
1047 a "Bad file descriptor" error.
1049 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1050 utilities have not been installed.
1052 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1053 temporary space as the ACL may create new global variables.
1055 TL/01 LDAP support uses per connection or global context settings, depending
1056 upon the detected version of the libraries at build time.
1058 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1059 to extract and use the src ip:port in logging and expansions as if it
1060 were a direct connection from the outside internet. PPv2 support was
1061 updated based on HAProxy spec change in May 2014.
1063 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1065 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1066 Properly escape header and check for NULL return.
1068 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1071 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1073 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1074 characters in header names, implemented as a verify condition.
1075 Contributed by Michael Fischer v. Mollard.
1077 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1078 results permerror and temperror. Previous values are deprecated but
1079 still accepted. In a future release, err_perm and err_temp will be
1080 completely removed, which will be a backward incompatibility if the
1081 ACL tests for either of these two old results. Patch contributed by
1082 user bes-internal on the mailing list.
1084 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1086 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1087 selectors, in both main and reject logs.
1089 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1092 JH/07 Add malware type "sock" for talking to simple daemon.
1094 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1096 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1097 routers/transports under cutthrough routing.
1099 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1100 numbers. Touch up "bool" conditional to keep the same definition.
1102 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1104 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1106 JH/12 Expand items in router/transport headers_add or headers_remove lists
1107 individually rather than the list as a whole. Bug 1452.
1109 Required for reasonable handling of multiple headers_ options when
1110 they may be empty; requires that headers_remove items with embedded
1111 colons must have them doubled (or the list-separator changed).
1113 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1114 view the policy declared in the DMARC record. Currently, $dmarc_status
1115 is a combined value of both the record presence and the result of the
1118 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1120 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1121 dnslookup router and the smtp transport (applying to the forward
1124 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1125 of ldap servers used for a specific lookup. Patch provided by Heiko
1128 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1129 New variable $lookup_dnssec_authenticated for observability.
1131 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1132 Patch submitted by Lars Timman.
1134 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1136 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1137 Requires trusted mode and valid format message id, aborts otherwise.
1138 Patch contributed by Heiko Schlichting.
1140 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1141 certextract with support for various fields. Bug 1358.
1143 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1144 is requested by default, modifiable by smtp transport option
1147 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1148 operate on certificate variables to give certificate fingerprints
1149 Also new ${sha256:cert_variable}.
1151 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1153 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1154 Christian Aistleitner.
1156 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1158 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1159 file. Patch from Wolfgang Breyha.
1161 JH/25 Expand the coverage of the delivery $host and $host_address to
1162 client authenticators run in verify callout. Bug 1476.
1164 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1165 align with daemon_smtp_ports. Bug 72.
1167 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1168 support and error reporting did not work properly.
1170 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1171 and is readable. Patch from Andrew Colin Kissa.
1173 TL/14 Enhance documentation of ${run expansion and how it parses the
1174 commandline after expansion, particularly in the case when an
1175 unquoted variable expansion results in an empty value.
1177 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1179 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1180 by renaming away from it. Addresses GH issue 15, reported by
1183 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1185 TL/15 SECURITY: prevent double expansion in math comparison functions
1186 (can expand unsanitized data). Not remotely exploitable.
1193 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1195 PP/02 Make -n do something, by making it not do something.
1196 When combined with -bP, the name of an option is not output.
1198 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1201 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1202 $sender_host_name and config options to manage this, and basic check
1205 PP/05 DSCP support for outbound connections and control modifier for inbound.
1207 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1208 (Only plugin which currently uses this is kerberos4, which nobody should
1209 be using, but we should make it available and other future plugins might
1210 conceivably use it, even though it would break NAT; stuff *should* be
1211 using channel bindings instead).
1213 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1214 name; added for Sendmail compatibility; requires admin caller.
1215 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1216 just ignore it); requires trusted caller.
1217 Also parse but ignore: -Ac -Am -X<logfile>
1220 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1222 TL/02 Add +smtp_confirmation as a default logging option.
1224 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1225 Patch by Magnus Holmgren from 2007-02-20.
1227 TL/04 Bugzilla 1281 - Spec typo.
1228 Bugzilla 1283 - Spec typo.
1229 Bugzilla 1290 - Spec grammar fixes.
1231 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1233 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1235 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1236 mailing list by by Dmitry Isaikin.
1238 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1240 JH/02 Support "G" suffix to numbers in ${if comparisons.
1242 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1244 NM/01 Bugzilla 1197 - Spec typo
1245 Bugzilla 1196 - Spec examples corrections
1247 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1249 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1250 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1253 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1254 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1256 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1257 "acl {{name}{arg}...}", and optional args on acl condition
1260 JH/05 Permit multiple router/transport headers_add/remove lines.
1262 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1264 JH/07 Avoid using a waiting database for a single-message-only transport.
1265 Performance patch from Paul Fisher. Bugzilla 1262.
1267 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1270 JH/09 Add $headers_added variable, with content from use of ACL modifier
1271 add_header (but not yet added to the message). Bugzilla 199.
1273 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1274 Pulled from Bugzilla 817 by Wolfgang Breyha.
1276 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1278 (nb: this is the same fix as in Exim 4.80.1)
1280 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1283 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1286 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1288 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1289 advertise SMTP AUTH mechanism to us, instead of a generic
1290 protocol violation error. Also, make Exim more robust to bad
1291 data from the Dovecot auth socket.
1293 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1295 When a queue runner is handling a message, Exim first routes the
1296 recipient addresses, during which it prunes them based on the retry
1297 hints database. After that it attempts to deliver the message to
1298 any remaining recipients. It then updates the hints database using
1301 So if a recipient address works intermittently, it can get repeatedly
1302 deferred at routing time. The retry hints record remains fresh so the
1303 address never reaches the final cutoff time.
1305 This is a fairly common occurrence when a user is bumping up against
1306 their storage quota. Exim had some logic in its local delivery code
1307 to deal with this. However it did not apply to per-recipient defers
1308 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1310 This change adds a proper retry rule check during routing so that the
1311 final cutoff time is checked against the message's age. We only do
1312 this check if there is an address retry record and there is not a
1313 domain retry record; this implies that previous attempts to handle
1314 the address had the retry_use_local_parts option turned on. We use
1315 this as an approximation for the destination being like a local
1316 delivery, as in LMTP.
1318 I suspect this new check makes the old local delivery cutoff check
1319 redundant, but I have not verified this so I left the code in place.
1321 TF/02 Correct gecos expansion when From: is a prefix of the username.
1323 Test 0254 submits a message to Exim with the header
1327 When I ran the test suite under the user fanf2, Exim expanded
1328 the header to contain my full name, whereas it should have added
1329 a Resent-Sender: header. It erroneously treats any prefix of the
1330 username as equal to the username.
1332 This change corrects that bug.
1334 GF/01 DCC debug and logging tidyup
1335 Error conditions log to paniclog rather than rejectlog.
1336 Debug lines prefixed by "DCC: " to remove any ambiguity.
1338 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1340 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1341 Bug spotted by Jeremy Harris; was flawed since initial commit.
1342 Would have resulted in OCSP responses post-SNI triggering an Exim
1343 NULL dereference and crash.
1345 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1347 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1348 Bug detection, analysis and fix by Samuel Thibault.
1349 Bugzilla 1331, Debian bug #698092.
1351 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1353 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1354 Server implementation by Todd Lyons, client by JH.
1355 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1356 config variable "prdr_enable" controls whether the server
1357 advertises the facility. If the client requests PRDR a new
1358 acl_data_smtp_prdr ACL is called once for each recipient, after
1359 the body content is received and before the acl_smtp_data ACL.
1360 The client is controlled by both of: a hosts_try_prdr option
1361 on the smtp transport, and the server advertisement.
1362 Default client logging of deliveries and rejections involving
1363 PRDR are flagged with the string "PRDR".
1365 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1366 fclose(). Diagnosis by Todd Lyons.
1368 PP/17 Update configure.default to handle IPv6 localhost better.
1369 Patch by Alain Williams (plus minor tweaks).
1372 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1373 This is now consistent with GnuTLS, and is now documented: the
1374 previous undocumented portable approach to treating the option as
1375 unset was to force an expansion failure. That still works, and
1376 an empty string is now equivalent.
1378 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1379 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1380 not performing validation itself.
1382 PP/20 Added force_command boolean option to pipe transport.
1383 Patch from Nick Koston, of cPanel Inc.
1385 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1388 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1390 PP/21 Fix eximon continuous updating with timestamped log-files.
1391 Broken in a format-string cleanup in 4.80, missed when I repaired the
1392 other false fix of the same issue.
1393 Report and fix from Heiko Schlichting.
1396 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1397 Report from Prashanth Katuri.
1399 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1400 It's SecureTransport, so affects any MacOS clients which use the
1401 system-integrated TLS libraries, including email clients.
1403 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1404 using a MIME ACL for non-SMTP local injection.
1405 Report and assistance in diagnosis by Warren Baker.
1407 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1409 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1411 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1412 last id that failed so it may be referenced in subsequent ACL's.
1414 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1417 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1418 ldap library initialization, allowing self-signed CA's to be
1419 used. Also properly sets require_cert option later in code by
1420 using NULL (global ldap config) instead of ldap handle (per
1421 session). Bug diagnosis and testing by alxgomz.
1423 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1424 the src/util/ subdirectory.
1426 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1427 renamed to Transport Post Delivery Action by Jeremy Harris, as
1430 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1431 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1432 redis_servers = needs to be configured which will be used by the redis
1433 lookup. Patch from Warren Baker, of The Packet Hub.
1435 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1437 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1438 hostname or reverse DNS when processing a host list. Used suggestions
1439 from multiple comments on this bug.
1441 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1443 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1444 Missed a few lines, added it to make the runtest require no keyboard
1447 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1448 contains upper case chars. Make router use caseful_local_part.
1450 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1451 support when GnuTLS has been built with p11-kit.
1457 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1459 This, or similar/improved, will also be change PP/11 of 4.82.
1465 PP/01 Handle short writes when writing local log-files.
1466 In practice, only affects FreeBSD (8 onwards).
1467 Bugzilla 1053, with thanks to Dmitry Isaikin.
1469 NM/01 Bugzilla 949 - Documentation tweak
1471 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1474 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1476 PP/02 Implemented gsasl authenticator.
1478 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1480 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1481 `pkg-config foo` for cflags/libs.
1483 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1484 with rest of GSASL and with heimdal_gssapi.
1486 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1487 `pkg-config foo` for cflags/libs for the TLS implementation.
1489 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1490 properties get this fed in as external SSF. A number of robustness
1491 and debugging improvements to the cyrus_sasl authenticator.
1493 PP/08 cyrus_sasl server now expands the server_realm option.
1495 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1496 Patch by Jeremy Harris.
1498 PP/10 Added dbmjz lookup type.
1500 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1502 PP/12 MAIL args handles TAB as well as SP, for better interop with
1503 non-compliant senders.
1504 Analysis and variant patch by Todd Lyons.
1506 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1507 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1508 Patch from Dirk Mueller <dmueller@suse.com>
1510 PP/13 tls_peerdn now print-escaped for spool files.
1511 Observed some $tls_peerdn in wild which contained \n, which resulted
1512 in spool file corruption.
1514 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1515 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1516 or write after TLS renegotiation, which otherwise led to messages
1519 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1520 as a tracking header (ie: a signed header comes before the signature).
1521 Patch from Wolfgang Breyha.
1523 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1524 comma-sep list; embedded commas doubled.
1526 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1528 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1530 Report and patch from Dmitry Banschikov.
1532 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1533 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1534 failures. We appear to now support TLS1.1+ with Exim.
1536 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1537 lets Exim select keys and certificates based upon TLS SNI from client.
1538 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1539 before an outbound SMTP session. New log_selector, +tls_sni.
1541 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1542 NULL dereference. Report and patch from Alun Jones.
1544 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1545 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1546 Not seeing resolver debug output on NetBSD, but suspect this is a
1547 resolver implementation change.
1549 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1550 Left warnings. Added "eximon gdb" invocation mode.
1552 PP/21 Defaulting "accept_8bitmime" to true, not false.
1554 PP/22 Added -bw for inetd wait mode support.
1556 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1557 locate the relevant includes and libraries. Made this the default.
1559 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1560 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1562 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1563 This may cause build issues on older platforms.
1565 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1566 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1567 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1568 Added SNI support via GnuTLS too.
1569 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1571 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1573 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1574 Applied second patch from Janne, implementing suggestion to default
1575 multiple-strings-in-record handling to match SPF spec.
1577 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1579 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1580 read-only, out of scope).
1581 Patch from Wolfgang Breyha, report from Stuart Northfield.
1583 PP/29 Fix three issues highlighted by clang analyser static analysis.
1584 Only crash-plausible issue would require the Cambridge-specific
1585 iplookup router and a misconfiguration.
1586 Report from Marcin Mirosław.
1588 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1590 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1591 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1592 As part of this, removing so much warning spew let me fix some minor
1593 real issues in debug logging.
1595 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1596 assignment on my part. Fixed.
1598 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1599 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1600 Janne Snabb (who went above and beyond: thank you).
1602 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1603 string otherwise requires a connection and a bunch more work and it's
1604 relatively easy to get wrong. Should also expose TLS library linkage
1607 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1608 64-bit ${eval} (JH/03).
1610 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1611 GNU libc to support some of the 64-bit stuff, should not lead to
1612 conflicts. Defined before os.h is pulled in, so if a given platform
1613 needs to override this, it can.
1615 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1616 protection layer was required, which is not implemented.
1617 Bugzilla 1254, patch from Wolfgang Breyha.
1619 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1620 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1621 tls_dhparam take prime identifiers. Also unbreak combination of
1622 OpenSSL+DH_params+TLSSNI.
1624 PP/39 Disable SSLv2 by default in OpenSSL support.
1630 PP/01 Solaris build fix for Oracle's LDAP libraries.
1631 Bugzilla 1109, patch from Stephen Usher.
1633 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1635 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1638 TF/02 Fix a couple more cases where we did not log the error message
1639 when unlink() failed. See also change 4.74-TF/03.
1641 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1642 lock up or crash if it happened to be inside a call to libc when it
1643 got a SIGUSR1 from exiwhat.
1645 The SIGUSR1 handler appends the current process status to the process
1646 log which is later printed by exiwhat. It used to use the general
1647 purpose logging code to do this, but several functions it calls are
1648 not safe for signals.
1650 The new output code in the SIGUSR1 handler is specific to the process
1651 log, and simple enough that it's easy to inspect for signal safety.
1652 Removing some special cases also simplifies the general logging code.
1653 Removing the spurious timestamps from the process log simplifies
1656 TF/04 Improved ratelimit ACL condition.
1658 The /noupdate option has been deprecated in favour of /readonly which
1659 has clearer semantics. The /leaky, /strict, and /readonly update modes
1660 are mutually exclusive. The update mode is no longer included in the
1661 database key; it just determines when the database is updated. (This
1662 means that when you upgrade Exim will forget old rate measurements.)
1664 Exim now checks that the per_* options are used with an update mode that
1665 makes sense for the current ACL. For example, when Exim is processing a
1666 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1667 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1668 must specify per_mail/readonly. If you omit the update mode it defaults to
1669 /leaky where that makes sense (as before) or /readonly where required.
1671 The /noupdate option is now undocumented but still supported for
1672 backwards compatibility. It is equivalent to /readonly except that in
1673 ACLs where /readonly is required you may specify /leaky/noupdate or
1674 /strict/noupdate which are treated the same as /readonly.
1676 A useful new feature is the /count= option. This is a generalization
1677 of the per_byte option, so that you can measure the throughput of other
1678 aggregate values. For example, the per_byte option is now equivalent
1679 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1681 The per_rcpt option has been generalized using the /count= mechanism
1682 (though it's more complicated than the per_byte equivalence). When it is
1683 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1684 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1685 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1686 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1687 Note that using per_rcpt with a non-readonly update mode in more than
1688 one ACL will cause the recipients to be double-counted. (The per_mail
1689 and per_byte options don't have this problem.)
1691 The handling of very low rates has changed slightly. If the computed rate
1692 is less than the event's count (usually one) then this event is the first
1693 after a long gap. In this case the rate is set to the same as this event's
1694 count, so that the first message of a spam run is counted properly.
1696 The major new feature is a mechanism for counting the rate of unique
1697 events. The new per_addr option counts the number of different
1698 recipients that someone has sent messages to in the last time period. It
1699 behaves like per_rcpt if all the recipient addresses are different, but
1700 duplicate recipient addresses do not increase the measured rate. Like
1701 the /count= option this is a general mechanism, so the per_addr option
1702 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1703 example, measure the rate that a client uses different sender addresses
1704 with the options per_mail/unique=$sender_address. There are further
1705 details in the main documentation.
1707 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1709 TF/06 Removed a few PCRE remnants.
1711 TF/07 Automatically extract Exim's version number from tags in the git
1712 repository when doing development or release builds.
1714 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1715 Bugzilla 879. Patch from Paul Fisher.
1717 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1718 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1721 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1723 PP/05 Variable $av_failed, true if the AV scanner deferred.
1724 Bugzilla 1078. Patch from John Horne.
1726 PP/06 Stop make process more reliably on build failure.
1727 Bugzilla 1087. Patch from Heiko Schlittermann.
1729 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1730 Bugzilla 1089. Patch from Heiko Schlittermann.
1732 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1733 Bugzilla 1131. Patch from Holger Weiß.
1735 PP/09 Handle IPv6 addresses with SPF.
1736 Bugzilla 860. Patch from Wolfgang Breyha.
1738 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1740 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1743 PP/11 match_* no longer expand right-hand-side by default.
1744 New compile-time build option, EXPAND_LISTMATCH_RHS.
1745 New expansion conditions, "inlist", "inlisti".
1747 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1749 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1751 PP/14 fix log_write() format string regression from TF/03.
1752 Bugzilla 1152. Patch from Dmitry Isaikin.
1758 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1760 PP/02 Harmonised TLS library version reporting; only show if debugging.
1761 Layout now matches that introduced for other libraries in 4.74 PP/03.
1763 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1765 PP/04 New "dns_use_edns0" global option.
1767 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1770 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1771 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1773 TK/01 Updated PolarSSL code to 0.14.2.
1774 Bugzilla 1097. Patch from Andreas Metzler.
1776 PP/07 Catch divide-by-zero in ${eval:...}.
1777 Fixes bugzilla 1102.
1779 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1782 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1783 format-string attack -- SECURITY: remote arbitrary code execution.
1785 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1786 time unintentionally subject to list matching rules, letting the header
1787 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1788 arbitrary string expansion). This allowed for information disclosure.
1790 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1791 INT_MIN/-1 -- value coerced to INT_MAX.
1797 NM/01 Workaround for PCRE version dependency in version reporting
1800 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1801 This fixes portability to compilers other than gcc, notably
1802 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1804 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1805 makefiles for portability to HP-UX and POSIX correctness.
1807 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1808 Also via indented variable definition in the Makefile.
1809 (Debugging by Oliver Heesakkers).
1811 PP/02 Restore caching of spamd results with expanded spamd_address.
1812 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1814 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1815 Improves build reliability. Fix from: Frank Elsner
1817 NM/02 Fix wide character breakage in the rfc2047 coding
1818 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1820 NM/03 Allow underscore in dnslist lookups
1821 Fixes bug 1026. Patch from Graeme Fowler
1823 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1824 Code patches from Adam Ciarcinski of NetBSD.
1826 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1829 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1830 is logged, to avoid truncation. Patch from John Horne.
1832 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1833 Patch from Jakob Hirsch.
1835 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1836 SQL string expansion failure details.
1837 Patch from Andrey Oktyabrski.
1839 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1840 Patch from Simon Arlott.
1842 PP/09 New lookups functionality failed to compile on old gcc which rejects
1843 extern declarations in function scope.
1844 Patch from Oliver Fleischmann
1846 PP/10 Use sig_atomic_t for flags set from signal handlers.
1847 Check getgroups() return and improve debugging.
1848 Fixed developed for diagnosis in bug 927 (which turned out to be
1851 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1852 Patch from Mark Zealey.
1854 PP/12 Bugzilla 1056: Improved spamd server selection.
1855 Patch from Mark Zealey.
1857 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1858 Based on patch from Heiko Schlittermann.
1860 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1861 Patch from Uwe Doering, sign-off by Michael Haardt.
1863 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1864 variable declaration deep within a block. Bug and patch from
1867 PP/15 lookups-Makefile IRIX compatibility coercion.
1869 PP/16 Make DISABLE_DKIM build knob functional.
1871 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1872 Patch by Simon Arlott
1874 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1875 variable argument macros. Our copy now differs from upstream.
1881 TF/01 Failure to get a lock on a hints database can have serious
1882 consequences so log it to the panic log.
1884 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1885 controlled using the smtp_confirmation log selector.
1887 TF/03 Include the error message when we fail to unlink a spool file.
1889 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1890 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1891 for maintaining out-of-tree patches for some time.
1893 PP/01 Bugzilla 139: Documentation and portability issues.
1894 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1895 Handle per-OS dynamic-module compilation flags.
1897 PP/02 Let /dev/null have normal permissions.
1898 The 4.73 fixes were a little too stringent and complained about the
1899 permissions on /dev/null. Exempt it from some checks.
1900 Reported by Andreas M. Kirchwitz.
1902 PP/03 Report version information for many libraries, including
1903 Exim version information for dynamically loaded libraries. Created
1904 version.h, now support a version extension string for distributors
1905 who patch heavily. Dynamic module ABI change.
1907 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1908 privilege escalation vulnerability whereby the Exim run-time user
1909 can cause root to append content of the attacker's choosing to
1912 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1915 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1916 If dropping privileges for untrusted macros, we disabled normal logging
1917 on the basis that it would fail; for the Exim run-time user, this is not
1918 the case, and it resulted in successful deliveries going unlogged.
1919 Fixed. Reported by Andreas Metzler.
1925 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1926 only prepend for the Resent-* case. Fixes regression introduced in
1927 Exim 4.70 by NM/22 for Bugzilla 607.
1929 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1930 increasing numbers of administrators be bitten by this.
1932 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1934 PP/03 Bugzilla 994: added openssl_options main configuration option.
1936 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1938 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1940 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1942 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1943 without a peer certificate, leading to a segfault because of an
1944 assumption that peers always have certificates. Be a little more
1945 paranoid. Problem reported by Martin Tscholak.
1947 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1948 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1949 NB: ClamAV planning to remove STREAM in "middle of 2010".
1950 CL also introduces -bmalware, various -d+acl logging additions and
1951 more caution in buffer sizes.
1953 PP/09 Implemented reverse_ip expansion operator.
1955 PP/10 Bugzilla 937: provide a "debug" ACL control.
1957 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1959 PP/12 Bugzilla 973: Implement --version.
1961 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1963 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1965 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1967 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1968 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1969 ignore trailing whitespace.
1971 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1973 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1976 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1977 Notification from Dr Andrew Aitchison.
1979 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1980 ExtendedDetectionInfo response format.
1981 Notification from John Horne.
1983 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1986 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1987 XSL and documented dependency on system catalogs, with examples of how
1990 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1993 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1994 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1995 configuration file which is writeable by the Exim user or group.
1997 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1998 of configuration files to cover files specified with the -C option if
1999 they are going to be used with root privileges, not just the default
2002 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2003 option (effectively making it always true).
2005 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2006 files to be used while preserving root privileges.
2008 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2009 that rogue child processes cannot use them.
2011 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2012 run-time user, instead of root.
2014 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2015 Exim run-time user without dropping privileges.
2017 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2018 result string, instead of calling string_vformat() twice with the same
2021 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2022 for other users. Others should always drop root privileges if they use
2023 -C on the command line, even for a whitelisted configure file.
2025 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2027 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2033 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2034 $data_path, and $header_path variables; fixed documentation bugs and
2037 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2038 exipick to access non-standard spools, including the "frozen" queue
2041 NM/01 Bugzilla 965: Support mysql stored procedures.
2042 Patch from Alain Williams
2044 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2046 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2047 Patch from Andreas Metzler
2049 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2050 Patch from Kirill Miazine
2052 NM/05 Bugzilla 671: Added umask to procmail example.
2054 JJ/03 installed exipick 20100323.0, fixing doc bug
2056 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2057 directory. Notification and patch from Dan Rosenberg.
2059 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2061 TK/02 Improve log output when DKIM signing operation fails.
2063 MH/01 Treat the transport option dkim_domain as a colon separated
2064 list, not as a single string, and sign the message with each element,
2065 omitting multiple occurences of the same signer.
2067 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2068 Bugzilla 985, 986. Patch by Simon Arlott
2070 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2071 Patch by Simon Arlott
2073 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2074 MBX locking. Notification from Dan Rosenberg.
2080 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2082 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2084 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2086 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2088 NM/04 Bugzilla 915: Flag broken perl installation during build.
2094 TK/01 Added patch by Johannes Berg that expands the main option
2095 "spamd_address" if it starts with a dollar sign.
2097 TK/02 Write list of recipients to X-Envelope-Sender header when building
2098 the mbox-format spool file for content scanning (suggested by Jakob
2101 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2102 (http://www.dcc-servers.net/) support via dccifd. Activated by
2103 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2105 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2106 by Mark Daniel Reidel <mr@df.eu>.
2108 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2109 When building exim an external PCRE library is now needed -
2110 PCRE is a system library on the majority of modern systems.
2111 See entry on PCRE_LIBS in EDITME file.
2113 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2114 conversation. Added nologin parameter to request.
2115 Patch contributed by Kirill Miazine.
2117 TF/01 Do not log submission mode rewrites if they do not change the address.
2119 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2121 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2122 log files in place. Contributed by Roberto Lima.
2124 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2126 TF/03 Bugzilla 615: When checking the local_parts router precondition
2127 after a local_part_suffix or local_part_prefix option, Exim now
2128 does not use the address's named list lookup cache, since this
2129 contains cached lookups for the whole local part.
2131 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2132 Robert Millan. Documentation is in experimental-spec.txt.
2134 TF/04 Bugzilla 668: Fix parallel build (make -j).
2136 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2138 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2139 Patch provided by Jan Srzednicki.
2141 TF/05 Leading white space used to be stripped from $spam_report which
2142 wrecked the formatting. Now it is preserved.
2144 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2145 that they are available at delivery time.
2147 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2149 TF/08 TLS error reporting now respects the incoming_interface and
2150 incoming_port log selectors.
2152 TF/09 Produce a more useful error message if an SMTP transport's hosts
2153 setting expands to an empty string.
2155 NM/06 Bugzilla 744: EXPN did not work under TLS.
2156 Patch provided by Phil Pennock.
2158 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2159 Patch provided by Richard Godbee.
2161 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2162 acl_smtp_notquit, added index entry.
2164 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2165 Patch provided by Eugene Bujak.
2167 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2168 accept(). Patch provided by Maxim Dounin.
2170 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2171 Patch provided by Phil Pennock.
2173 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2175 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2176 Patch provided by Brad "anomie" Jorsch.
2178 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2179 Patch provided by Dean Brooks.
2181 TK/05 Add native DKIM support (does not depend on external libraries).
2183 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2184 Patch provided by Graeme Fowler.
2186 NM/16 Bugzilla 851: Documentation example syntax fix.
2188 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2190 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2193 NM/19 Bugzilla 745: TLS version reporting.
2194 Patch provided by Phil Pennock.
2196 NM/20 Bugzilla 167: bool: condition support.
2197 Patch provided by Phil Pennock.
2199 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2200 clients. Patch provided by Phil Pennock.
2202 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2203 Patch provided by Brad "anomie" Jorsch.
2205 NM/23 Bugzilla 687: Fix misparses in eximstats.
2206 Patch provided by Heiko Schlittermann.
2208 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2209 Patch provided by Heiko Schlittermann.
2211 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2212 plus update to original patch.
2214 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2216 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2217 Patch provided by David Brownlee.
2219 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2221 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2223 NM/30 Bugzilla 888: TLS documentation bugfixes.
2225 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2227 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2228 Unlike the original bugzilla I have changed all shell scripts in src tree.
2230 NM/33 Bugzilla 898: Transport filter timeout fix.
2231 Patch by Todd Rinaldo.
2233 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2234 Patch by Serge Demonchaux.
2236 NM/35 Bugzilla 39: Base64 decode bug fixes.
2237 Patch by Jakob Hirsch.
2239 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2241 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2243 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2245 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2251 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2252 ALT-N's libdkim that I have put here:
2253 http://duncanthrax.net/exim-experimental/
2255 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2256 were called 'true' and it seems that C99 defines that as a reserved
2257 keyword to be used with 'bool' variable types. That means you could
2258 not include C99-style headers which use bools without triggering
2259 build errors in sieve.c.
2261 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2262 as mailq or other aliases. Changed the --help handling significantly
2263 to do whats expected. exim_usage() emits usage/help information.
2265 SC/01 Added the -bylocaldomain option to eximstats.
2267 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2269 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2271 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2277 PH/01 Another patch from the Sieve maintainer.
2279 PH/02 When an IPv6 address is converted to a string for single-key lookup
2280 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2281 dots are used instead of colons so that keys in lsearch files need not
2282 contain colons. This was done some time before quoting was made available
2283 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2284 (notated using the quote facility) so as to distinguish them from IPv4
2285 keys. This meant that lookups for IP addresses in host lists did not work
2286 for iplsearch lookups.
2288 This has been fixed by arranging for IPv6 addresses to be expressed with
2289 colons if the lookup type is iplsearch. This is not incompatible, because
2290 previously such lookups could never work.
2292 The situation is now rather anomalous, since one *can* have colons in
2293 ordinary lsearch keys. However, making the change in all cases is
2294 incompatible and would probably break a number of configurations.
2296 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2299 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2300 conversion specification without a maximum field width, thereby enabling
2301 a rogue spamd server to cause a buffer overflow. While nobody in their
2302 right mind would setup Exim to query an untrusted spamd server, an
2303 attacker that gains access to a server running spamd could potentially
2304 exploit this vulnerability to run arbitrary code as the Exim user.
2306 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2307 $primary_hostname instead of what libspf2 thinks the hosts name is.
2309 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2310 a directory entry by the name of the lookup key. Previously, if a
2311 symlink pointed to a non-existing file or a file in a directory that
2312 Exim lacked permissions to read, a lookup for a key matching that
2313 symlink would fail. Now it is enough that a matching directory entry
2314 exists, symlink or not. (Bugzilla 503.)
2316 PH/03 The body_linecount and body_zerocount variables are now exported in the
2319 PH/04 Added the $dnslist_matched variable.
2321 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2322 This means they are set thereafter only if the connection becomes
2325 PH/06 Added the client_condition to authenticators so that some can be skipped
2326 by clients under certain conditions.
2328 PH/07 The error message for a badly-placed control=no_multiline_responses left
2329 "_responses" off the end of the name.
2331 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2333 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2334 (without spaces) instead of just copying the configuration text.
2336 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2338 PH/11 Added $max_received_linelength.
2340 PH/12 Added +ignore_defer and +include_defer to host lists.
2342 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2343 way in which PCRE > 7.0 is built.
2345 PH/14 Implemented queue_only_load_latch.
2347 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2348 MAIL command. The effect was to mangle the value on 64-bit systems.
2350 PH/16 Another patch from the Sieve maintainer.
2352 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2354 PH/18 If a system quota error occurred while trying to create the file for
2355 a maildir delivery, the message "Mailbox is full" was not appended to the
2356 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2357 only to a quota excession during the actual writing of the file.
2359 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2360 characters?) which causes problems in log lines. The DN values are now
2361 passed through string_printing() before being added to log lines.
2363 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2364 and InterBase are left for another time.)
2366 PH/21 Added message_body_newlines option.
2368 PH/22 Guard against possible overflow in moan_check_errorcopy().
2370 PH/23 POSIX allows open() to be a macro; guard against that.
2372 PH/24 If the recipient of an error message contained an @ in the local part
2373 (suitably quoted, of course), incorrect values were put in $domain and
2374 $local_part during the evaluation of errors_copy.
2380 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2381 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2384 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2385 issue a MAIL command.
2387 PH/02 In an ACL statement such as
2389 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2391 if a client was not listed at all, or was listed with a value other than
2392 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2393 the condition was not true (as it should be), so access was not denied.
2394 The bug was that the ! inversion was incorrectly passed on to the second
2395 item. This has been fixed.
2397 PH/03 Added additional dnslists conditions == and =& which are different from
2398 = and & when the dns lookup returns more than one IP address.
2400 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2401 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2403 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2404 FSYNC, which compiles an option called disable_fsync that allows for
2405 bypassing fsync(). The documentation is heavily laced with warnings.
2407 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2409 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2410 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2411 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2412 including adding "make clean"; (3) Added -fPIC when compiling the test
2413 dynamically loaded module, to get rid of a warning.
2415 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2416 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2417 = 0s. The bug is otherwise harmless.
2419 PH/07 There was a bug in the dovecot authenticator such that the value of
2420 $auth1 could be overwritten, and so not correctly preserved, after a
2421 successful authentication. This usually meant that the value preserved by
2422 the server_setid option was incorrect.
2424 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2426 PH/09 Installed PCRE release 7.0.
2428 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2429 run for batched SMTP input. It is now run at the start of every message
2430 in the batch. While fixing this I discovered that the process information
2431 (output by running exiwhat) was not always getting set for -bs and -bS
2432 input. This is fixed, and it now also says "batched" for BSMTP.
2434 PH/11 Added control=no_pipelining.
2436 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2437 patch, slightly modified), and move the expansion of helo_data till after
2438 the connection is made in the smtp transport (so it can use these
2441 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2443 PH/14 Added log_selector = +pid.
2445 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2447 PH/16 Add ${if forany and ${if forall.
2449 PH/17 Added dsn_from option to vary the From: line in DSNs.
2451 PH/18 Flush SMTP output before performing a callout, unless control =
2452 no_callout_flush is set.
2454 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2455 was true (the default) a successful delivery failed to delete the retry
2456 item, thus causing premature timeout of the address. The bug is now
2459 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2461 PH/21 Long custom messages for fakedefer and fakereject are now split up
2462 into multiline responses in the same way that messages for "deny" and
2463 other ACL rejections are.
2465 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2466 with slight modification.
2468 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2469 draft, changing the syntax and factoring some duplicate code".
2471 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2472 for deliveries of the second and subsequent messages over the same SMTP
2475 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2476 ${reduce, with only minor "tidies".
2478 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2480 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2481 expansion side effects.
2483 PH/27 When a message times out after an over-quota error from an Exim-imposed
2484 quota, the bounce message says "mailbox is full". This message was not
2485 being given when it was a system quota that was exceeded. It now should
2488 MH/03 Made $recipients available in local_scan(). local_scan() already has
2489 better access to the recipient list through recipients_list[], but
2490 $recipients can be useful in postmaster-provided expansion strings.
2492 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2493 in the case of a MAIL command with additional options following the
2494 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2495 were accidentally chopped off.
2497 PH/29 SMTP synchronization checks are implemented when a command is read -
2498 there is a check that no more input is waiting when there shouldn't be
2499 any. However, for some commands, a delay in an ACL can mean that it is
2500 some time before the response is written. In this time, more input might
2501 arrive, invalidly. So now there are extra checks after an ACL has run for
2502 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2503 pipelining has not been advertised.
2505 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2507 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2508 returned in the SMTP response when smtp_return_error_details was false.
2509 This has been fixed.
2511 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2512 instead of the C I/O that was originally supplied, because problems were
2513 reported on Solaris.
2515 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2516 Exim which did not show up earlier: it was assuming that a call to
2517 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2518 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2519 was a macro that became an assignment, so it seemed to work. This has
2520 changed to a proper function call with a void return, hence the compile
2521 error. Exim's code has been fixed.
2523 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2526 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2528 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2530 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2531 because it turns out to be too big or there is a timeout) there is no
2532 "Completed" line in the log. When some messages of this type were
2533 selected by exigrep, they were listed as "not completed". Others were
2534 picked up by some special patterns. I have improved the selection
2535 criteria to be more general.
2537 PH/38 The host_find_failed option in the manualroute router can now be set
2538 to "ignore", to completely ignore a host whose IP address cannot be
2539 found. If all hosts are ignored, the behaviour is controlled by the new
2540 host_all_ignored option.
2542 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2543 homing or because of multiple MX records with /mx) generated more than
2544 one IP address, and the following item turned out to be the local host,
2545 all the secondary addresses of the first item were incorrectly removed
2546 from the list, along with the local host and any following hosts (which
2547 is what is supposed to happen).
2549 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2550 whoever called Exim into the -H file. In the case of the daemon it was
2551 behaving confusingly. When first started, it used values for whoever
2552 started the daemon, but after a SIGHUP it used the Exim user (because it
2553 calls itself on a restart). I have changed the code so that it now always
2556 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2557 message are rejected with the same error (e.g. no authentication or bad
2558 sender address), and a DATA command is nevertheless sent (as can happen
2559 with PIPELINING or a stupid MUA), the error message that was given to the
2560 RCPT commands is included in the rejection of the DATA command. This is
2561 intended to be helpful for MUAs that show only the final error to their
2564 PH/42 Another patch from the Sieve maintainer.
2566 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2567 Eximstats - Fixed some broken HTML links and added missing column headers
2569 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2570 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2572 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2574 PH/43 Yet another patch from the Sieve maintainer.
2576 PH/44 I found a way to check for a TCP/IP connection going away before sending
2577 the response to the final '.' that terminates a message, but only in the
2578 case where the client has not sent further data following the '.'
2579 (unfortunately, this is allowed). However, in many cases there won't be
2580 any further data because there won't be any more messages to send. A call
2581 to select() can be used: if it shows that the input is "ready", there is
2582 either input waiting, or the socket has been closed. An attempt to read
2583 the next input character can distinguish the two cases. Previously, Exim
2584 would have sent an OK response which the client would never have see.
2585 This could lead to message repetition. This fix should cure that, at
2586 least in a lot of common cases.
2588 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2589 advertised in response to EHLO.
2595 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2596 fixed by 4.65/MH/01 (is this a record?) are fixed:
2598 (i) An empty string was always treated as zero by the numeric comparison
2599 operators. This behaviour has been restored.
2601 (ii) It is documented that the numeric comparison operators always treat
2602 their arguments as decimal numbers. This was broken in that numbers
2603 starting with 0 were being interpreted as octal.
2605 While fixing these problems I realized that there was another issue that
2606 hadn't been noticed. Values of message_size_limit (both the global option
2607 and the transport option) were treated as octal if they started with 0.
2608 The documentation was vague. These values are now always treated as
2609 decimal, and I will make that clear in the documentation.
2615 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2616 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2619 MH/01 Don't check that the operands of numeric comparison operators are
2620 integers when their expansion is in "skipping" mode (fixes bug
2621 introduced by 4.64-PH/07).
2623 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2624 child addresses, Exim now panics and dies. Previously, because the count
2625 is held in a short int, deliveries were likely to be lost. As such a
2626 large number of recipients for a single message is ridiculous
2627 (performance will be very, very poor), I have chosen to impose a limit
2628 rather than extend the field.
2634 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2635 leftover -K file (the existence of which was triggered by #402).
2636 While we were at it, introduced process PID as part of the -K
2637 filename. This should rule out race conditions when creating
2640 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2641 processing considerably. Previous code took too long for large mails,
2642 triggering a timeout which in turn triggers #401.
2644 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2645 in the DK code in transports.c. sendfile() is not really portable,
2646 hence the _LINUX specificness.
2648 TF/01 In the add_headers option to the mail command in an Exim filter,
2649 there was a bug that Exim would claim a syntax error in any
2650 header after the first one which had an odd number of characters
2653 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2654 callout verification, Exim cached a "reject" for the entire domain. This
2655 is correct for most verifications, but it is not correct for a recipient
2656 verification with use_sender or use_postmaster set, because in that case
2657 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2658 case of MAIL FROM:<> rejection from other early rejections (e.g.
2659 rejection of HELO). When verifying a recipient using a non-null MAIL
2660 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2661 Whatever the result of the callout, the value of the domain cache is
2662 left unchanged (for any other kind of callout, getting as far as trying
2663 RCPT means that the domain itself is ok).
2665 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2668 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2669 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2670 session, the daemon ignores ECONNECT errors and logs others; it now
2671 ignores EPIPE as well.
2673 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2674 (quoted-printable decoding).
2676 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2677 later the small subsequent patch to fix an introduced bug.
2679 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2681 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2683 PH/08 An error is now given if message_size_limit is specified negative.
2685 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2686 to be given (somewhat) arbitrary names.
2688 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2691 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2692 miscellaneous code fixes
2694 PH/10 Added the log_reject_target ACL modifier to specify where to log
2697 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2698 hostname. This is wrong, because it relates to the incoming message (and
2699 probably the interface on which it is arriving) and not to the outgoing
2700 callout (which could be using a different interface). This has been
2701 changed to use the value of the helo_data option from the smtp transport
2702 instead - this is what is used when a message is actually being sent. If
2703 there is no remote transport (possible with a router that sets up host
2704 addresses), $smtp_active_hostname is used.
2706 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2707 tweaks were necessary in order to get it to work (see also 21 below):
2708 (a) The code assumed that strncpy() returns a negative number on buffer
2709 overflow, which isn't the case. Replaced with Exim's string_format()
2711 (b) There were several signed/unsigned issues. I just did the minimum
2712 hacking in of casts. There is scope for a larger refactoring.
2713 (c) The code used strcasecmp() which is not a standard C function.
2714 Replaced with Exim's strcmpic() function.
2715 (d) The code set only $1; it now sets $auth1 as well.
2716 (e) A simple test gave the error "authentication client didn't specify
2717 service in request". It would seem that Dovecot has changed its
2718 interface. Fortunately there's a specification; I followed it and
2719 changed what the client sends and it appears to be working now.
2721 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2724 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2725 address A is aliased to B and C, where B exists and C does not. Without
2726 -v the output is "A verified" because verification stops after a
2727 successful redirection if more than one address is generated. However,
2728 with -v the child addresses are also verified. Exim was outputting "A
2729 failed to verify" and then showing the successful verification for C,
2730 with its parentage. It now outputs "B failed to verify", showing B's
2731 parentage before showing the successful verification of C.
2733 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2734 look up a TXT record in a specific list after matching in a combined
2737 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2738 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2739 they consult the DNS. I had assumed they would set it the way they
2740 wanted; and indeed my experiments on Linux seem to show that in some
2741 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2742 To be on the safe side, however, I have now made the interface to
2743 host_find_byname() similar to host_find_bydns(), with an argument
2744 containing the DNS resolver options. The host_find_byname() function now
2745 sets these options at its start, just as host_find_bydns() does. The smtp
2746 transport options dns_qualify_single and dns_search_parents are passed to
2747 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2748 of host_find_byname() use the default settings of RES_DEFNAMES
2749 (qualify_single) but not RES_DNSRCH (search_parents).
2751 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2752 spool_read_header() do less string testing, by means of a preliminary
2753 switch on the second character of optional "-foo" lines. (This is
2754 overdue, caused by the large number of possibilities that now exist.
2755 Originally there were few.) While I was there, I also converted the
2756 str(n)cmp tests so they don't re-test the leading "-" and the first
2757 character, in the hope this might squeeze out yet more improvement.
2759 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2760 flag allowing group syntax was set by the header_syntax check but not
2761 turned off, possible causing trouble later; (2) The flag was not being
2762 set at all for the header_verify test, causing "group"-style headers to
2763 be rejected. I have now set it in this case, and also caused header_
2764 verify to ignore an empty address taken from a group. While doing this, I
2765 came across some other cases where the code for allowing group syntax
2766 while scanning a header line wasn't quite right (mostly, not resetting
2767 the flag correctly in the right place). These bugs could have caused
2768 trouble for malformed header lines. I hope it is now all correct.
2770 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2771 with the "reply" argument non-NULL. The code, however (which originally
2772 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2773 but it didn't always do it. This confused somebody who was copying the
2774 code for some other use. I have removed all the tests.
2776 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2777 feature that was used to support insecure browsers during the U.S. crypto
2778 embargo. It requires special client support, and Exim is probably the
2779 only MTA that supported it -- and would never use it because real RSA is
2780 always available. This code has been removed, because it had the bad
2781 effect of slowing Exim down by computing (never used) parameters for the
2782 RSA_EXPORT functionality.
2784 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2785 authenticator to fail if there's a tab character in the incoming data
2786 (there should never be unless someone is messing about, as it's supposed
2787 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2788 if the connection is using TLS or if the remote IP is the same as the
2789 local IP, and the "valid-client-cert option" if a client certificate has
2792 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2793 authenticators. This can be used for authorization after authentication
2794 succeeds. (In the case of plaintext, it servers for both authentication
2797 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2798 if any retry times were supplied.
2800 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2801 connection, where there is no client IP address to check. In this
2802 situation, the verify now always succeeds.
2804 PH/25 Applied John Jetmore's -Mset patch.
2806 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2808 PH/27 In a string expansion for a processed (not raw) header when multiple
2809 headers of the same name were present, leading whitespace was being
2810 removed from all of them, but trailing whitespace was being removed only
2811 from the last one. Now trailing whitespace is removed from each header
2812 before concatenation. Completely empty headers in a concatenation (as
2813 before) are ignored.
2815 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2816 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2818 PH/29 [Removed. This was a change that I later backed out, and forgot to
2819 correct the ChangeLog entry (that I had efficiently created) before
2820 committing the later change.]
2822 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2823 address errors (4xx response to RCPT) over the same connection as other
2824 messages routed to the same hosts. Such deliveries are always "forced",
2825 so retry times are not inspected. This resulted in far too many retries
2826 for the affected addresses. The effect occurred only when there were more
2827 hosts than the hosts_max_try setting in the smtp transport when it had
2828 the 4xx errors. Those hosts that it had tried were not added to the list
2829 of hosts for which the message was waiting, so if all were tried, there
2830 was no problem. Two fixes have been applied:
2832 (i) If there are any address or message errors in an SMTP delivery, none
2833 of the hosts (tried or untried) are now added to the list of hosts
2834 for which the message is waiting, so the message should not be a
2835 candidate for sending over the same connection that was used for a
2836 successful delivery of some other message. This seems entirely
2837 reasonable: after all the message is NOT "waiting for some host".
2838 This is so "obvious" that I'm not sure why it wasn't done
2839 previously. Hope I haven't missed anything, but it can't do any
2840 harm, as the worst effect is to miss an optimization.
2842 (ii) If, despite (i), such a delivery is accidentally attempted, the
2843 routing retry time is respected, so at least it doesn't keep
2844 hammering the server.
2846 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2847 in ${readsocket because some servers need this prod.
2849 PH/32 Added some extra debug output when updating a wait-xxx database.
2851 PH/33 The hint "could be header name not terminated by colon", which has been
2852 given for certain expansion errors for a long time, was not being given
2853 for the ${if def:h_colon_omitted{... case.
2855 PH/34 The spec says: "With one important exception, whenever a domain list is
2856 being scanned, $domain contains the subject domain." There was at least
2857 one case where this was not true.
2859 PH/35 The error "getsockname() failed: connection reset by peer" was being
2860 written to the panic log as well as the main log, but it isn't really
2861 panic-worthy as it just means the connection died rather early on. I have
2862 removed the panic log writing for the ECONNRESET error when getsockname()
2865 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2866 runs only) independently of the message's sender address. This meant
2867 that, if the 4xx error was in fact related to the sender, a different
2868 message to the same recipient with a different sender could confuse
2869 things. In particular, this can happen when sending to a greylisting
2870 server, but other circumstances could also provoke similar problems.
2871 I have changed the default so that the retry time for these errors is now
2872 based a combination of the sender and recipient addresses. This change
2873 can be overridden by setting address_retry_include_sender=false in the
2876 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2877 remote server are returned as part of bounce messages. This was not
2878 happening for LMTP over a pipe (the lmtp transport), but now it is the
2879 same for both kinds of LMTP.
2881 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2882 in header lines that were in fact CNAMEs. This is no longer the case.
2884 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2885 and queue runs started by the daemon processed all messages. This has
2886 been fixed so that -R and -S can now usefully be given with -q<time>.
2888 PH/40 Import PCRE release 6.7 (fixes some bugs).
2890 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2892 PH/42 Give an error if -q is specified more than once.
2894 PH/43 Renamed the variables $interface_address and $interface_port as
2895 $received_ip_address and $received_port, to make it clear that these
2896 values apply to message reception, and not to the outgoing interface when
2897 a message is delivered. (The old names remain recognized, of course.)
2899 PH/44 There was no timeout on the connect() call when using a Unix domain
2900 socket in the ${readsocket expansion. There now is.
2902 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2903 be meaningful with "accept".
2905 SC/01 Eximstats V1.43
2906 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2908 SC/02 Eximstats V1.44
2909 Use a glob alias rather than an array ref in the generated
2910 parser. This improves both readability and performance.
2912 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2913 Collect SpamAssassin and rejection statistics.
2914 Don't display local sender or destination tables unless
2915 there is data to show.
2916 Added average volumes into the top table text output.
2918 SC/04 Eximstats V1.46
2919 Collect data on the number of addresses (recipients)
2920 as well as the number of messages.
2922 SC/05 Eximstats V1.47
2923 Added 'Message too big' to the list of mail rejection
2924 reasons (thanks to Marco Gaiarin).
2926 SC/06 Eximstats V1.48
2927 Mainlog lines which have GMT offsets and are too short to
2928 have a flag are now skipped.
2930 SC/07 Eximstats V1.49 (Alain Williams)
2931 Added the -emptyok flag.
2933 SC/08 Eximstats V1.50
2934 Fixes for obtaining the IP address from reject messages.
2936 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2937 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2938 whitespace changes from 4.64-PH/27
2940 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2943 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2944 are found, allow negative numbers in numeric criteria)
2946 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2948 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2949 to match changes made in 4.64-PH/43
2951 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2953 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2954 "domain" is actually an IP address, and force a failure. This locks out
2955 those revolvers/nameservers that support "A-for-A" lookups, in
2956 contravention of the specifications.
2958 PH/48 When a host name was looked up from an IP address, and the subsequent
2959 forward lookup of the name timed out, the host name was left in
2960 $sender_host_name, contrary to the specification.
2962 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2963 restricted to single-key lookups, Exim was not diagnosing an error if
2964 * or *@ was used with a query-style lookup.
2966 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2968 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2969 long ago, but noone interested enough thought of it. Let's just say that
2970 the "1.1" means that there are some new functions that weren't there at
2971 some point in the past.
2973 PH/51 Error processing for expansion failure of helo_data from an smtp
2974 transport during callout processing was broken.
2976 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2977 tested/used via the -bh/-bhc/-bs options.
2979 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2980 bug, fixed in subsequent PCRE releases).
2982 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2983 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2985 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2991 SC/01 Use a glob alias rather than an array ref in eximstats generated
2992 parser. This improves both readability and performance.
2994 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2995 Don't display local sender or destination tables in eximstats unless
2996 there is data to show.
2997 Added average volumes into the eximstats top table text output.
2999 SC/03 Collect data on the number of addresses (recipients) as well
3000 as the number of messages in eximstats.
3002 TF/01 Correct an error in the documentation for the redirect router. Exim
3003 does (usually) call initgroups() when daemonizing.
3005 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3006 with consistent privilege compared to when running as a daemon.
3008 TF/03 Note in the spec that $authenticated_id is not set for local
3009 submissions from trusted users.
3011 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3012 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3014 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3015 by adding some example configuration directives to the default
3016 configuration file. A little bit of work is required to uncomment the
3017 directives and define how usernames and passwords are checked, but
3018 there is now a framework to start from.
3020 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3021 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3022 without this. I don't know how relevant this is to other LDAP libraries.
3024 PH/02 Add the verb name to the "unknown ACL verb" error.
3026 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3028 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3030 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3031 directory not expanded when it should be if an expanded home directory
3032 was set for the address (which is overridden by the transport).
3034 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3037 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3038 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3039 because it is too late at that time, and has no effect.
3041 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3042 security issue with \' (bugzilla #107). I could not use the
3043 PQescapeStringConn() function, because it needs a PGconn value as one of
3046 PH/08 When testing addresses using -bt, indicate those final addresses that
3047 are duplicates that would not cause an additional delivery. At least one
3048 person was confused, thinking that -bt output corresponded to deliveries.
3049 (Suppressing duplicates isn't a good idea as you lose the information
3050 about possibly different redirections that led to the duplicates.)
3052 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3053 systems where poll() doesn't work, in particular OS X.
3055 PH/10 Added more information to debugging output for retry time not reached.
3057 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3058 operations in malware.c.
3060 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3063 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3064 syslog_duplication set false, log lines that would normally be written
3065 both the the main log and to the reject log were not written to syslog at
3068 PH/14 In the default configuration, change the use of "message" in ACL warn
3069 statements to "add_header".
3071 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3072 not followed by a command (e.g. "seen endif").
3074 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3075 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3078 PH/17 Added extra conditions to the default value of delay_warning_condition
3082 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3083 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3084 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3087 The Auto-Submitted: and various List- headers are standardised, whereas I
3088 don't think Precedence: ever was.
3090 PH/18 Refactored debugging code in route_finduser() to show more information,
3091 in particular, the error code if getpwnam() issues one.
3093 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3094 This is apparently needed in addition to the PH/07 change above to avoid
3095 any possible encoding problems.
3097 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3098 but not after initializing Perl.
3100 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3101 output them only if debugging. By default they are written stderr,
3102 apparently, which is not desirable.
3104 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3107 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3110 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3112 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3113 authenticated or an ident call has been made. Suppress the default
3114 values for $authenticated_id and $authenticated_sender (but permit -oMai
3115 and -oMas) when testing with -bh.
3117 PH/24 Re-jigged the order of the tests in the default configuration so that the
3118 tests for valid domains and recipients precede the DNS black list and CSA
3119 tests, on the grounds that those ones are more expensive.
3121 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3122 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3123 This bug exists in every version of Exim that I still have, right back to
3126 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3127 However, an attempt to turn on case-sensitivity in a regex key by
3128 including (?-i) didn't work because the subject string was already
3129 lowercased, and the effects were non-intuitive. It turns out that a
3130 one-line patch can be used to allow (?-i) to work as expected.
3136 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3137 other effects) broke the use of negated acl sub-conditions.
3139 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3142 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3143 "Deny" causes Exim to reject the incoming connection with a 554 error.
3144 Unfortunately, if there is a major crisis, such as a disk failure,
3145 tcp-wrappers gives "deny", whereas what one would like would be some
3146 kind of temporary error. A kludge has been added to help with this.
3147 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3148 554 error is used if errno is still zero or contains ENOENT (which occurs
3149 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3152 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3154 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3155 errors. Otherwise a message that provokes a temporary error (when other
3156 messages do not) can cause a whole host to time out.
3158 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3159 addresses were routed directly to files or pipes from a redirect router.
3160 File deliveries just didn't batch; pipe deliveries might have suffered
3163 PH/06 A failure to get a lock for a hints database would erroneously always say
3164 "Failed to get write lock", even when it was really a read lock.
3166 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3167 of 0600. This has been changed to use the value of the lockfile_mode
3168 option (which defaults to 0600).
3170 PH/08 Applied small patch from the Sieve maintainer.
3172 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3173 folder from quota calculations, a direct delivery into this folder messed
3174 up the contents of the maildirsize file. This was because the regex was
3175 used only to exclude .Trash (or whatever) when the size of the mailbox
3176 was calculated. There was no check that a delivery was happening into an
3177 excluded directory. This bug has been fixed by ignoring all quota
3178 processing for deliveries into excluded directories.
3180 PH/10 Added the maildirfolder_create_regex option to appendfile.
3186 PH/01 The code for finding all the local interface addresses on a FreeBSD
3187 system running IPv6 was broken. This may well have applied to all BSD
3188 systems, as well as to others that have similar system calls. The broken
3189 code found IPv4 interfaces correctly, but gave incorrect values for the
3190 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3191 that it would not match correctly against @[] and not recognize the IPv6
3194 PH/02 The ipliteral router was not recognizing addresses of the form user@
3195 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3197 PH/03 Added disable_ipv6.
3199 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3200 decoded form, because it is most often used to construct To: headers
3201 lines in autoreplies, and the decoded form may well be syntactically
3202 invalid. However, $reply_address has leading white space removed, and all
3203 newlines turned into spaces so that the autoreply transport does not
3206 PH/05 If group was specified without a user on a router, and no group or user
3207 was specified on a transport, the group from the router was ignored.
3209 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3210 for visible compile-time settings that can be used to change these
3211 numbers, for those that want even more. Backwards compatibility with old
3212 spool files has been maintained. However, going back to a previous Exim
3213 release will lost any variables that are in spool files.
3215 PH/07 Two small changes when running in the test harness: increase delay when
3216 passing a TCP/IP connection to a new process, in case the original
3217 process has to generate a bounce, and remove special handling of
3218 127.0.0.2 (sic), which is no longer necessary.
3220 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3221 be the same on different OS.
3223 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3226 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3227 whether --show-vars was specified or not
3229 JJ/02 exipick: Added support for new ACL variable spool format introduced
3232 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3233 syntactically invalid From: or Reply-to: line, and a filter used this to
3234 generate an autoreply, and therefore failed to obtain an address for the
3235 autoreply, Exim could try to deliver to a non-existent relative file
3236 name, causing unrelated and misleading errors. What now happens is that
3237 it logs this as a hard delivery error, but does not attempt to create a
3240 PH/11 The exinext utility has a -C option for testing purposes, but although
3241 the given file was scanned by exinext itself; it wasn't being passed on
3242 when Exim was called.
3244 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3245 an end-of-file indication when reading a command response.
3247 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3248 compiled. In many other places in Exim, IPv6 addresses are always
3249 recognized, so I have changed this. It also means that IPv4 domain
3250 literals of the form [IPV4:n.n.n.n] are now always recognized.
3252 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3253 used if the router is not running as root, for example, when verifying at
3254 ACL time, or when using -bh. The debugging output from this situation was
3255 non-existent - all you got was a failure to exec. I have made two
3258 (a) Failures to set uid/gid, the current directory, or a process leader
3259 in a subprocess such as that created by queryprogram now generate
3260 suitable debugging output when -d is set.
3262 (b) The queryprogram router detects when it is not running as root,
3263 outputs suitable debugging information if -d is set, and then runs
3264 the subprocess without attempting to change uid/gid.
3266 PH/15 Minor change to Makefile for building test_host (undocumented testing
3269 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3270 additional section of a DNS packet that returns MX or SRV records.
3271 Instead, it always explicitly searches for A/AAAA records. This avoids
3272 major problems that occur when a DNS server includes only records of one
3273 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3274 fixed another bug: if SRV records were looked up and the corresponding
3275 address records were *not* found in the additional section, the port
3276 values from the SRV records were lost.
3278 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3279 using the correct key (the original address) when searching the retry
3280 rules in order to find which one to use for generating the retry hint.
3282 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3283 adding the default one. Similarly, if it contains a Reply-To: header, the
3284 errors_reply_to option, if set, is not used.
3286 PH/19 When calculating a retry time, Exim used to measure the "time since
3287 failure" by looking at the "first failed" field in the retry record. Now
3288 it does not use this if it is later than than the arrival time of the
3289 message. Instead it uses the arrival time. This makes for better
3290 behaviour in cases where some deliveries succeed, thus re-setting the
3291 "first failed" field. An example is a quota failure for a huge message
3292 when small messages continue to be delivered. Without this change, the
3293 "time since failure" will always be short, possible causing more frequent
3294 delivery attempts for the huge message than are intended.
3295 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3297 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3298 $1, $2, $3) because the numerical variables can be reset during some
3299 expansion items (e.g. "match"), thereby losing the authentication data.
3301 PH/21 Make -bV show the size of off_t variables so that the test suite can
3302 decide whether to run tests for quotas > 2G.
3304 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3305 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3306 filecount value is greater than 2G or if a quota value is greater than 2G
3307 on a system where the size of off_t is not greater than 4, a panic error
3310 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3311 never match. The debug and -bh output now contains an explicit error
3312 message indicating a malformed IPv4 address or mask.
3314 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3315 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3316 PH/23 above applies.
3318 PH/25 Do not write to syslog when running in the test harness. The only
3319 occasion when this arises is a failure to open the main or panic logs
3320 (for which there is an explicit test).
3322 PH/26 Added the /no_tell option to "control=freeze".
3324 PH/27 If a host name lookup failed very early in a connection, for example, if
3325 the IP address matched host_lookup and the reverse lookup yielded a name
3326 that did not have a forward lookup, an error message of the form "no IP
3327 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3328 could be logged. Now it outputs the IP address instead of "NULL".
3330 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3331 allows the sender and the authenticated sender to be set when
3332 submitting a message from within Exim. Since child_open_exim() is
3333 documented for local_scan(), the new function should be too.
3335 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3336 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3337 results in an empty string is now treated as unset.
3339 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3341 PH/31 Added sender_verify_fail logging option.
3343 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3344 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3345 tidied the source and removed it altogether.
3347 PH/33 When a queue run was abandoned because the load average was too high, a
3348 log line was always written; now it is written only if the queue_run log
3349 selector is set. In addition, the log line for abandonment now contains
3350 information about the queue run such as the pid. This is always present
3351 in "start" and "stop" lines but was omitted from the "abandon" line.
3353 PH/34 Omit spaces between a header name and the colon in the error message that
3354 is given when verify = headers_syntax fails (if there are lots of them,
3355 the message gets confusing).
3357 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3358 names, as there are now some PTR records that contain slashes. This check
3359 is only to protect against broken name servers that fall over on strange
3360 characters, so the fact that it applies to all lookups doesn't matter.
3362 PH/36 Now that the new test suite is complete, we can remove some of the
3363 special code in Exim that was needed for the old test suite. For example,
3364 sorting DNS records because real resolvers return them in an arbitrary
3365 order. The new test suite's fake resolver always returns records in the
3368 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3369 bounces) except when queue_only is set, to avoid logging races between
3370 the different processes.
3372 PH/38 Panic-die if .include specifies a non-absolute path.
3374 PH/39 A tweak to the "H" retry rule from its user.
3376 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3377 a label. They prevented compilation on older perls.
3379 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3380 a warning to be raised on newish perls.
3382 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3383 on queue. Changes to match documented behaviour of showing count of
3384 messages matching specified criteria.
3386 PH/40 Changed the default ident timeout from 30s to 5s.
3388 PH/41 Added support for the use of login_cap features, on those BSD systems
3389 that have them, for controlling the resources used by pipe deliveries.
3391 PH/42 The content-scanning code uses fopen() to create files in which to put
3392 message data. Previously it was not paying any attention to the mode of
3393 the files. Exim runs with umask(0) because the rest of the code creates
3394 files with open(), and sets the required mode explicitly. Thus, these
3395 files were ending up world-writeable. This was not a big issue, because,
3396 being within the spool directory, they were not world-accessible. I have
3397 created a function called modefopen, which takes an additional mode
3398 argument. It sets umask(777), creates the file, chmods it to the required
3399 mode, then resets the umask. All the relevant calls to fopen() in the
3400 content scanning code have been changed to use this function.
3402 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3403 to 24 hours. This avoids potential overflow problems when processing G
3404 and H retry rules. I suspect nobody ever tinkers with this value.
3406 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3408 PH/45 When the plaintext authenticator is running as a client, the server's
3409 challenges are checked to ensure they are valid base64 strings. By
3410 default, the authentication attempt is cancelled if an invalid string is
3411 received. Setting client_ignore_invalid_base64 true ignores these errors.
3412 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3413 they are received. Thus, the responses can be made to depend on the
3414 challenges. If an invalid string is ignored, an empty string is placed in
3417 PH/46 Messages that are created by the autoreply transport now contains a
3418 References: header, in accordance with RFCs 2822 and 3834.
3420 PH/47 Added authenticated_sender_force to the smtp transport.
3422 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3424 PH/49 Installed latest patch from the Sieve maintainer.
3426 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3427 was also set, the appendfile transport was unnecessarily scanning a
3428 directory of message files (e.g. for maildir delivery) to find the count
3429 of files (along with the size), even though it did not need this
3430 information. It now does the scan only if it needs to find either the
3431 size of the count of files.
3433 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3435 PH/52 Two bugs concerned with error handling when the smtp transport is
3438 (i) Exim was not creating retry information for temporary errors given
3439 for individual recipients after the DATA command when the smtp transport
3440 was used in LMTP mode. This meant that they could be retried too
3441 frequently, and not timed out correctly.
3443 (ii) Exim was setting the flag that allows error details to be returned
3444 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3445 recipients that were returned after the DATA command.
3447 PH/53 This is related to PH/52, but is more general: for any failing address,
3448 when detailed error information was permitted to be returned to the
3449 sender, but the error was temporary, then after the final timeout, only
3450 "retry timeout exceeded" was returned. Now it returns the full error as
3451 well as "retry timeout exceeded".
3453 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3454 do this, and (what is worse) MTAs that accept it.
3456 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3457 will now be deprecated.
3459 PH/56 New os.c-cygwin from the Cygwin maintainer.
3461 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3462 formats (previously only available in exim formats via -bpr, -bpru,
3463 and -bpra. Now also available in native and exiqgrep formats)
3465 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3466 with very large, slow to parse queues
3468 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3470 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3472 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3473 responses to authentication challenges, though it was showing the
3474 challenges; (ii) I've removed the CR characters from the debug output for
3477 PH/58 Allow for the insertion of a newline as well as a space when a string
3478 is turned into more than one encoded-word during RFC 2047 encoding. The
3479 Sieve code now uses this.
3481 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3482 data_4xx, lost_connection, tls_required.
3484 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3485 message was being sent as an SMTP response.
3487 PH/61 Add -l and -k options to exicyclog.
3489 PH/62 When verifying, if an address was redirected to one new address, so that
3490 verification continued, and the new address failed or deferred after
3491 having set something in $address_data, the value of $address_data was not
3492 passed back to the ACL. This was different to the case when no
3493 redirection occurred. The value is now passed back in both cases.
3495 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3496 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3497 use login_cap.h, so on its own it isn't the distinguishing feature. The
3498 new name refers directly to the setclassresources() function.
3500 PH/65 Added configuration files for NetBSD3.
3502 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3504 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3505 is preferred over IPv4.
3507 PH/68 The bounce_return_message and bounce_return_body options were not being
3508 honoured for bounces generated during the reception of non-SMTP messages.
3509 In particular, this applied to messages rejected by the ACL. This bug has
3510 been fixed. However, if bounce_return_message is true and bounce_return_
3511 body is false, the headers that are returned for a non-SMTP message
3512 include only those that have been read before the error was detected.
3513 (In the case of an ACL rejection, they have all been read.)
3515 PH/69 The HTML version of the specification is now built in a directory called
3516 spec_html instead of spec.html, because the latter looks like a path with
3517 a MIME-type, and this confuses some software.
3519 PH/70 Catch two compiler warnings in sieve.c.
3521 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3522 function verify_get_ident() calls ip_connect() to connect a socket, but
3523 if the "connect()" function timed out, ip_connect() used to close the
3524 socket. However, verify_get_ident() also closes the socket later, and in
3525 between Exim writes to the log, which may get opened at this point. When
3526 the socket was closed in ip_connect(), the log could get the same file
3527 descriptor number as the socket. This naturally causes chaos. The fix is
3528 not to close the socket in ip_connect(); the socket should be closed by
3529 the function that creates it. There was only one place in the code where
3530 this was missing, in the iplookup router, which I don't think anybody now
3531 uses, but I've fixed it anyway.
3533 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3534 well as to direct DNS lookups. Otherwise the handling of names in host
3535 lists is inconsistent and therefore confusing.
3541 PH/01 Two changes to the default runtime configuration:
3543 (1) Move the checks for relay_from_hosts and authenticated clients from
3544 after to before the (commented out) DNS black list checks.
3546 (2) Add control=submission to the relay_from_hosts and authenticated
3547 clients checks, on the grounds that messages accepted by these
3548 statements are most likely to be submissions.
3550 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3552 (1) Generate an error if the third argument for the ${prvs expansion is
3555 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3558 (3) Reset the variables that are obtained from the first argument of
3559 ${prvscheck and used in the second argument before leaving the code,
3560 because their memory is reclaimed, so using them afterwards may do
3563 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3564 one (it's much easier than Tom thought :-).
3566 (5) Because of (4), we can now allow for the use of $prvscheck_result
3567 inside the third argument.
3569 PH/03 For some reason, the default setting of PATH when running a command from
3570 a pipe transport was just "/usr/bin". I have changed it to
3573 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3574 anything to be listed in the output from -bV.
3576 PH/05 When a filter generated an autoreply, the entire To: header line was
3577 quoted in the delivery log line, like this:
3579 => >A.N.Other <ano@some.domain> <original@ddress> ...
3581 This has been changed so that it extracts the operative address. There
3582 may be more than one such address. If so, they are comma-separated, like
3585 => >ano@some.domain,ona@other.domain <original@ddress> ...
3587 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3588 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3589 not being looked up in the rDNS to get a host name, Exim was showing the
3590 IP address twice in Received: lines, even though the IP addresses were
3591 identical. For example:
3593 Received: from [1.2.3.4] (helo=[1.2.3.4])
3595 However, if the real host name was known, it was omitting the HELO data
3596 if it matched the actual IP address. This has been tidied up so that it
3597 doesn't show the same IP address twice.
3599 PH/07 When both +timestamp and +memory debugging was on, the value given by
3600 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3601 called by the debug printing, thereby overwriting the timestamp buffer.
3602 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3604 PH/08 When the original message was included in an autoreply transport, it
3605 always said "this is a copy of the message, including all the headers",
3606 even if body_only or headers_only was set. It now gives an appropriate
3609 PH/09 Applied a patch from the Sieve maintainer which:
3611 o fixes some comments
3612 o adds the (disabled) notify extension core
3613 o adds some debug output for the result of if/elsif tests
3614 o points to the current vacation draft in the documentation
3615 and documents the missing references header update
3619 o fixes a bug in processing the envelope test (when testing
3620 multiple envelope elements, the last element determined the
3623 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3624 Electronic Mail") by including:
3626 Auto-submitted: auto-generated
3628 in the messages that it generates (bounce messages and others, such as
3629 warnings). In the case of bounce messages for non-SMTP messages, there was
3630 also a typo: it was using "Auto_submitted" (underscore instead of
3631 hyphen). Since every message generated by Exim is necessarily in response
3632 to another message, thes have all been changed to:
3634 Auto-Submitted: auto-replied
3636 in accordance with these statements in the RFC:
3638 The auto-replied keyword:
3640 - SHOULD be used on messages sent in direct response to another
3641 message by an automatic process,
3643 - MUST NOT be used on manually-generated messages,
3645 - MAY be used on Delivery Status Notifications (DSNs) and Message
3646 Disposition Notifications (MDNs),
3648 - MUST NOT be used on messages generated by automatic or periodic
3649 processes, except for messages which are automatic responses to
3652 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3653 to the default Received: header definition.
3655 PH/12 Added log selector acl_warn_skipped (default on).
3657 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3658 variables because (a) they are in the wrong storage pool and (b) even if
3659 they were copied, it wouldn't work properly because of the caching.
3661 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3662 checking when decoding. Apparently there are clients that generate
3663 overlong encoded strings. Why am I not surprised?
3665 PH/15 If the first argument of "${if match_address" was not empty, but did not
3666 contain an "@" character, Exim crashed. Now it writes a panic log message
3667 and treats the condition as false.
3669 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3671 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3672 extension "envelope-auth". The code is finished and in agreement with
3673 other implementations, but there is no documentation so far and in fact,
3674 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3675 not changing the active code.
3677 Print executed "if" and "elsif" statements when debugging is used. This
3678 helps a great deal to understand what a filter does.
3680 Document more things not specified clearly in RFC3028. I had all this
3681 sorted out, when out of a sudden new issues came to my mind. Oops."
3683 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3686 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3687 canonical form (as documented). However, after a host name lookup from
3688 the IP address, check_host() was doing a simple string comparison with
3689 addresses acquired from the DNS when checking that the found name did
3690 have the original IP as one of its addresses. Since any found IPv6
3691 addresses are likely to be in abbreviated form, the comparison could
3692 fail. Luckily, there already exists a function for doing the comparison
3693 by converting both addresses to binary, so now that is used instead of
3694 the text comparison.
3696 PH/20 There was another similar case to PH/19, when a complete host name was
3697 given in a host list; looking up its IP address could give an abbreviated
3698 form, whereas the current host's name might or might not be abbreviated.
3699 The same fix has been applied.
3705 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3706 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3709 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3710 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3712 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3714 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3715 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3716 Also, for "Auto-Submitted", treat anything other than "no" as
3717 non-personal, in accordance with RFC 3834. (Previously it treated
3718 anything starting "auto-" as non-personal.)
3720 TF/01 The control=submission/name=... option had a problem with syntax
3721 errors if the name included a slash character. The /name= option
3722 now slurps the rest of the string, so it can include any characters
3723 but it must come last in the list of options (after /sender_retain
3726 PH/05 Some modifications to the interface to the fake nameserver for the new
3734 TK/01 Added the "success_on_redirect" address verification option. See
3735 NewStuff for rationale and an example.
3737 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3739 PH/02 Patch to exigrep to allow it to work on syslog lines.
3741 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3742 fread() to skip over the body file's header line, because in Cygwin the
3743 header line is locked and is inaccessible.
3745 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3746 co-exist for some time) to make it clear that it is the Exim ID that is
3747 referenced, not the Message-ID: header line.
3749 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3750 string_format() function, because snprintf() does not exist on all
3753 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3754 ${stat: expansion item.
3756 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3757 protocol synchronization error", to keep the pedants happy.
3759 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3760 well as for IRIX systems, when gcc is being used. See the host.c source
3763 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3765 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3768 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3769 transport and to the smtp transport in LMTP mode.
3771 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3773 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3774 run a filter in a subprocess. This could lead to confusion in subsequent
3775 lookups in the parent process. There should also be a search_tidyup() at
3776 the end of the subprocess.
3778 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3779 only if the host matched helo_try_verify_hosts, which caused the
3780 verification to occur when the EHLO/HELO command was issued. The ACL just
3781 tested the remembered result. Now, if a previous verification attempt has
3782 not happened, "verify = helo" does it there and then.
3784 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3786 TK/03 Fix log output including CR from clamd.
3788 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3789 exist provoked a memory error which could cause a segfault.
3791 PH/15 Installed PCRE 6.2
3793 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3795 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3796 of the problem. Specifically, suggested +O2 rather than +O1 for the
3799 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3801 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3802 record, the port was not correctly logged when the outgoing_port log
3803 selector was set (it logged the transort's default port).
3805 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3806 fallback_hosts, and "hosts" in the smtp transport.
3808 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3809 host errors such as "Connection refused".
3811 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3812 authentication with radiusclient 0.4.9:
3814 - Error returned from rc_read_config was caught wrongly
3815 - Username/password not passed on to radius server due to wrong length.
3817 The presumption is that some radiusclient API changes for 4.51/PH/17
3818 were not taken care of correctly. The code is still untested by me (my
3819 Linux distribution still has 0.3.2 of radiusclient), but it was
3820 contributed by a Radius user.
3822 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3823 expanding the "port" option of the smtp transport.
3825 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3826 while reading a MIME header. Thanks to Tom Hughes for a patch.
3828 PH/24 Include config.h inside local_scan.h so that configuration settings are
3831 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3832 that in an ACL for RCPT (for example), you can examine exactly what was
3835 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3836 commands, but it was not correctly comparing the address with the actual
3837 client host address. Thus, it would show the EHLO address in Received:
3838 header lines when this was not necessary.
3840 PH/27 Added the % operator to ${eval:}.
3842 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3843 it should be ignoring failures (because with -C, for example, it has lost
3844 privilege). It wasn't ignoring creation failures other than "already
3847 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3850 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3851 because an input line was too long, either on its own, or by virtue of
3852 too many continuations, the temporary file was not being removed, and the
3853 return code was incorrect.
3855 PH/31 Missing "BOOL" in function definition in filtertest.c.
3857 PH/32 Applied Sieve patches from the maintainer.
3859 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3861 PH/33 Added "verify = not_blind".
3863 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3864 Local/Makefile (with some defaults set). These are used in built scripts
3865 such as exicyclog, but they have never been used in the exim_install
3866 script (though there are many overriding facilities there). I have
3867 arranged that the exim_install script now takes note of these two
3870 PH/35 Installed configuration files for Dragonfly.
3872 PH/36 When a locally submitted message by a trusted user did not contain a
3873 From: header, and the sender address was obtained from -f or from an SMTP
3874 MAIL command, and the trusted user did not use -F to supply a sender
3875 name, $originator_name was incorrectly used when constructing a From:
3876 header. Furthermore, $originator_name was used for submission mode
3877 messages from external hosts without From: headers in a similar way,
3878 which is clearly wrong.
3880 PH/37 Added control=suppress_local_fixups.
3882 PH/38 When log_selector = +received_sender was set, and the addition of the
3883 sender made the log line's construction buffer exactly full, or one byte
3884 less than full, an overflow happened when the terminating "\n" was
3887 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3888 when the result of a list match is failure because a DNS lookup failed.
3890 PH/40 RM_COMMAND is now used in the building process.
3892 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3893 the "build-* directories that it finds.
3895 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3896 address in a domain literal was a prefix of an interface address.
3898 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3899 when verifying a sender address, unless rewrite_headers is false.
3901 PH/44 Wrote a long comment about why errors_to addresses are verified as
3902 recipients, not senders.
3904 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3905 the ratelimit ACL was added.
3907 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3909 PH/46 Added extra information about PostgreSQL errors to the error string.
3911 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3912 avoiding the need to install special zones in a real server. This is
3913 backwards compatible; if it can't find the fake resolver, it drops back.
3914 Thus, both old and new test suites can be run.
3916 TF/02 Added util/ratelimit.pl
3918 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3919 clock is set back in time.
3921 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3922 Candler <B.Candler@pobox.com>.
3924 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3925 OK for addresses that are the result of redirections.
3927 PH/48 A number of further additions for the benefit of the new test suite,
3928 including a fake gethostbyname() that interfaces to the fake DNS resolver
3931 TF/06 The fix for widen_domains has also been applied to qualify_single and
3932 search_parents which are the other dnslookup options that can cause
3935 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3938 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3940 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3941 service" if it encountered an MX record with an empty target hostname.
3942 The message is now "an MX or SRV record indicated no SMTP service".
3944 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3945 if the DNS of the sending site is misconfigured. This is quite a
3946 common situation. This change restores the behaviour of treating a
3947 helo verification defer as a failure.
3949 PH/51 If self=fail was set on a router, the bounce message did not include the
3950 actual error message.
3956 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3958 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3959 command itself ended in error, the underlying message about the transport
3960 filter timeout was being overwritten with the pipe command error. Now the
3961 underlying error message should be appended to the second error message.
3963 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3965 PH/02 Reduce the amount of output that "make" produces by default. Full output
3966 can still be requested.
3968 PH/03 The warning log line about a condition test deferring for a "warn" verb
3969 was being output only once per connection, rather than after each
3970 occurrence (because it was using the same function as for successful
3971 "warn" verbs). This seems wrong, so I have changed it.
3973 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3974 it should not have, which might have caused a crash in the right
3975 circumstances, but probably never did.
3977 PH/04 Installed a modified version of Tony Finch's patch to make submission
3978 mode fix the return path as well as the Sender: header line, and to
3979 add a /name= option so that you can make the user's friendly name appear
3982 TF/03 Added the control = fakedefer ACL modifier.
3984 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3985 Mark Lowes for thorough testing.
3987 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3989 TK/03 Merged latest SRS patch from Miles Wilton.
3991 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3992 with the definition in sysexits.h (which is #included earlier).
3993 Fortunately, Exim does not actually use EX_OK. The code used to try to
3994 preserve the sysexits.h value, by assuming that macro definitions were
3995 scanned for macro replacements. I have been disabused of this notion,
3996 so now the code just undefines EX_OK before #including unistd.h.
3998 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3999 in the smtp transport. When a block could not be written in a single
4000 write() function, the timeout was being re-applied to each part-write.
4001 This seems wrong - if the receiver was accepting one byte at a time it
4002 would take for ever. The timeout is now adjusted when this happens. It
4003 doesn't have to be particularly precise.
4005 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4006 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4008 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4009 without a domain if the check to <postmaster@domain> fails.
4011 SC/01 Eximstats: added -xls and the ability to specify output files
4012 (patch written by Frank Heydlauf).
4014 SC/02 Eximstats: use FileHandles for outputting results.
4016 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4018 SC/04 Eximstats: fixed display of large numbers with -nvr option
4020 SC/05 Eximstats: fixed merging of reports with empty tables.
4022 SC/06 Eximstats: added the -include_original_destination flag
4024 SC/07 Eximstats: removed tabs and trailing whitespace.
4026 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4028 TK/06 MBOX spool code: Add real "From " MBOX separator line
4029 so the .eml file is really in mbox format (even though
4030 most programs do not really care). Patch from Alex Miller.
4032 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4033 The latter is generated from $received_to and is only set if the
4034 message has one envelope recipient. SA can use these headers,
4035 obviously out-of-the-box. Patch from Alex Miller.
4037 PH/08 The ${def test on a variable was returning false if the variable's
4038 value was "0", contrary to what the specification has always said!
4039 The result should be true unless the variable is empty.
4041 PH/09 The syntax error of a character other than { following "${if
4042 def:variable_name" (after optional whitespace) was not being diagnosed.
4043 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4044 accidental colon was present, for example, could give incorrect results.
4046 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4047 result is used (not including appendfile, where other changes are about
4050 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4051 This involved changing a lot of size variables from int to off_t. It
4052 should work with maildirs and everything.
4054 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4055 spamd dying while we are connected to it.
4057 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4060 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4061 function for BDB 4.3.
4063 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4065 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4066 the sources for PCRE have been re-arranged and more files are now
4069 PH/15 The code I had for printing potentially long long variables in PH/11
4070 above was not the best (it lost precision). The length of off_t variables
4071 is now inspected at build time, and an appropriate printing format (%ld
4072 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4073 to be "long long int" or "long int". This is needed for the internal
4074 formatting function string_vformat().
4076 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4077 the configuration file to be ":syslog", then the script "guesses" where
4078 the logs files are, rather than using the compiled in default. In our
4079 case the guess is not the same as the compiled default, so the script
4080 suddenly stopped working when I started to use syslog. The patch checks
4081 to see if log_file_path is "". If so, it attempts to read it from exim
4082 with no configuration file to get the compiled in version, before it
4083 falls back to the previous guessing code."
4085 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4086 implementing BATV in an Exim configuration. See NewStuff for the gory
4089 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4090 Makefile that are specific to HP-UX.
4092 PH/18 If the "use_postmaster" option was set for a recipient callout together
4093 with the "random" option, the postmaster address was used as the MAIL
4094 FROM address for the random test, but not for the subsequent recipient
4095 test. It is now used for both.
4097 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4098 patch removes a few documentation additions to RFC 3028, because the
4099 latest draft now contains them. It adds the new en;ascii-case comparator
4100 and a new error check for 8bit text in MIME parts. Comparator and
4101 require names are now matched exactly. I enabled the subaddress
4102 extension, but it is not well tested yet (read: it works for me)."
4104 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4105 rework some of the code of TK/09 above to avoid the hardwired use of
4106 "%lld" and "long long". Replaced the call to snprintf() with a call to
4109 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4110 records point to non-existent hosts", "retry timeout exceeded", and
4111 "retry time not reached for any host after a long failure period".
4113 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4114 experimental DomainKeys support:
4116 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4117 (2) On an error such as an illegally used "control", the wrong name for
4118 the control was given.
4120 These problems did NOT occur unless DomainKeys support was compiled.
4122 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4124 PH/24 Added ${if match_ip condition.
4126 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4127 databases so that it will be absolutely obvious if a crash occurs in the
4128 DB library. This is a regular occurrence (often caused by mis-matched
4131 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4132 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4133 picked up on a user's system that detects such things. There doesn't seem
4134 to be a gcc warning option for this - only an attribute that has to be
4135 put on the function's prototype. It seems that in Fedora Core 4 they have
4136 set this on a number of new functions. No doubt there will be more in due
4139 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4140 specify a transport. However, if an address that was verified by such a
4141 router was the subject of a callout, Exim crashed because it tried to
4142 read the rcpt_include_affixes from the non-existent transport. Now it
4143 just assumes that the setting of that option is false. This bug was
4144 introduced by 4.51/PH/31.
4146 PH/28 Changed -d+all to exclude +memory, because that information is very
4147 rarely of interest, but it makes the output a lot bigger. People tend to
4148 do -d+all out of habit.
4150 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4151 code in os-type was giving problems when libc.so lives in lib64, like on
4154 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4155 aren't the modern standard, and it seems that some systems' include files
4156 don't always have them. Exim was already checking for some of the newer
4157 ones like T_AAAA, and defining it itself. I've added checks for all the
4158 record types that Exim uses.
4160 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4161 not automatically generating a new one, as it is supposed to. This
4162 prevented TLS from working. If the file did exist, but contained invalid
4163 data, a new version was generated, as expected. It was only the case of a
4164 non-existent file that was broken.
4166 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4167 with a change in libdomainkeys > 0.64.
4169 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4170 from DNS. If the selector record carries the flag, it now has
4171 precedence over the domain-wide flag.
4173 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4175 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4176 the use of a port name that isn't defined in /etc/services) occurred, the
4177 message was deferred as in a normal delivery, and thus remained on the
4178 spool, instead of being failed because of the mua_wrapper setting. This
4179 is now fixed, and I tidied up some of the mua_wrapper messages at the
4182 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4183 the messages in a hash of arrays rather than using individual hashes.
4184 This is a bit cleaner and results in dramatic memory savings, albeit
4185 at a slight CPU cost.
4187 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4188 as requested by Marc Sherman.
4190 SC/10 Eximstats: added histograms for user specified patterns as requested
4193 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4195 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4196 fopen() in the content-scanning modules that did not already have it.
4202 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4203 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4205 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4207 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4209 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4210 to test Sieve filters that use "vacation".
4212 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4213 that changes the way the GnuTLS parameters are stored in the cache file.
4214 The new format can be generated externally. For backward compatibility,
4215 if the data in the cache doesn't make sense, Exim assumes it has read an
4216 old-format file, and it generates new data and writes a new file. This
4217 means that you can't go back to an older release without removing the
4220 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4221 work if there are any delivery delays because "one_time" forces the
4222 parent to be marked "delivered", so its unseen clone is never tried
4223 again. For this reason, Exim now forbids the simultaneous setting of
4226 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4227 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4228 entry if you want to know the details. The fix, however, neglected to
4229 consider the case when local delivery batching is involved. The test for
4230 "previously delivered" was not happening when checking to see if an
4231 address could be batched with a previous (undelivered) one; under
4232 certain circumstances this could lead to multiple deliveries to the same
4235 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4236 in its include files, and this causes problems building Exim.
4238 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4239 header_syntax) but Exim was just ignoring anything given after a slash.
4240 In particular, this caused confusion with an attempt to use "verify =
4241 reverse_host_lookup/defer_ok". An error is now given when options are
4242 supplied for verify items that do not have them. (Maybe reverse_host_
4243 lookup should have a defer_ok option, but that's a different point.)
4245 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4246 defined by RFC 821) to 2048, because there were problems with some AUTH
4247 commands, and RFC 1869 says the size should be increased for extended
4248 SMTP commands that take arguments.
4250 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4253 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4254 "unknown" error; now it says that the functionality isn't in the binary.
4256 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4257 an address' error message when a string expansion fails (syntax or
4258 whatever). Otherwise the password may appear in the log. Following change
4259 PH/42 below, there is no longer a chance of it appearing in a bounce
4262 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4264 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4265 of its addresses was ever tried. (Bugzilla bug #2.)
4267 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4268 the result incorrectly in the debug output. (It correctly added a newline
4269 to what was transported.)
4271 TF/01 Added $received_time.
4273 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4274 commented out examples of how to interface to a virus scanner and to
4275 SpamAssassin. Also added commented examples of av_scanner and
4276 spamd_address settings.
4278 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4279 and controls are allowed in which ACLs. There were a couple of minor
4280 errors. Some of the entries in the conditions table (which is a table of
4281 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4282 negation of where the condition IS allowed.
4284 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4286 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4287 header file does not have a version number, so I've had to invent a new
4288 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4289 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4290 radiusclient), but it was contributed by a Radius user.
4292 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4293 files or directories, for OpenSSL.
4295 PH/19 When an Exim process that is running as root has to create an Exim log
4296 file, it does so in a subprocess that runs as exim:exim so as to get the
4297 ownership right at creation (otherwise, other Exim processes might see
4298 the file with the wrong ownership). There was no test for failure of this
4299 fork() call, which would lead to the process getting stuck as it waited
4300 for a non-existent subprocess. Forks do occasionally fail when resources
4301 run out. I reviewed all the other calls to fork(); they all seem to check
4304 PH/20 When checking for unexpected SMTP input at connect time (before writing
4305 the banner), Exim was not dealing correctly with a non-positive return
4306 from the read() function. If the client had disconnected by this time,
4307 the result was a log entry for a synchronization error with an empty
4308 string after "input=" when read() returned zero. If read() returned -1
4309 (an event I could not check), uninitialized data bytes were printed.
4310 There were reports of junk text (parts of files, etc) appearing after
4313 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4315 PH/22 Added support for macro redefinition, and (re)definition in between
4316 driver and ACL definitions.
4318 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4319 forgetting to use the resulting value; it was using the unexpanded value.
4321 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4322 hadn't been configured. The fix is from Juergen Kreileder, who
4323 understands it better than I do:
4325 "Here's what I see happening with three configured cyrus_sasl
4326 authenticators configured (plain, login, cram-md5):
4328 On startup auth_cyrus_sasl_init() gets called for each of these.
4329 This means three calls to sasl_listmech() without a specified mech_list.
4330 => SASL tests which mechs of all available mechs actually work
4331 => three warnings about OTP not working
4332 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4334 With the patch, sasl_listmech() also gets called three times. But now
4335 SASL's mech_list option is set to the server_mech specified in the the
4336 authenticator. Or in other words, the answer from sasl_listmech()
4337 gets limited to just the mech you're testing for (which is different
4339 => the return list contains just 'plain' or 'login', 'cram-md5' or
4340 nothing depending on the value of ob->server_mech.
4342 I've just tested the patch: Authentication still works fine,
4343 unavailable mechs specified in the exim configuration are still
4344 caught, and the auth.log warnings about OTP are gone."
4346 PH/25 When debugging is enabled, the contents of the command line are added
4347 to the debugging output, even when log_selector=+arguments is not
4350 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4351 answer is "GNU", and only if the return is "GNU/something" is the answer
4354 PH/27 $acl_verify_message is now set immediately after the failure of a
4355 verification in an ACL, and so is available in subsequent modifiers. In
4356 particular, the message can be preserved by coding like this:
4358 warn !verify = sender
4359 set acl_m0 = $acl_verify_message
4361 Previously, $acl_verify_message was set only while expanding "message"
4362 and "log_message" when a very denied access.
4364 PH/28 Modified OS/os.c-Linux with
4366 -#ifndef OS_LOAD_AVERAGE
4367 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4369 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4370 nomenclature these days.)
4372 PH/29 Installed patch from the Sieve maintainer that adds the options
4373 sieve_useraddress and sieve_subaddress to the redirect router.
4375 PH/30 In these circumstances:
4376 . Two addresses routed to the same list of hosts;
4377 . First host does not offer TLS;
4378 . First host accepts first address;
4379 . First host gives temporary error to second address;
4380 . Second host offers TLS and a TLS session is established;
4381 . Second host accepts second address.
4382 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4383 and peerdn, if requested) that were in fact used only for the second
4386 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4387 attention to any local part prefix or suffix that was matched by the
4388 router that accepted the address. It now behaves in the same way as it
4389 does for delivery: the affixes are removed from the local part unless
4390 rcpt_include_affixes is set on the transport.
4392 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4393 timeout during the DATA phase of an incoming message.
4395 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4396 applied a patch sanctioned by the Sieve maintainer.
4398 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4399 the uid or gid is negative. A case of a negative gid caused this to be
4400 noticed. The fix allows for either to be negative.
4402 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4403 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4406 PH/36 The change PH/12 above was broken. Fixed it.
4408 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4409 the grounds that routing the same address twice would always produce the
4410 same answer. This might have been true once, but it is certainly no
4411 longer true now. Routing a child address may depend on the previous
4412 routing that produced that child. Some complicated redirection strategies
4413 went wrong when messages had multiple recipients, and made Exim's
4414 behaviour dependent on the order in which the addresses were given.
4416 I have moved the duplicate checking until after the routing is complete.
4417 Exim scans the addresses that are assigned to local and remote
4418 transports, and removes any duplicates. This means that more work will be
4419 done, as duplicates will always all be routed, but duplicates are
4420 presumably rare, so I don't expect this is of any significance.
4422 For deliveries to pipes, files, and autoreplies, the duplicate checking
4423 still happens during the routing process, since they are not going to be
4426 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4427 It corrects a timeout issue with spamd. This is Ian's comment: "The
4428 background is that sometimes spamd either never reads data from a
4429 connection it has accepted, or it never writes response data. The exiscan
4430 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4431 blindly assumes that writes won't block so it may never time out."
4433 PH/39 Allow G after quota size as well as K and M.
4435 PH/40 The value set for $authenticated_id in an authenticator may not contain
4436 binary zeroes or newlines because the value is written to log lines and
4437 to spool files. There was no check on this. Now the value is run through
4438 the string_printing() function so that such characters are converted to
4439 printable escape sequences.
4441 PH/41 $message_linecount is a new variable that contains the total number of
4442 lines in the message. Compare $body_linecount, which is the count for the
4445 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4446 bounce and delay warning messages, except in certain special cases, which
4449 (a) An SMTP error message from a remote host;
4450 (b) A message specified in a :fail: redirection;
4451 (c) A message specified in a "fail" command in a system filter;
4452 (d) A message specified in a FAIL return from the queryprogram router;
4453 (e) A message specified by the cannot_route_message router option.
4455 In these cases only, Exim does include the error details in bounce and
4456 warning messages. There are also a few cases where bland messages such
4457 as "unrouteable address" or "local delivery error" are given.
4459 PH/43 $value is now also set for the "else" part of a ${run expansion.
4461 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4462 being worked on, but at least Exim now implements the latest version to
4465 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4466 process to complete was treated as a delivery failure, a timeout while
4467 writing the message to the pipe was logged, but erroneously treated as a
4468 successful delivery. Such timeouts include transport filter timeouts. For
4469 consistency with the overall process timeout, these timeouts are now
4470 treated as errors, giving rise to delivery failures by default. However,
4471 there is now a new Boolean option for the pipe transport called
4472 timeout_defer, which, if set TRUE, converts the failures into defers for
4473 both kinds of timeout. A transport filter timeout is now identified in
4476 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4477 systems where "make" and "gmake" are different, calling "gmake" at top
4478 level broke things. I've arranged for the value of $(MAKE) to be passed
4479 from the Makefile to this script so that it can call the same version of
4483 A note about Exim versions 4.44 and 4.50
4484 ----------------------------------------
4486 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4487 changes of various kinds. As a consequence, a big documentation update was
4488 needed. This delayed the release for rather longer than seemed good, especially
4489 in the light of a couple of (minor) security issues. Therefore, the changes
4490 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4491 release. So 4.44 and 4.50 are in effect two different branches that both start
4494 I have left the 4.50 change log unchanged; it contains all the changes since
4495 4.43. The change log for 4.44 is below; many of its items are identical to
4496 those for 4.50. This seems to be the most sensible way to preserve the
4497 historical information.
4503 1. Minor wording change to the doc/README.SIEVE file.
4505 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4506 computation of the current number of files was incorrect.
4508 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4509 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4512 4. Give more explanation in the error message when the command for a transport
4513 filter fails to execute.
4515 5. There are several places where Exim runs a non-Exim command in a
4516 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4517 was being done only for the command run by the queryprogram router. It is
4518 now done for all such subprocesses. The other cases are: ${run, transport
4519 filters, and the commands run by the lmtp and pipe transports.
4521 6. Added CONFIGURE_GROUP build-time option.
4523 7. Some older OS have a limit of 256 on the maximum number of file
4524 descriptors. Exim was using setrlimit() to set 1000 as a large value
4525 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4526 systems. I've change it so that if it can't get 1000, it tries for 256.
4528 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4529 was an oversight, and furthermore, ever since the addition of extra
4530 controls (e.g. 4.43/32), the checks on when to allow different forms of
4531 "control" were broken. There should now be diagnostics for all cases when a
4532 control that does not make sense is encountered.
4534 9. Added the /retain_sender option to "control=submission".
4536 10. $recipients is now available in the predata ACL (oversight).
4538 11. Tidy the search cache before the fork to do a delivery from a message
4539 received from the command line. Otherwise the child will trigger a lookup
4540 failure and thereby defer the delivery if it tries to use (for example) a
4541 cached ldap connection that the parent has called unbind on.
4543 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4544 of $address_data from the recipient verification was clobbered by the
4545 sender verification.
4547 13. The value of address_data from a sender verification is now available in
4548 $sender_address_data in subsequent conditions in the ACL statement.
4550 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4552 15. Added a new option "connect=<time>" to callout options, to set a different
4555 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4556 was its contents. (It was OK if the option was not defined at all.)
4558 17. A "Completed" log line is now written for messages that are removed from
4559 the spool by the -Mrm option.
4561 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4562 information about exactly what failed.
4564 19. Added -dd to debug only the daemon process.
4566 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4567 handles timeouts, both on the server side and network timeouts. Renamed the
4568 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4570 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4571 It is now set to "smtps".
4573 22. $host_address is now set to the target address during the checking of
4574 ignore_target_hosts.
4576 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4577 being passed; this would have caused $sender_host_name to have been used if
4578 matching the list had actually called for a host name (not very likely,
4579 since this list is usually IP addresses). A host name is now passed as
4582 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4583 code that specifies a non-restarting handler (typically sigaction() in
4584 modern systems) in an attempt to fix a rare and obscure crash bug.
4586 25. Narrowed the window for a race in the daemon that could cause it to ignore
4587 SIGCHLD signals. This is not a major problem, because they are used only to
4588 wake it up if nothing else does.
4590 26. A malformed maildirsize file could cause Exim to calculate negative values
4591 for the mailbox size or file count. Odd effects could occur as a result.
4592 The maildirsize information is now recalculated if the size or filecount
4595 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4596 support for a long time. Removed HAVE_SYS_VFS_H.
4598 28. Installed the latest version of exipick from John Jetmore.
4600 29. In an address list, if the pattern was not a regular expression, an empty
4601 subject address (from a bounce message) matched only if the pattern was an
4602 empty string. Non-empty patterns were not even tested. This was the wrong
4603 because it is perfectly reasonable to use an empty address as part of a
4604 database query. An empty address is now tested by patterns that are
4605 lookups. However, all the other forms of pattern expect the subject to
4606 contain a local part and a domain, and therefore, for them, an empty
4607 address still always fails if the pattern is not itself empty.
4609 30. Exim went into a mad DNS loop when attempting to do a callout where the
4610 host was specified on an smtp transport, and looking it up yielded more
4611 than one IP address.
4613 31. Re-factored the code for checking spool and log partition space into a
4614 function that finds that data and another that does the check. The former
4615 is then used to implement four new variables: $spool_space, $log_space,
4616 $spool_inodes, and $log_inodes.
4618 32. The RFC2047 encoding function was originally intended for short strings
4619 such as real names; it was not keeping to the 75-character limit for
4620 encoded words that the RFC imposes. It now respects the limit, and
4621 generates multiple encoded words if necessary. To be on the safe side, I
4622 have increased the buffer size for the ${rfc2047: expansion operator from
4625 33. It is now permitted to omit both strings after an "if" condition; if the
4626 condition is true, the result is "true". As before, when the second string
4627 is omitted, a false condition yields an empty string. This makes it less
4628 cumbersome to write custom ACL and router conditions.
4630 34. Failure to deliver a bounce message always caused it to be frozen, even if
4631 there was an errors_to setting on the router. The errors_to setting is now
4634 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4635 canonical form (fully expanded) before being placed in
4636 $sender_host_address.
4638 36. The table in the code that translates DNS record types into text (T_A to
4639 "A" for instance) was missing entries for NS and CNAME. It is just possible
4640 that this could have caused confusion if both these types were looked up
4641 for the same domain, because the text type is used as part of Exim's
4642 per-process caching. But the chance of anyone hitting this buglet seems
4645 37. The dnsdb lookup has been extended in a number of ways.
4647 (1) There is a new type, "zns", which walks up the domain tree until it
4648 finds some nameserver records. It should be used with care.
4650 (2) There is a new type, "mxh", which is like "mx" except that it returns
4651 just the host names, not the priorities.
4653 (3) It is now possible to give a list of domains (or IP addresses) to be
4654 looked up. The behaviour when one of the lookups defers can be
4655 controlled by a keyword.
4657 (4) It is now possible to specify the separator character for use when
4658 multiple records are returned.
4660 38. The dnslists ACL condition has been extended: it is now possible to supply
4661 a list of IP addresses and/or domains to be looked up in a particular DNS
4664 39. Added log_selector=+queue_time_overall.
4666 40. When running the queue in the test harness, wait just a tad after forking a
4667 delivery process, to get repeatability of debugging output.
4669 41. Include certificate and key file names in error message when GnuTLS fails
4670 to set them up, because the GnuTLS error message doesn't include the name
4671 of the failing file when there is a problem reading it.
4673 42. Allow both -bf and -bF in the same test run.
4675 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4677 44. The "Exiscan patch" is now merged into the mainline Exim source.
4679 45. Sometimes the final signoff response after QUIT could fail to get
4680 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4681 < 0 before doing a fflush(). This bug looks as though it goes back to the
4682 introduction of TLS in release 3.20, but "sometimes" must have been rare
4683 because the tests only now provoked it.
4685 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4686 (this can affect the format of dates).
4688 47. exim_tidydb, when checking for the continued existence of a message for
4689 which it has found a message-specific retry record, was not finding
4690 messages that were in split spool directories. Consequently, it was
4691 deleting retry records that should have stayed in existence.
4693 48. Steve fixed some bugs in eximstats.
4695 49. The SPA authentication driver was not abandoning authentication and moving
4696 on to the next authenticator when an expansion was forced to fail,
4697 contradicting the general specification for all authenticators. Instead it
4698 was generating a temporary error. It now behaves as specified.
4700 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4701 (the order specifies the preference for clients). The order is now AES256,
4702 AES128, 3DES, ARCFOUR128.
4704 51. Small patch to Sieve code - explicitly set From: when generating an
4707 52. Exim crashed if a remote delivery caused a very long error message to be
4708 recorded - for instance if somebody sent an entire SpamAssassin report back
4709 as a large number of 550 error lines. This bug was coincidentally fixed by
4710 increasing the size of one of Exim's internal buffers (big_buffer) that
4711 happened as part of the Exiscan merge. However, to be on the safe side, I
4712 have made the code more robust (and fixed the comments that describe what
4715 53. Now that there can be additional text after "Completed" in log lines (if
4716 the queue_time_overall log selector is set), a one-byte patch to exigrep
4717 was needed to allow it to recognize "Completed" as not the last thing in
4720 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4721 patch that reportedly fixes this has been added. I am not expert enough to
4722 create a test for it. This is what the patch creator wrote:
4724 "I found a little strange behaviour of ldap code when working with
4725 Windows 2003 AD Domain, where users was placed in more than one
4726 Organization Units. When I tried to give exim partial DN, the exit code
4727 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4728 But simultaneously result of request was absolutely normal ldap result,
4729 so I produce this patch..."
4731 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4732 REFERENCE, so I have modified the code to exclude the patch when that macro
4735 55. Some experimental protocols are using DNS PTR records for new purposes. The
4736 keys for these records are domain names, not reversed IP addresses. The
4737 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4738 leaves it alone. Component reversal etc. now happens only for IP addresses.
4741 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4743 57. Double the size of the debug message buffer (to 2048) so that more of very
4744 long debug lines gets shown.
4746 58. The exicyclog utility now does better if the number of log files to keep
4747 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4749 59. Two changes related to the smtp_active_hostname option:
4751 (1) $smtp_active_hostname is now available as a variable.
4752 (2) The default for smtp_banner uses $smtp_active_hostname instead
4753 of $primary_hostname.
4755 60. The host_aton() function is supposed to be passed a string that is known
4756 to be a valid IP address. However, in the case of IPv6 addresses, it was
4757 not checking this. This is a hostage to fortune. Exim now panics and dies
4758 if the condition is not met. A case was found where this could be provoked
4759 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4760 components; fortuitously, this particular loophole had already been fixed
4761 by change 4.50/55 above.
4763 If there are any other similar loopholes, the new check in host_aton()
4764 itself should stop them being exploited. The report I received stated that
4765 data on the command line could provoke the exploit when Exim was running as
4766 exim, but did not say which command line option was involved. All I could
4767 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4768 running as the user.
4771 61. There was a buffer overflow vulnerability in the SPA authentication code
4772 (which came originally from the Samba project). I have added a test to the
4773 spa_base64_to_bits() function which I hope fixes it.
4776 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4777 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4779 63. The daemon start-up calls getloadavg() while still root for those OS that
4780 need the first call to be done as root, but it missed one case: when
4781 deliver_queue_load_max is set with deliver_drop_privilege. This is
4782 necessary for the benefit of the queue runner, because there is no re-exec
4783 when deliver_drop_privilege is set.
4785 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4786 This has been fixed.
4788 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4789 was in use, was not putting the data itself into the right store pool;
4790 consequently, it could be overwritten for a subsequent message in the same
4791 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4794 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4796 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4797 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4798 the function were treating the return as a boolean value, which happened to
4799 work because 0=false and not-0=true, but is not correct code.
4801 68. The host_aton() function was not handling scoped IPv6 addresses (those
4802 with, for example, "%eth0" on the end) correctly.
4804 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4805 negated items (that is, ~something) in unsigned ints. Some compilers
4806 apparently mutter when there is no cast.
4808 70. If an address verification called from an ACL failed, and did not produce a
4809 user-specific message (i.e. there was only a "system" message), nothing was
4810 put in $acl_verify_message. In this situation, it now puts the system
4813 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4814 session; change 4.31/43 added the unwanted input to the log line - except
4815 that it did not do this in the start of session case. It now does.
4817 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4818 This is wrong and can cause the other end to generate a synchronization
4819 error if it is another Exim or anything else that does the synchronization
4820 check. A QUIT command is no longer sent after a timeout.
4822 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4823 during host lookups.
4825 74. The defer_ok option of callout verification was not working if it was used
4826 when verifying addresses in header lines, that is, for this case:
4828 verify = header_sender/callout=defer_ok
4830 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4831 those file descriptors could be used for SMTP connections. If anything
4832 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4833 could be sent to the SMTP client, causing chaos. The daemon now opens
4834 stdin, stdout, and stderr to /dev/null when it puts itself into the
4837 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4838 log by default. The user can override this with suitable Perl magic.
4840 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4841 the log message when discard triggers, was not working for the DATA ACL or
4842 for the non-SMTP ACL.
4844 78. Error message wording change in sieve.c.
4846 79. If smtp_accept_max_per_host was set, the number of connections could be
4847 restricted to fewer than expected, because the daemon was trying to set up
4848 a new connection before checking whether the processes handling previous
4849 connections had finished. The check for completed processes is now done
4850 earlier. On busy systems, this bug wouldn't be noticed because something
4851 else would have woken the daemon, and it would have reaped the completed
4854 80. If a message was submitted locally by a user whose login name contained one
4855 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4856 It caused a spool format error. I have fixed the spool reading code. A
4857 related problem was that the "from" clause in the Received: line became
4858 illegal because of the space(s). It is now covered by ${quote_local_part.
4860 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4863 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4865 83. Patch from Sieve maintainer to fix unterminated string problem in
4866 "vacation" handling.
4868 84. Some minor changes to the Linux configuration files to help with other
4869 OS variants using glibc.
4871 85. One more patch for Sieve to update vacation handling to latest spec.
4874 ----------------------------------------------------
4875 See the note above about the 4.44 and 4.50 releases.
4876 ----------------------------------------------------
4882 1. Change 4.43/35 introduced a bug that caused file counts to be
4883 incorrectly computed when quota_filecount was set in an appendfile
4886 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4887 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4890 3. Give more explanation in the error message when the command for a transport
4891 filter fails to execute.
4893 4. There are several places where Exim runs a non-Exim command in a
4894 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4895 was being done only for the command run by the queryprogram router. It is
4896 now done for all such subprocesses. The other cases are: ${run, transport
4897 filters, and the commands run by the lmtp and pipe transports.
4899 5. Some older OS have a limit of 256 on the maximum number of file
4900 descriptors. Exim was using setrlimit() to set 1000 as a large value
4901 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4902 systems. I've change it so that if it can't get 1000, it tries for 256.
4904 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4905 was an oversight, and furthermore, ever since the addition of extra
4906 controls (e.g. 4.43/32), the checks on when to allow different forms of
4907 "control" were broken. There should now be diagnostics for all cases when a
4908 control that does not make sense is encountered.
4910 7. $recipients is now available in the predata ACL (oversight).
4912 8. Tidy the search cache before the fork to do a delivery from a message
4913 received from the command line. Otherwise the child will trigger a lookup
4914 failure and thereby defer the delivery if it tries to use (for example) a
4915 cached ldap connection that the parent has called unbind on.
4917 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4918 of $address_data from the recipient verification was clobbered by the
4919 sender verification.
4921 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4922 was its contents. (It was OK if the option was not defined at all.)
4924 11. A "Completed" log line is now written for messages that are removed from
4925 the spool by the -Mrm option.
4927 12. $host_address is now set to the target address during the checking of
4928 ignore_target_hosts.
4930 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4931 being passed; this would have caused $sender_host_name to have been used if
4932 matching the list had actually called for a host name (not very likely,
4933 since this list is usually IP addresses). A host name is now passed as
4936 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4937 code that specifies a non-restarting handler (typically sigaction() in
4938 modern systems) in an attempt to fix a rare and obscure crash bug.
4940 15. Narrowed the window for a race in the daemon that could cause it to ignore
4941 SIGCHLD signals. This is not a major problem, because they are used only to
4942 wake it up if nothing else does.
4944 16. A malformed maildirsize file could cause Exim to calculate negative values
4945 for the mailbox size or file count. Odd effects could occur as a result.
4946 The maildirsize information is now recalculated if the size or filecount
4949 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4950 support for a long time. Removed HAVE_SYS_VFS_H.
4952 18. Updated exipick to current release from John Jetmore.
4954 19. Allow an empty sender to be matched against a lookup in an address list.
4955 Previously the only cases considered were a regular expression, or an
4958 20. Exim went into a mad DNS lookup loop when doing a callout where the
4959 host was specified on the transport, if the DNS lookup yielded more than
4962 21. The RFC2047 encoding function was originally intended for short strings
4963 such as real names; it was not keeping to the 75-character limit for
4964 encoded words that the RFC imposes. It now respects the limit, and
4965 generates multiple encoded words if necessary. To be on the safe side, I
4966 have increased the buffer size for the ${rfc2047: expansion operator from
4969 22. Failure to deliver a bounce message always caused it to be frozen, even if
4970 there was an errors_to setting on the router. The errors_to setting is now
4973 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4974 canonical form (fully expanded) before being placed in
4975 $sender_host_address.
4977 24. Updated eximstats to version 1.33
4979 25. Include certificate and key file names in error message when GnuTLS fails
4980 to set them up, because the GnuTLS error message doesn't include the name
4981 of the failing file when there is a problem reading it.
4983 26. Expand error message when OpenSSL has problems setting up cert/key files.
4986 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4987 (this can affect the format of dates).
4989 28. exim_tidydb, when checking for the continued existence of a message for
4990 which it has found a message-specific retry record, was not finding
4991 messages that were in split spool directories. Consequently, it was
4992 deleting retry records that should have stayed in existence.
4994 29. eximstats updated to version 1.35
4995 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4996 1.35 - bugfix such that pie charts by volume are generated correctly
4998 30. The SPA authentication driver was not abandoning authentication and moving
4999 on to the next authenticator when an expansion was forced to fail,
5000 contradicting the general specification for all authenticators. Instead it
5001 was generating a temporary error. It now behaves as specified.
5003 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5004 (the order specifies the preference for clients). The order is now AES256,
5005 AES128, 3DES, ARCFOUR128.
5007 31. Small patch to Sieve code - explicitly set From: when generating an
5010 32. Exim crashed if a remote delivery caused a very long error message to be
5011 recorded - for instance if somebody sent an entire SpamAssassin report back
5012 as a large number of 550 error lines. This bug was coincidentally fixed by
5013 increasing the size of one of Exim's internal buffers (big_buffer) that
5014 happened as part of the Exiscan merge. However, to be on the safe side, I
5015 have made the code more robust (and fixed the comments that describe what
5018 33. Some experimental protocols are using DNS PTR records for new purposes. The
5019 keys for these records are domain names, not reversed IP addresses. The
5020 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5021 leaves it alone. Component reversal etc. now happens only for IP addresses.
5024 34. The host_aton() function is supposed to be passed a string that is known
5025 to be a valid IP address. However, in the case of IPv6 addresses, it was
5026 not checking this. This is a hostage to fortune. Exim now panics and dies
5027 if the condition is not met. A case was found where this could be provoked
5028 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5029 components; fortuitously, this particular loophole had already been fixed
5030 by change 4.50/55 or 4.44/33 above.
5032 If there are any other similar loopholes, the new check in host_aton()
5033 itself should stop them being exploited. The report I received stated that
5034 data on the command line could provoke the exploit when Exim was running as
5035 exim, but did not say which command line option was involved. All I could
5036 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5037 running as the user.
5040 35. There was a buffer overflow vulnerability in the SPA authentication code
5041 (which came originally from the Samba project). I have added a test to the
5042 spa_base64_to_bits() function which I hope fixes it.
5045 36. The daemon start-up calls getloadavg() while still root for those OS that
5046 need the first call to be done as root, but it missed one case: when
5047 deliver_queue_load_max is set with deliver_drop_privilege. This is
5048 necessary for the benefit of the queue runner, because there is no re-exec
5049 when deliver_drop_privilege is set.
5051 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5052 was in use, was not putting the data itself into the right store pool;
5053 consequently, it could be overwritten for a subsequent message in the same
5054 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5057 38. Sometimes the final signoff response after QUIT could fail to get
5058 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5059 < 0 before doing a fflush(). This bug looks as though it goes back to the
5060 introduction of TLS in release 3.20, but "sometimes" must have been rare
5061 because the tests only now provoked it.
5067 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5068 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5069 Now it returns an integer. A number of calls were still expecting a T/F
5070 return. Fortuitously, in all cases, the tests worked in OK situations,
5071 which is the norm. However, things would have gone wrong on any write
5072 failures on the smtp file descriptor. This function is used when sending
5073 messages over SMTP and also when doing verify callouts.
5075 2. When Exim is called to do synchronous delivery of a locally submitted
5076 message (the -odf or -odi options), it no longer closes stderr before doing
5079 3. Implemented the mua_wrapper option.
5081 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5083 5. Implemented the functions header_remove(), header_testname(),
5084 header_add_at_position(), and receive_remove_recipient(), and exported them
5087 6. If an ACL "warn" statement specified the addition of headers, Exim already
5088 inserted X-ACL-Warn: at the start if there was no header name. However, it
5089 was not making this test for the second and subsequent header lines if
5090 there were newlines in the string. This meant that an invalid header could
5091 be inserted if Exim was badly configured.
5093 7. Allow an ACL "warn" statement to add header lines at the start or after all
5094 the Received: headers, as well as at the end.
5096 8. Added the rcpt_4xx retry error code.
5098 9. Added postmaster_mailfrom=xxx to callout verification option.
5100 10. Added mailfrom=xxxx to the callout verification option, for verify=
5103 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5104 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5106 12. Inserted some casts to stop certain compilers warning when using pointer
5107 differences as field lengths or precisions in printf-type calls (mostly
5108 affecting debugging statements).
5110 13. Added optional readline() support for -be (dynamically loaded).
5112 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5113 same clock tick as a message's arrival, so that its received time was the
5114 same as the "first fail" time on the retry record, and that message
5115 remained on the queue past the ultimate address timeout, every queue runner
5116 would try a delivery (because it was past the ultimate address timeout) but
5117 after another failure, the ultimate address timeout, which should have then
5118 bounced the address, did not kick in. This was a "< instead of <=" error;
5119 in most cases the first failure would have been in the next clock tick
5120 after the received time, and all would be well.
5122 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5123 being recognized when the domain list was tested by the match_domain
5124 condition in an expansion string.
5126 16. Added the ${str2b64: operator.
5128 17. Exim was always calling setrlimit() to set a large limit for the number of
5129 processes, without checking whether the existing limit was already
5130 adequate. (It did check for the limit on file descriptors.) Furthermore,
5131 errors from getrlimit() and setrlimit() were being ignored. Now they are
5132 logged to the main and panic logs, but Exim does carry on, to try to do its
5133 job under whatever limits there are.
5135 18. Imported PCRE 5.0.
5137 19. Trivial typo in log message " temporarily refused connection" (the leading
5140 20. If the log selector return_path_on_delivery was set and an address was
5141 redirected to /dev/null, the delivery process crashed because it assumed
5142 that a return path would always be set for a "successful" delivery. In this
5143 case, the whole delivery is bypassed as an optimization, and therefore no
5146 21. Internal re-arrangement: the function for sending a challenge and reading
5147 a response while authentication was assuming a zero-terminated challenge
5148 string. It's now changed to take a pointer and a length, to allow for
5149 binary data in such strings.
5151 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5153 23. Exim was not respecting finduser_retries when seeking the login of the
5154 uid under which it was called; it was always trying 10 times. (The default
5155 setting of finduser_retries is zero.) Also, it was sleeping after the final
5156 failure, which is pointless.
5158 24. Implemented tls_on_connect_ports.
5160 25. Implemented acl_smtp_predata.
5162 26. If the domain in control=submission is set empty, Exim assumes that the
5163 authenticated id is a complete email address when it generates From: or
5164 Sender: header lines.
5166 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5167 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5168 chown and chgrp in /bin and hostname in /usr/bin.
5170 28. Exim was keeping the "process log" file open after each use, just as it
5171 does for the main log. This opens the possibility of it remaining open for
5172 long periods when the USR1 signal hits a daemon. Occasional processlog
5173 errors were reported, that could have been caused by this. Anyway, it seems
5174 much more sensible not to leave this file open at all, so that is what now
5177 29. The long-running daemon process does not normally write to the log once it
5178 has entered its main loop, and it closes the log before doing so. This is
5179 so that log files can straightforwardly be renamed and moved. However,
5180 there are a couple of unusual error situations where the daemon does write
5181 log entries, and I had neglected to close the log afterwards.
5183 30. The text of an SMTP error response that was received during a remote
5184 delivery was being truncated at 512 bytes. This is too short for some of
5185 the long messages that one sometimes sees. I've increased the limit to
5188 31. It is now possible to make retry rules that apply only when a message has a
5189 specific sender, in particular, an empty sender.
5191 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5192 it possible to be selective about when SMTP synchronization is enforced.
5194 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5196 32. Implemented hosts_connection_nolog.
5198 33. Added an ACL for QUIT.
5200 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5203 35. Added mailbox_size and mailbox_filecount to appendfile.
5205 36. Added control = no_multiline_responses to ACLs.
5207 37. There was a bug in the logic of the code that waits for the clock to tick
5208 in the case where the clock went backwards by a substantial amount such
5209 that the microsecond fraction of "now" was more than the microsecond
5210 fraction of "then" (but the whole seconds number was less).
5212 38. Added support for the libradius Radius client library this is found on
5213 FreeBSD (previously only the radiusclient library was supported).
5219 1. When certain lookups returned multiple values in the form name=value, the
5220 quoting of the values was not always being done properly. Specifically:
5221 (a) If the value started with a double quote, but contained no whitespace,
5223 (b) If the value contained whitespace other than a space character (i.e.
5224 tabs or newlines or carriage returns) it was not quoted.
5225 This fix has been applied to the mysql and pgsql lookups by writing a
5226 separate quoting function and calling it from the lookup code. The fix
5227 should probably also be applied to nisplus, ibase and oracle lookups, but
5228 since I cannot test any of those, I have not disturbed their existing code.
5230 2. A hit in the callout cache for a specific address caused a log line with no
5231 reason for rejecting RCPT. Now it says "Previous (cached) callout
5232 verification failure".
5234 3. There was an off-by-one bug in the queryprogram router. An over-long
5235 return line was truncated at 256 instead of 255 characters, thereby
5236 overflowing its buffer with the terminating zero. As well as fixing this, I
5237 have increased the buffer size to 1024 (and made a note to document this).
5239 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5240 when Exim is waiting for an SMTP response from a remote server, Exim
5241 restarts its select() call on the socket, thereby resetting its timeout.
5242 This is not a problem when such interrupts are rare. Somebody set up a cron
5243 job to run exiwhat every 2 minutes, which is less than the normal select()
5244 timeout (5 or 10 minutes). This meant that the select() timeout never
5245 kicked in because it was always reset. I have fixed this by comparing the
5246 time when an interrupt arrives with the time at the start of the first call
5247 to select(). If more time than the timeout has elapsed, the interrupt is
5248 treated as a timeout.
5250 5. Some internal re-factoring in preparation for the addition of Sieve
5251 extensions (by MH). In particular, the "personal" test is moved to a
5252 separate function, and given an option for scanning Cc: and Bcc: (which is
5253 not set for Exim filters).
5255 6. When Exim created an email address using the login of the caller as the
5256 local part (e.g. when creating a From: or Sender: header line), it was not
5257 quoting the local part when it contained special characters such as @.
5259 7. Installed new OpenBSD configuration files.
5261 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5262 try to make them clearer.
5264 9. Callout options, other than the timeout value, were being ignored when
5265 verifying sender addresses in header lines. For example, when using
5267 verify = header_sender/callout=no_cache
5269 the cache was (incorrectly) being used.
5271 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5272 only the Cygwin environment.
5274 11. When return_path_on_delivery was set as a log selector, if different remote
5275 addresses in the same message used different return paths and parallel
5276 remote delivery occurred, the wrong values would sometimes be logged.
5277 (Whenever a remote delivery process finished, the return path value from
5278 the most recently started remote delivery process was logged.)
5280 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5281 lines when AUTH and/or TLS are in use. This is the "received protocol"
5282 field. Exim used to use "asmtp" for authenticated SMTP, without any
5283 indication (in the protocol name) for TLS use. Now it follows the RFC and
5284 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5285 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5286 names appear in log lines as well as in Received: header lines.
5288 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5289 extensions, and comparison tests, and to fix some bugs.
5291 14. Changes to the "personal" filter test:
5293 (1) The test was buggy in that it was just doing the equivalent of
5294 "contains" tests on header lines. For example, if a user's address was
5295 anne@some.where, the "personal" test would incorrectly be true for
5297 To: susanne@some.where
5299 This test is now done by extracting each address from the header in turn,
5300 and checking the entire address. Other tests that are part of "personal"
5301 are now done using regular expressions (for example, to check local parts
5302 of addresses in From: header lines).
5304 (2) The list of non-personal local parts in From: addresses has been
5305 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5306 taken from the Sieve specification recommendations.
5308 (3) If the message contains any header line starting with "List-" it is
5309 treated as non-personal.
5311 (4) The test for "circular" in the Subject: header line has been removed
5312 because it now seems ill-conceived.
5314 15. Minor typos in src/EDITME comments corrected.
5316 16. Installed latest exipick from John Jetmore.
5318 17. If headers_add on a router specified a text string that was too long for
5319 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5320 of string_sprintf() is now avoided.
5322 18. $message_body_size was not set (it was always zero) when running the DATA
5323 ACL and the local_scan() function.
5325 19. For the "mail" command in an Exim filter, no default was being set for
5326 the once_repeat time, causing a random time value to be used if "once" was
5327 specified. (If the value happened to be <= 0, no repeat happened.) The
5328 default is now 0s, meaning "never repeat". The "vacation" command was OK
5329 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5330 (I found it when inspecting the code).
5332 20. There is now an overall timeout for performing a callout verification. It
5333 defaults to 4 times the callout timeout, which applies to individual SMTP
5334 commands during the callout. The overall timeout applies when there is more
5335 than one host that can be tried. The timeout is checked before trying the
5336 next host. This prevents very long delays if there are a large number of
5337 hosts and all are timing out (e.g. when the network connections are timing
5338 out). The value of the overall timeout can be changed by specifying an
5339 additional sub-option for "callout", called "maxwait". For example:
5341 verify = sender/callout=5s,maxwait=20s
5343 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5344 to the end before writing, but this should make it even safer).
5346 22. Exim was forgetting that it had advertised PIPELINING for the second and
5347 subsequent messages on an SMTP connection. It was also not resetting its
5348 memory on STARTTLS and an internal HELO.
5350 23. When Exim logs an SMTP synchronization error within a session, it now
5351 records whether PIPELINING has been advertised or not.
5353 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5354 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5355 rather than long int.
5357 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5359 26. Added the never_mail option to autoreply.
5365 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5366 crash if the getsockname() call failed; this can happen if a connection is
5367 closed very soon after it is established. The problem was simply in the
5368 order in which certain operations were done, causing Exim to try to write
5369 to the SMTP stream before it had set up the file descriptor. The bug has
5370 been fixed by making things happen in the correct order.
5376 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5377 before the connection was closed, thus losing the rejection response.
5379 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5380 some early Solaris releases, but causes trouble in current releases where
5381 socklen_t is defined.
5383 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5386 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5389 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5390 that releasing the top of it at the end releases what was used for sub-
5391 expansions (unless the block got too big). However, discard this block if
5392 the first thing is a variable or header, so that we can use its block when
5393 it is dynamic (useful for very large $message_headers, for example).
5395 6. Lookups now cache *every* query, not just the most recent. A new, separate
5396 store pool is used for this. It can be recovered when all lookup caches are
5397 flushed. Lookups now release memory at the end of their result strings.
5398 This has involved some general refactoring of the lookup sources.
5400 7. Some code has been added to the store_xxx() functions to reduce the amount
5401 of flapping under certain conditions.
5403 8. log_incoming_interface used to affect only the <= reception log lines. Now
5404 it causes the local interface and port to be added to several more SMTP log
5405 lines, for example "SMTP connection from", and rejection lines.
5407 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5409 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5411 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5412 could be overwritten at the end of the current message (or the start of a
5413 new message if it was set in a HELO ACL). The value is now preserved for
5414 the duration of the SMTP connection.
5416 12. If a transport had a headers_rewrite setting, and a matching header line
5417 contained an unqualified address, that address was qualified, even if it
5418 did not match any rewriting rules. The underlying bug was that the values
5419 of the flags that permit the existence of unqualified sender and recipient
5420 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5421 non-local messages, and by -bnq for local messages) were not being
5422 preserved with the message after it was received.
5424 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5425 "next input=" as part of the text comprising the host identity instead of
5426 the correct text. The code was using the same buffer for two different
5427 strings. However, depending on which order the printing function evaluated
5428 its arguments, the bug did not always show up. Under Linux, for example, my
5429 test suite worked just fine.
5431 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5432 doesn't work with some older versions of Perl. It has been changed to "my",
5433 which in any case is probably the better facility to use.
5435 15. A really picky compiler found some instances of statements for creating
5436 error messages that either had too many or two few arguments for the format
5439 16. The size of the buffer for calls to the DNS resolver has been increased
5440 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5441 for addresses that have a lot of PTR records. This alleviates a problem; it
5442 does not fully solve it.
5444 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5445 buffer now truncates the list and logs the incident, which is the same
5446 action as happens when Exim is looking up a host name and its aliases.
5447 Previously in this situation something unpredictable would happen;
5448 sometimes it was "internal error: store_reset failed".
5450 18. If a server dropped the connection unexpectedly when an Exim client was
5451 using GnuTLS and trying to read a response, the client delivery process
5452 crashed while trying to generate an error log message.
5454 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5455 string, for example:
5457 warn message = H1: something\nH2: something
5459 the text was added as a single header line from Exim's point of view
5460 though it ended up OK in the delivered message. However, searching for the
5461 second and subsequent header lines using $h_h2: did not work. This has been
5462 fixed. Similarly, if a system filter added multiple headers in this way,
5463 the routers could not see them.
5465 20. Expanded the error message when iplsearch is called with an invalid key to
5466 suggest using net-iplsearch in a host list.
5468 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5469 ACLs are no longer actually imposed (and a message to that effect is
5472 22. If a "gecos" field in a passwd entry contained escaped characters, in
5473 particular, if it contained a \" sequence, Exim got it wrong when building
5474 a From: or a Sender: header from that name. A second bug also caused
5475 incorrect handling when an unquoted " was present following a character
5476 that needed quoting.
5478 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5479 was not being matched caselessly.
5481 24. Arranged for all hyphens in the exim.8 source to be escaped with
5484 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5485 properly. Recipient callout cache records were still being keyed to include
5486 the sender, even when use_sender was set false. This led to far more
5487 callouts that were necessary. The sender is no longer included in the key
5488 when use_sender is false.
5490 26. Added "control = submission" modifier to ACLs.
5492 27. Added the ${base62d: operator to decode base 62 numbers.
5494 28. dnsdb lookups can now access SRV records.
5496 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5497 the configuration file.
5499 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5500 (-v) mode. This makes the output for a verbose queue run more intelligible.
5502 31. Added a use_postmaster feature to recipient callouts.
5504 32. Added the $body_zerocount variable, containing the number of binary zero
5505 bytes in the message body.
5507 33. The time of last modification of the "new" subdirectory is now used as the
5508 "mailbox time last read" when there is a quota error for a maildir
5511 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5513 35. Added +ignore_unknown as a special item in host lists.
5515 36. Code for decoding IPv6 addresses in host lists is now included, even if
5516 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5517 address was recognized as an IP address, but was then not correctly decoded
5518 into binary, causing unexpected and incorrect effects when compared with
5525 1. Very minor rewording of debugging text in manualroute to say "list of
5526 hosts" instead of "hostlist".
5528 2. If verify=header_syntax was set, and a header line with an unqualified
5529 address (no domain) and a large number of spaces between the end of the
5530 name and the colon was received, the reception process suffered a buffer
5531 overflow, and (when I tested it) crashed. This was caused by some obsolete
5532 code that should have been removed. The fix is to remove it!
5534 3. When running in the test harness, delay a bit after writing a bounce
5535 message to get a bit more predictability in the log output.
5537 4. Added a call to search_tidyup() just before forking a reception process. In
5538 theory, someone could use a lookup in the expansion of smtp_accept_max_
5539 per_host which, without the tidyup, could leave open a database connection.
5541 5. Added the variables $recipient_data and $sender_data which get set from a
5542 lookup success in an ACL "recipients" or "senders" condition, or a router
5543 "senders" option, similar to $domain_data and $local_part_data.
5545 6. Moved the writing of debug_print from before to after the "senders" test
5548 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5549 problems for message scanning, either using a data ACL, or using
5550 local_scan() because the Received: header was not generated till after they
5551 were called (in order to set the time as the time of reception completion).
5552 I have revised the way this works. The header is now generated after the
5553 body is received, but before the ACL or local_scan() are called. After they
5554 are run, the timestamp in the header is updated.
5560 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5561 before starting a queue runner without re-exec. This happened only when
5562 deliver_drop_privilege was set or when the Exim user was set to root. The
5563 effect of the bug was that timeouts during subsequent deliveries caused
5564 crashes instead of being properly handled. The handler is now left at its
5565 default (and expected) setting.
5567 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5568 message, again when deliver_drop_privilege is set or Exim is run as root.
5569 The bug described in (1) was not present in this case, but the tidying up
5570 of the other signals was missing. I have made the two cases consistent.
5572 3. The ignore_target_hosts setting on a manualroute router was being ignored
5573 for hosts that were looked up using the /MX notation.
5575 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5578 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5579 operated on the sender address. After changing the $sender_address to <>
5580 for the sender address verify, Exim was re-instated it as the original
5581 (before rewriting) address, but remembering that it had rewritten it, so it
5582 wasn't rewriting it again. This bug also had the effect of breaking the
5583 sender address verification caching when the sender address was rewritten.
5585 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5586 This has been changed so that if the ip literal address matches
5587 ignore_target_hosts, the router declines.
5589 7. Added expansion conditions match_domain, match_address, and match_local_
5590 part (NOT match_host).
5592 8. The placeholder for the Received: header didn't have a length field set.
5594 9. Added code to Exim itself and to exim_lock to test for a specific race
5595 condition that could lead to file corruption when using MBX delivery. The
5596 issue is with the lockfile that is created in /tmp. If this file is removed
5597 after a process has opened it but before that process has acquired a lock,
5598 there is the potential for a second process to recreate the file and also
5599 acquire a lock. This could lead to two Exim processes writing to the file
5600 at the same time. The added code performs the same test as UW imapd; it
5601 checks after acquiring the lock that its file descriptor still refers to
5602 the same named file.
5604 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5605 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5608 11. Added the smtp_active_hostname option. If used, this will typically be made
5609 to depend on the incoming interface address. Because $interface_address is
5610 not set up until the daemon has forked a reception process, error responses
5611 that can happen earlier (such as "too many connections") no longer contain
5614 12. If an expansion in a condition on a "warn" statement fails because a lookup
5615 defers, the "warn" statement is abandoned, and the next ACL statement is
5616 processed. Previously this caused the whole ACL to be aborted.
5618 13. Added the iplsearch lookup type.
5620 14. Added ident_timeout as a log selector.
5622 15. Added tls_certificate_verified as a log selector.
5624 16. Added a global option tls_require_ciphers (compare the smtp transport
5625 option of the same name). This controls incoming TLS connections.
5627 17. I finally figured out how to make tls_require_ciphers do a similar thing
5628 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5629 before starting the TLS session.
5631 18. Tabs are now shown as \t in -bP output.
5633 19. If the log selector return_path_on_delivery was set, Exim crashed when
5634 bouncing a message because it had too many Received: header lines.
5636 20. If two routers both had headers_remove settings, and the first one included
5637 a superfluous trailing colon, the final name in the first list and the
5638 first name in the second list were incorrectly joined into one item (with a
5639 colon in the middle).
5645 1. Added -C and -D options to the exinext utility, mainly to make it easier
5646 to include in the automated testing, but these could be helpful when
5647 multiple configurations are in use.
5649 2. The exinext utility was not formatting the output nicely when there was
5650 an alternate port involved in the retry record key, nor when there was a
5651 message id as well (for retries that were specific to a specific message
5652 and a specific host). It was also confused by IPv6 addresses, because of
5653 the additional colons they contain. I have fixed the IPv4 problem, and
5654 patched it up to do a reasonable job for IPv6.
5656 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5657 delivery, the log line now contains "pipelined" if PIPELINING was used.
5659 4. An SMTP transport process used to panic and die if the bind() call to set
5660 an explicit outgoing interface failed. This has been changed; it is now
5661 treated in the same way as a connect() failure.
5663 5. A reference to $sender_host_name in the part of a conditional expansion
5664 that was being skipped was still causing a DNS lookup. This no longer
5667 6. The def: expansion condition was not recognizing references to header lines
5668 that used bh_ and bheader_.
5670 7. Added the _cache feature to named lists.
5672 8. The code for checking quota_filecount in the appendfile transport was
5673 allowing one more file than it should have been.
5675 9. For compatibility with Sendmail, the command line option
5683 and sets the incoming protocol and host name (for trusted callers). The
5684 host name and its colon can be omitted when only the protocol is to be set.
5685 Note the Exim already has two private options, -pd and -ps, that refer to
5686 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5687 "s", but I don't think that's a major issue.
5689 10. A number of refactoring changes to the code, none of which should affect
5692 (a) The number of logging options was getting close to filling up the
5693 32-bit word that was used as a bit map. I have split them into two classes:
5694 those that are passed in the argument to log_write(), and those that are
5695 only ever tested independently outside of that function. These are now in
5696 separate 32-bit words, so there is plenty of room for expansion again.
5697 There is no change in the user interface or the logging behaviour.
5699 (b) When building, for example, log lines, the code previously used a
5700 macro that called string_cat() twice, in order to add two strings. This is
5701 not really sufficiently general. Furthermore, there was one instance where
5702 it was actually wrong because one of the argument was used twice, and in
5703 one call a function was used. (As it happened, calling the function twice
5704 did not affect the overall behaviour.) The macro has been replaced by a
5705 function that can join an arbitrary number of extra strings onto a growing
5708 (c) The code for expansion conditions now uses a table and a binary chop
5709 instead of a serial search (which was left over from when there were very
5710 few conditions). Also, it now recognizes conditions like "pam" even when
5711 the relevant support is not compiled in: a suitably worded error message is
5712 given if an attempt is made to use such a condition.
5714 11. Added ${time_interval:xxxxx}.
5716 12. A bug was causing one of the ddress fields not to be passed back correctly
5717 from remote delivery subprocesses. The field in question was not being
5718 subsequently used, so this caused to problems in practice.
5720 13. Added new log selectors queue_time and deliver_time.
5722 14. Might have fixed a bug in maildirsizefile handling that threw up
5723 "unexpected character" debug warnings, and recalculated the data
5724 unnecessarily. In any case, I expanded the warning message to give more
5727 15. Added the message "Restricted characters in address" to the statements in
5728 the default ACL that block characters like @ and % in local parts.
5730 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5731 Three changes have been made:
5733 (a) There was a serious bug; a negative response to MAIL caused the whole
5734 recipient domain to be cached as invalid, thereby blocking all messages
5735 to all local parts at the same domain, from all senders. This bug has
5736 been fixed. The domain is no longer cached after a negative response to
5737 MAIL if the sender used is not empty.
5739 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5742 (c) A new callout option, "use_sender" has been added for people who want
5743 the modified behaviour.
5749 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5752 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5753 indeed breaks things for older releases.
5755 3. Added additional logging to the case where there is a problem reading data
5756 from a filter that is running in a subprocess using a pipe, in order to
5757 try to track down a specific problem.
5759 4. Testing facility fudge: when running in the test harness and attempting
5760 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5761 getting "No route to host". Convert this to a timeout.
5763 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5766 6. Some OS don't have socklen_t but use size_t instead. This affects the
5767 fifth argument of getsockopt() amongst other things. This is now
5768 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5769 can be set for individual OS. I have set it for SunOS5, OSF1, and
5770 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5771 some earlier ones do not.
5773 7. Change 4.30/15 was not doing the test caselessly.
5775 8. The standard form for an IPv6 address literal was being rejected by address
5776 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5777 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5778 this, as well as the form without the "IPv6" on the front (but only when
5779 address literals are enabled, of course).
5781 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5783 10. Exim crashed if a message with an empty sender address specified by -f
5784 encountered a router with an errors_to setting. This could be provoked only
5785 by a command such as
5789 where an empty string was supplied; "<>" did not hit this bug.
5791 11. Installed PCRE release 4.5.
5793 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5794 remained set. It is now erased.
5796 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5797 times from message ids (which are base 36 rather than the normal 62).
5799 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5800 were being counted as actual protocol errors, and logged if the log
5801 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5802 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5803 and DATA following a set of rejected RCPTs do not count as protocol errors.
5804 In other words, Exim assumes they were pipelined, though this may not
5805 actually be the case. Of course, in all cases the client gets an
5806 appropriate error code.
5808 15. If a lookup fails in an ACL condition, a message about the failure may
5809 be available; it is used if testing the ACL cannot continue, because most
5810 such messages specify what the cause of the deferral is. However, some
5811 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5812 that caused an old message to be retained and used if a later statement
5813 caused a defer, replacing the real cause of the deferral.
5815 16. If an IP address had so many PTR records that the DNS lookup buffer
5816 was not large enough to hold them, Exim could crash while trying to process
5817 the truncated data. It now detects and logs this case.
5819 17. Further to 4.21/58, another change has been made: if (and only if) the
5820 first line of a message (the first header line) ends with CRLF, a bare LF
5821 in a subsequent header line has a space inserted after it, so as not to
5822 terminate the header.
5824 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5825 unnecessarily, used atoi() instead of strtol(), and didn't check the
5826 termination when getting file sizes from file names by regex.
5828 19. Completely re-implemented the support for maildirsize files, in the light
5829 of a number of problems with the previous contributed implementation
5830 (4.30/29). In particular:
5832 . If the quota is zero, the maildirsize file is maintained, but no quota is
5835 . If the maildir directory does not exist, it is created before any attempt
5836 to write a maildirsize file.
5838 . The quota value in the file is just a cache; if the quota is changed in
5839 the transport, the new value overrides.
5841 . A regular expression is available for excluding directories from the
5844 20. The autoreply transport checks the characters in options that define the
5845 message's headers; it allows continued headers, but it was checking with
5846 isspace() after an embedded newline instead of explicitly looking for a
5849 21. If all the "regular" hosts to which an address was routed had passed their
5850 expiry times, and had not reached their retry times, the address was
5851 bounced, even if fallback hosts were defined. Now Exim should go on to try
5854 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5855 equivalent code in the SMTP transport. Some hosts send humungous responses
5856 to HELO/EHLO, more than 1024 it seems.
5858 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5859 gives compiler warnings in some environments. I've now done it "properly",
5862 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5863 (because of problems with the built-in one) was declared to return uschar *
5864 instead of char *, causing compiler failure.
5866 25. Fixed a file descriptor leak when processing alias/forward files.
5868 26. Fixed a minor format string issue in dbfn.c.
5870 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5872 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5873 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5876 29. When a sender address is verified, it is cached, to save repeating the test
5877 when there is more than one recipient in a message. However, when the
5878 verification involves a callout, it is possible for different callout
5879 options to be set for different recipients. It is too complicated to keep
5880 track of this in the cache, so now Exim always runs a verification when a
5881 callout is required, relying on the callout cache for the optimization.
5882 The overhead is duplication of the address routing, but this should not be
5885 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5886 to be verified with callout=postmaster, and the main callout worked but the
5887 postmaster check failed, the verification correctly failed. However, if a
5888 subsequent RCPT command asked for sender verification *without* the
5889 postmaster check, incorrect caching caused this verification also to fail,
5892 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5893 it was not caching the DNS options (qualify_single, search_parents) that
5894 were used when the lookup failed. A subsequent lookup with different
5895 options therefore always gave the same answer, though there were cases
5896 where it should not have. (Example: a "domains = !$mx_any" option on a
5897 dnslookup router: the "domains" option is always processed without any
5898 widening, but the router might have qualify_single set.) Now Exim uses the
5899 cached value only when the same options are set.
5901 32. Added John Jetmore's "exipick" utility to the distribution.
5903 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5904 than a timeout (e.g. a certificate is required, and is not provided), an
5905 Exim server now closes the connection immediately. Previously it waited for
5906 the client to close - but if the client is SSL, it seems that they each
5907 wait for each other, leading to a delay before one of them times out.
5909 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5910 maintained 0.8.x compatibility because I don't think many are using it, and
5911 it is clearly obsolete.
5913 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5916 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5917 client certificate was expired. A simple patch fixes this, though I don't
5918 understand the full logic of why the verify callback is called multiple
5921 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5922 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5923 which causes problems with some clients (such as the Certicom SSL Plus
5924 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5925 disables the coutermeasure allowing Eudora to connect."
5927 38. Exim was not checking that a write() to a log file succeeded. This could
5928 lead to Bad Things if a log got too big, in particular if it hit a file
5929 size limit. Exim now panics and dies if it cannot write to a log file, just
5930 as it does if it cannot open a log file.
5932 39. Modified OS/Makefile-Linux so that it now contains
5934 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5936 The two -D definitions ensure that Exim is compiled with large file
5937 support, which makes it possible to handle log files that are bigger than
5940 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5941 instance) a domain was checked against a named list that involved a lookup,
5942 causing $domain_data to be set, then another domain was checked against the
5943 same list, then the first domain was re-checked, the value of $domain_data
5944 after the final check could be wrong. In particular, if the second check
5945 failed, it could be set empty. This bug probably also applied to
5948 41. The strip_trailing_dot option was not being applied to the address given
5949 with the -f command-line option.
5951 42. The code for reading a message's header from the spool was incrementing
5952 $received_count, but never initializing it. This meant that the value was
5953 incorrect (doubled) while delivering a message in the same process in which
5954 it was received. In the most common configuration of Exim, this never
5955 happens - a fresh exec is done - but it can happen when
5956 deliver_drop_privilege is set.
5958 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5959 it now includes up to 150 characters of the unexpected data in the log
5962 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5963 and building data strings. The size of both of these buffers was 10 000
5964 bytes - far larger than anybody would *ever* want, thought I. Needless to
5965 say, somebody hit the limit. I have increased the maximum line length to
5966 20 000 and the maximum data length of concatenated lines to 100 000. I have
5967 also fixed two bugs, because there was no checking on these buffers. Tsk,
5968 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5969 buffer is too small.
5971 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5972 lsearch lookups. Now it does.
5974 46. When parsing a route_list item in a manualroute router, a fixed-length
5975 buffer was used for the list of hosts. I made this 1024 bytes long,
5976 thinking that nobody would ever have a list of hosts that long. Wrong.
5977 Somebody had a whole pile of complicated expansion conditions, and the
5978 string was silently truncated, leading to an expansion error. It turns out
5979 that it is easier to change to an unlimited length (owing to other changes
5980 that have happened since this code was originally written) than to build
5981 structure for giving a limitation error. The length of the item that
5982 expands into the list of hosts is now unlimited.
5984 47. The lsearch lookup could not handle data where the length of text line was
5985 more than 4095 characters. Such lines were truncated, leading to shortened
5986 data being returned. It should now handle lines of any length.
5988 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5989 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5992 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5993 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5994 no longer happens in comment lines. A list of replacements is now placed
5995 at the head of all of the source files, except those whose only change is
5996 to replace PERL_COMMAND in the very first #! line.
5998 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5999 messages on the queue, with a bottom-up merge sort, using code contributed
6000 by Michael Haardt. This should make operations like -bp somewhat faster on
6001 large queues. It won't affect queue runners, except when queue_run_in_order
6004 51. Installed eximstats 1.31 in the distribution.
6006 52. Added support for SRV lookups to the dnslookup router.
6008 53. If an ACL referred to $message_body or $message_body_end, the value was not
6009 reset for any messages that followed in the same SMTP session.
6011 54. The store-handling optimization for building very long strings was not
6012 differentiating between the different store pools. I don't think this
6013 actually made any difference in practice, but I've tidied it.
6015 55. While running the routers to verify a sender address, $sender_address
6016 was still set to the sender address. This is wrong, because when routing to
6017 send a bounce to the sender, it would be empty. Therefore, I have changed
6018 it so that, while verifying a sender address, $sender_address is set to <>.
6019 (There is no change to what happens when verifying a recipient address.)
6021 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6022 target A or AAAA records (if not already returned) without resetting the
6023 qualify_single or search_parents options of the DNS resolver. These are
6024 inappropriate in this case because the targets of MX and SRV records must
6025 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6026 target that, when qualified, matched something in the local domain. These
6027 two options are now turned off when doing these lookups.
6029 57. It seems that at least some releases of Reiserfs (which does not have the
6030 concept of a fixed number of inodes) returns zero and not -1 for the
6031 number of available inodes. This interacted badly with check_spool_inodes,
6032 which assumed that -1 was the "no such thing" setting. What I have done is
6033 to check that the total number of inodes is greater than zero before doing
6034 the test of how many are available.
6036 58. When a "warn" ACL statement has a log_message modifier, the message is
6037 remembered, and not repeated. This is to avoid a lot of repetition when a
6038 message has many recipients that cause the same warning to be written.
6039 However, Exim was preserving the list of already written lines for an
6040 entire SMTP session, which doesn't seem right. The memory is now reset if a
6041 new message is started.
6043 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6044 debugging output unless log_rewrite was also set.
6046 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6047 of (int)(handle) when we know that handle contains (void *)(-1).
6049 61. The Exim daemon panic-logs an error return when it closes the incoming
6050 connection. However "connection reset by peer" seems to be common, and
6051 isn't really an error worthy of noting specially, so that particular error
6054 62. When Exim is trying to find all the local interfaces, it used to panic and
6055 die if the ioctl to get the interface flags failed. However, it seems that
6056 on at least one OS (Solaris 9) it is possible to have an interface that is
6057 included in the list of interfaces, but for which you get a failure error
6058 for this call. This happens when the interface is not "plumbed" into a
6059 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6060 failure of the "get flags" call assumes that the interface is down.
6062 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6063 makes life easier for people who are doing arithmetic on fields extracted
6064 from dates, where you often get leading zeros that should not be
6065 interpreted as octal.
6067 64. Added qualify_domain to the redirect router, to override the global
6070 65. If a pathologically long header line contained very many addresses (the
6071 report of this problem mentioned 10 000) and each of them was rewritten,
6072 Exim could use up a very large amount of memory. (It kept on making new
6073 copies of the header line as it rewrote, and never released the old ones.)
6074 At the expense of a bit more processing, the header rewriting function has
6075 been changed so that it no longer eats memory in this way.
6077 66. The generation of the Received: header has been moved from the time that a
6078 message starts to be received, to the time that it finishes. The timestamp
6079 in the Received: header should now be very close to that of the <= log
6080 line. There are two side-effects of this change:
6082 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6083 logged header lines no longer include the local Received: line, because
6084 it has not yet been created. The same applies to a copy of the message
6085 that is returned to a non-SMTP sender when a message is rejected.
6087 (b) When a filter file is tested using -bf, no additional Received: header
6088 is added to the test message. After some thought, I decided that this
6091 This change does not affect the value of $received_for. It is still set
6092 after address rewriting, but before local_scan() is called.
6094 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6096 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6097 gave an unhelpful panic error message, and a defer error. I have managed to
6098 change this behaviour so that it now rejects any supplied certificate,
6099 which seems right, as the list of acceptable certificates is empty.
6101 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6102 gave an unhelpful defer error. I have not managed to make this reject any
6103 supplied certificates, but the error message it gives is "no certificate
6104 supplied", which is not helpful.
6106 70. exigrep's output now also includes lines that are not associated with any
6107 message, but which match the given pattern. Implemented by a patch from
6108 Martin Sluka, which also tidied up the Perl a bit.
6110 71. Recipient callout verification, like sender verification, was using <> in
6111 the MAIL FROM command. This isn't really the right thing, since the actual
6112 sender may affect whether the remote host accepts the recipient or not. I
6113 have changed it to use the actual sender in the callout; this means that
6114 the cache record is now keyed on a recipient/sender pair, not just the
6115 recipient address. There doesn't seem to be a real danger of callout loops,
6116 since a callout by the remote host to check the sender would use <>.
6117 [SEE ABOVE: changed after hitting problems.]
6119 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6120 temporary errors. However, in the case of such a code being given after
6121 the end of a data transmission (i.e. after ".") Exim was failing to write
6122 a retry record for the message. (Yes, there was some broken host that was
6123 actually sending 8xx at this point.)
6125 73. An unknown lookup type in a host list could cause Exim to panic-die when
6126 the list was checked. (An example that provoked this was putting <; in the
6127 middle of a list instead of at the start.) If this happened during a DATA
6128 ACL check, a -D file could be left lying around. This kind of configuration
6129 error no longer causes Exim to die; instead it causes a defer error. The
6130 incident is still logged to the main and panic logs.
6132 74. Buglet left over from Exim 3 conversion. The message "too many messages
6133 in one connection" was written to the rejectlog but not the mainlog, except
6134 when address rewriting (yes!) was being logged.
6136 75. Added write_rejectlog option.
6138 76. When a system filter was run not as root (that is, when system_filter_user
6139 was set), the values of the $n variables were not being returned to the
6140 main process; thus, they were not subsequently available in the $sn
6143 77. Added +return_path_on_delivery log selector.
6145 78. A connection timeout was being treated differently from recipients deferred
6146 when testing hosts_max_try with a message that was older than the host's
6147 retry timeout. (The host should not be counted, thus allowing all hosts to
6148 be tried at least once before bouncing.) This may have been the cause of an
6149 occasionally reported bug whereby a message would remain on the queue
6150 longer than the retry timeout, but would be bounced if a delivery was
6151 forced. I say "may" because I never totally pinned down the problem;
6152 setting up timeout/retry tests is difficult. See also the next item.
6154 79. The ultimate address timeout was not being applied to errors that involved
6155 a combination of host plus message (for example, a timeout on a MAIL
6156 command). When an address resolved to a number of possible hosts, and they
6157 were not all tried for each delivery (e.g. because of hosts_max_try), a
6158 message could remain on the queue longer than the retry timeout.
6160 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6163 81. Fixed an obscure SMTP outgoing bug which required at least the following
6164 conditions: (a) there was another message waiting for the same server;
6165 (b) the server returned 5xx to all RCPT commands in the first message so
6166 that the message was not completed; (c) the server dropped the connection
6167 or gave a negative response to the RSET that Exim sends to abort the
6168 transaction. The observed case was a dropped connection after DATA that had
6169 been sent in pipelining mode. That is, the server had advertised PIPELINING
6170 but was not implementing it correctly. The effect of the bug was incorrect
6171 behaviour, such as trying another host, and this could lead to a crash.
6177 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6178 and daemon.c were passed as pointers to ints; they should have been
6179 pointers to socklen_t variables (which are typically unsigned ints).
6181 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6184 3. Fixed a really odd bug that affected only the testing scheme; patching a
6185 certain fixed string in the binary changed the value of another string that
6186 happened to be identical to the end of the original first string.
6188 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6189 name", it returns that address as the IP address. On some operating
6190 systems (e.g. Solaris), it also passes back the IP address string as the
6191 "host name". However, on others (e.g. Linux), it passes back an empty
6192 string. Exim wasn't checking for this, and was changing the host name to an
6193 empty string, assuming it had been canonicalized.
6195 5. Although rare, it is permitted to have more than one PTR record for a given
6196 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6197 all the names associated with an address, because they do in Solaris.
6198 However, it seems that they do not in Linux for data that comes from the
6199 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6200 I found this out when I moved to a new Linux workstation and tried to run
6201 the Exim test suite.
6203 To get round this problem I have changed the code so that it now does its
6204 own call to the DNS to look up PTR records when searching for a host name.
6205 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6206 addresses that are only in /etc/hosts are still found.
6208 This behaviour is, however, controlled by an option called host_lookup_
6209 order, which defaults to "bydns:byaddr". If people want to use the other
6210 order, or indeed, just use one or the other means of lookup, they can
6211 specify it in this variable.
6213 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6214 some operating systems, this comes back from gethostbyaddr() as an empty
6215 string, and this is what Exim used to test for. However, it seems that in
6216 other systems, "." is yielded. Exim now tests for this case too.
6218 7. The values of check_spool_space and check_log_space are now held internally
6219 as a number of kilobytes instead of an absolute number of bytes. If a
6220 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6221 kilobyte. This means that much larger values can be stored.
6223 8. Exim monitor: an attempt to get the action menu when not actually pointing
6224 at a message produces an empty menu entitled "No message selected". This
6225 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6226 no entries in it ("Shell widget menu has zero width and/or height"). So I
6227 have added a single, blank menu entry in this case.
6229 9. Added ${quote_local_part.
6231 10. MIME decoding is now applied to the contents of Subject: header lines when
6234 11. Now that a reference to $sender_host_address automatically causes a reverse
6235 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6236 host lookup before query-style lookups in lists that might use this
6237 variable. This has therefore been abolished, and the "net-" prefix is no
6238 longer necessary for query-style lookups.
6240 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6241 have been a typo for LFLAGS, so it has been changed.
6243 13. The install script calls Exim with "-C /dev/null" in order to find the
6244 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6245 to be output. However, since Exim outputs its version number before the
6246 error, it didn't break the script. It just looked ugly. I fixed this by
6247 always allowing "-C /dev/null" if the caller is root.
6249 14. Ignore overlarge ACL variable number when reading spool file - insurance
6250 against a later release with more variables having written the file.
6252 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6253 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6254 this, as well as the form without the "IPv6" on the front.
6256 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6257 OS/Makefile-Darwin file.
6259 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6260 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6262 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6263 variable. However, some OS use ints for inodes. I've added cast to long int
6264 to get rid of the compiler warning.
6266 19. I had forgotten to lock out "/../" in configuration file names when
6267 ALT_CONFIG_PREFIX was set.
6269 20. Routers used for verification do not need to specify transports. However,
6270 if such a router generated a host list, and callout was configured, Exim
6271 crashed, because it could not find a port number from the (non-existent)
6272 transport. It now assumes port 25 in this circumstance.
6274 21. Added the -t option to exigrep.
6276 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6277 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6278 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6279 them is set, LOOKUP_LSEARCH is forced.
6281 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6283 24. Added sender and host information to the "rejected by local_scan()" log
6284 line; previously there was no indication of these.
6286 25. Added .include_if_exists.
6288 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6289 Linux. It turns out that not all file systems support this. Apparently some
6290 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6291 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6292 not supported on the file descriptor, is now ignored when Exim is trying to
6293 sync a directory. This applies only to Linux.
6295 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6297 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6298 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6301 29. Added support for maildirsize files from supplied patch (modified a bit).
6303 30. The use of :fail: followed by an empty string could lead Exim to respond to
6304 sender verification failures with (e.g.):
6306 550 Verification failed for <xxx>
6307 550 Sender verify failed
6309 where the first response line was missing the '-' that indicates it is not
6310 the final line of the response.
6312 31. The loop for finding the name of the user that called Exim had a hardwired
6313 limit of 10; it now uses the value of finduser_retries, which is used for
6314 all other user lookups.
6316 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6319 33. Exim was neglecting to zero errno before one call of strtol() when
6320 expanding a string and expecting an integer value. On some systems this
6321 resulted in spurious "integer overflow" errors. Also, it was casting the
6322 result into an int without checking.
6324 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6325 did not work. The code looks as if it has *never* worked, though it appears
6326 to have been documented since at least release 1.62. I have made it work.
6328 35. The "timeout_DNS" error in retry rules, also documented since at least
6329 1.62, also never worked. As it isn't clear exactly what this means, and
6330 clearly it isn't a major issue, I have abolished the feature by treating it
6331 as "timeout", and writing a warning to the main and panic logs.
6333 36. The display of retry rules for -brt wasn't always showing the error code
6336 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6337 timeout_connect_A, timeout_connect_MX.
6339 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6340 to the empty sender.
6342 39. The daemon was not analysing the content of -oX till after it had closed
6343 stderr and disconnected from the controlling terminal. This meant that any
6344 syntax errors were only noted on the panic log, and the return code from
6345 the command was 0. By re-arranging the code a little, I've made the
6346 decoding happen first, so such errors now appear on stderr, and the return
6347 code is 1. However, the actual setting up of the sockets still happens in
6348 the disconnected process, so errors there are still only recorded on the
6351 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6352 connections (as happens on some IP stacks) was logged at start up time as
6353 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6354 it from "IPv6 and IPv4", which means that two separate sockets are being
6357 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6358 says whether AF_INET or AF_INET6 was passed as an argument.
6360 42. Exiwhat output was messed up when time zones were included in log
6363 43. Exiwhat now gives more information about the daemon's listening ports,
6364 and whether -tls-on-connect was used.
6366 44. The "port" option of the smtp transport is now expanded.
6368 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6369 silently ignored. Now an error message is written to the main and panic
6372 46. There's a new ACL modifier called "logwrite" which writes to a log file
6373 as soon as it is encountered.
6375 47. Added $local_user_uid and $local_user_gid at routing time.
6377 48. Exim crashed when trying to verify a sender address that was being
6380 49. Exim was recognizing only a space character after ".include". It now also
6381 recognizes a tab character.
6383 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6384 extracting the relevant information from the specification. The man page no
6385 longer contains scrambled data for the -d option, and I've added a section
6386 at the front about calling Exim under different names.
6388 51. Added "extra_headers" argument to the "mail" command in filter files.
6390 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6393 53. Installed eximstats 1.29.
6395 54. Added transport_filter_timeout as a generic transport option.
6397 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6398 Cc: header lines. This was required by RFC 822, but it not required by RFC
6401 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6402 incoming messages that did not have them. Now it does so only if the
6403 message originates locally, that is, if there is no associated remote host
6404 address. When Resent- header lines are present, this applies to the Resent-
6405 lines rather than the non-Resent- lines.
6407 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6408 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6410 58. Messages for configuration errors now include the name of the main
6411 configuration file - useful now that there may be more than one file in a
6412 list (.included file names were always shown).
6414 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6415 for those rare installations that do not start the daemon as root or run it
6416 setuid root. I've cut out the call to initgroups() if the daemon is not
6419 60. The Exim user and group can now be bound into the binary as text strings
6420 that are looked up at the start of Exim's processing.
6422 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6424 62. Added $mailstore_basename variable.
6426 63. Installed patch to sieve.c from Michael Haardt.
6428 64. When Exim failed to open the panic log after failing to open the main log,
6429 the original message it was trying to log was written to stderr and debug
6430 output, but if they were not available (the usual case in production), it
6431 was lost. Now it is written to syslog before the two lines that record the
6432 failures to open the logs.
6434 65. Users' Exim filters run in subprocesses under the user's uid. It is
6435 possible for a "deliver" command or an alias in a "personal" command to
6436 provoke an address rewrite. If logging of address rewriting is configured,
6437 this fails because the process is not running as root or exim. There may be
6438 a better way of dealing with this, but for the moment (because 4.30 needs
6439 to be released), I have disabled address rewrite logging when running a
6440 filter in a non-root, non-exim process.
6446 1. The buildconfig auxiliary program wasn't quoting the value set for
6447 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6448 not defined. This bug was masked in 4.22 by the effect that was fixed in
6451 2. Some messages that were rejected after a message id was allocated were
6452 shown as "incomplete" by exigrep. It no longer does this for messages that
6453 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6455 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6456 have allow_domain_literals set, the ID did not get logged in the <= line.
6457 Domain literals are now always recognized in Message-ID: header lines.
6459 4. The first argument for a ${extract expansion item is the key name or field
6460 number. Leading and trailing spaces in this item were not being ignored,
6461 causing some misleading effects.
6463 5. When deliver_drop_privilege was set, single queue runner processes started
6464 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6465 same command in the process it spins off) were not dropping privilege.
6467 6. When the daemon running as "exim" started a queue runner, it always
6468 re-executed Exim in the spun-off process. This is a waste of effort when
6469 deliver_drop_privilege is set. The new process now just calls the
6470 queue-runner function directly.
6476 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6479 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6480 ignored. Though the use of -oP was forcing the writing of a pid file, it
6481 was always written to the default place.
6483 3. If the message "no IP address found for host xxxx" is generated during
6484 incoming verification, it is now followed by identification of the incoming
6485 connection (so you can more easily find what provoked it).
6487 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6489 5. Added some features to "harden" Exim a bit more against certain attacks:
6491 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6492 be put in Local/Makefile. This is like the never_users runtime option,
6493 but it cannot be overridden. The default setting is "root".
6495 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6496 prefix string with which any file named in a -C command line option
6499 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6500 is retained for -C and -D only if the caller of Exim is root. Without
6501 it, the exim user may also use -C and -D and retain privilege.
6503 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6504 command line option is disabled.
6506 6. Macro names set by the -D option must start with an upper case letter, just
6507 like macro names defined in the configuration file.
6509 7. Added "dereference=" facility to LDAP.
6511 8. Two instances of the typo "uknown" in the source files are fixed.
6513 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6514 the Configure-Makefile script screwed up while processing it.
6516 10. Incorporated PCRE 4.4.
6518 11. The SMTP synchronization check was not operating right at the start of an
6519 SMTP session. For example, it could not catch a HELO sent before the client
6520 waited for the greeting. There is now a check for outstanding input at the
6521 point when the greeting is written. Because of the duplex, asynchronous
6522 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6523 way, but not yet received, when the check is performed.
6525 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6526 on TCP/IP sockets, because this apparently causes some broken clients to
6529 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6530 unchanged) from the Cygwin maintainer.
6532 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6533 was supported instead of testing for mbx. Effectively a typo.
6535 15. The spa authenticator server code was not checking that the input it
6536 received was valid base64.
6538 16. The debug output line for the "set" modifier in ACLs was not showing the
6539 name of the variable that was being set.
6541 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6543 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6544 lookup on its own. Something else was needed to trigger the lookup. For
6545 example, a match in host_lookup or the need for a host name in a host list.
6546 Now, if $sender_host_name is referenced and the host name has not yet been
6547 looked up, a lookup is performed. If the lookup fails, the variable remains
6548 empty, and $host_lookup_failed is set to "1".
6550 19. Added "eqi" as a case-independent comparison operator.
6552 20. The saslauthd authentication condition could segfault if neither service
6553 nor realm was specified.
6555 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6556 error message that was logged was misleading, and incoming SMTP
6557 connections were dropped. The message is now more accurate, and temporary
6558 errors are given to SMTP connections.
6560 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6561 (except RSET) with a 421 error, until QUIT is received. However, it was
6562 failing to send a response to QUIT.
6564 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6565 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6566 ACL is now run after the helo_try_verify_hosts code.
6568 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6569 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6570 case-independent, so other case variants are also recognized). Apparently
6571 some systems use these upper case variants.
6573 25. If more than two messages were waiting for the same host, and a transport
6574 filter was specified for the transport, Exim sent two messages over the
6575 same TCP/IP connection, and then failed with "socket operation on non-
6576 socket" when it tried to send the third.
6578 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6580 27. The extern definition of crypt16() in expand.c was not being excluded when
6581 the OS had its own crypt16() function.
6583 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6584 as a preferred synonym for return_size_limit, both as an option and as an
6587 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6589 30. Changed the default configuration ACL to relax the local part checking rule
6590 for addresses that are not in any local domains. For these addresses,
6591 slashes and pipe symbols are allowed within local parts, but the sequence
6592 /../ is explicitly forbidden.
6594 31. SPA server authentication was not clearing the challenge buffer before
6597 32. log_message in a "warn" ACL statement was writing to the reject log as
6598 well as to the main log, which contradicts the documentation and doesn't
6599 seem right (because no rejection is happening). So I have stopped it.
6601 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6602 However, I am unable to do any testing of this.
6604 34. Fixed an infelicity in the appendfile transport. When checking directories
6605 for a mailbox, to see if any needed to be created, it was accidentally
6606 using path names with one or more superfluous leading slashes; tracing
6607 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6609 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6610 contents are added to the log line that is written for every discarded
6611 recipient. (Previously a log_message setting was ignored.)
6613 36. The ${quote: operator now quotes the string if it is empty.
6615 37. The install script runs exim in order to find its version number. If for
6616 some reason other than non-existence or emptiness, which it checks, it
6617 could not run './exim', it was installing it with an empty version number,
6618 i.e. as "exim-". This error state is now caught, and the installation is
6621 38. An argument was missing from the function that creates an error message
6622 when Exim fails to connect to the socket for saslauthd authentication.
6623 This could cause Exim to crash, or give a corrupted message.
6625 39. Added isip, isip4, and isip6 to ${if conditions.
6627 40. The ACL variables $acl_xx are now saved with the message, and can be
6628 accessed later in routers, transports, and filters.
6630 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6631 strings in the file are not string-expanded.
6633 42. If a MAIL command specified a SIZE value that was too large to fit into an
6634 int variable, the check against message_size_limit failed. Such values are
6635 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6636 day this will have to be increased, but I don't think I want to be around
6637 when emails are that large.
6644 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6645 iconv() is not standard in FreeBSD.
6647 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6648 IPv6 enabled. The observed symptom was a segmentation fault on return from
6649 the function os_common_find_running_interfaces() in src/os.c.
6651 3. In the check_special_case() function in daemon.c I had used "errno" as an
6652 argument name, which causes warnings on some systems. This was basically a
6653 typo, since it was named "eno" in the comments!
6655 4. The code that waits for the clock to tick (at a resolution of some fraction
6656 of a second) so as to ensure message-id uniqueness was always waiting for
6657 at least one whole tick, when it could have waited for less. [This is
6658 almost certainly not relevant at current processor speeds, where it is
6659 unlikely to ever wait at all. But we try to future-proof.]
6661 5. The function that sleeps for a time interval that includes fractions of a
6662 second contained a race. It did not block SIGALRM between setting the
6663 timer, and suspending (a couple of lines later). If the interval was short
6664 and the sigsuspend() was delayed until after it had expired, the suspension
6665 never ended. On busy systems this could lead to processes getting stuck for
6668 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6669 process, before it forks any delivery processes. The open lookup caching
6670 mechanism meant that the open file or database connection was passed into
6671 the delivery process. The problem was that delivery processes always tidy
6672 up cached lookup data. This could cause a problem for the next delivery
6673 process started by the queue runner, because the external queue runner
6674 process does not know about the closure. So the next delivery process
6675 still has data in the lookup cache. In the case of a file lookup, there was
6676 no problem because closing a file descriptor in a subprocess doesn't affect
6677 the parent. However, if the lookup was caching a connection to a database,
6678 the connection was closed, and the second delivery process was likely to
6679 see errors such as "PGSQL: query failed: server closed the connection
6680 unexpectedly". The problem has been fixed by closing all cached lookups
6681 in a queue runner before running a delivery process.
6683 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6684 seem to have the "const" qualifier which it has on other OS. I've
6687 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6688 *of the same type* (client or server) with the same public name that an
6689 error should be diagnosed.
6691 9. When Exim looked up a host name for an IP address, but failed to find the
6692 original IP address when looking up the host name (a safety check), it
6693 output the message "<ip address> does not match any IP for NULL", which was
6694 confusing, to say the least. The bug was that the host name should have
6695 appeared instead of "NULL".
6697 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6698 user that is built into the binary, and the -C or -D options is used, root
6699 privilege is dropped before the configuration file is read. In addition,
6700 logging is switched to stderr instead of the normal log files. If the
6701 configuration then re-defines the Exim user, the unprivileged environment
6702 is probably not what is expected, so Exim logs a panic warning message (but
6705 However, if deliver_drop_privilege is set, the unprivileged state may well
6706 be exactly what is intended, so the warning has been cut out in that case,
6707 and Exim is allowed to try to write to its normal log files.
6713 1. smtp_return_error_details was not giving details for temporary sender
6714 or receiver verification errors.
6716 2. Diagnose a configuration error if two authenticators have the same public
6719 3. Exim used not to create the message log file for a message until the first
6720 delivery attempt. This could be confusing when incoming messages were held
6721 for policy or load reasons. The message log file is now created at the time
6722 the message is received, and an initial "Received" line is written to it.
6724 4. The automatically generated man page for command line options had a minor
6725 bug that caused no ill effects; however, a more serious problem was that
6726 the procedure for building the man page automatically didn't always
6727 operate. Consequently, release 4.20 contains an out-of-date version. This
6728 shouldn't happen again.
6730 5. When building Exim with embedded Perl support, the script that builds the
6731 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6732 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6734 6. The freeze_tell option was not being used for messages that were frozen on
6735 arrival, either by an ACL or by local_scan().
6737 7. Added the smtp_incomplete_transaction log selector.
6739 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6740 was accepting AUTH without a new EHLO.
6742 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6743 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6746 10. Logging of TCP/IP connections (when configured) now happens in the main
6747 daemon process instead of the child process, so that the TCP/IP connection
6748 count is more accurate (but it can never be perfect).
6750 11. The use of "drop" in a nested ACL was not being handled correctly in the
6751 outer ACL. Now, if condition failure induced by the nested "drop" causes
6752 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6753 or "require"), the connection is dropped.
6755 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6756 that yield "discard" can now be used with an "accept" or a "discard" verb,
6757 but an error is generated for any others (because I can't see a useful way
6758 to define what should happen).
6760 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6761 are now processed in the same way as lines in the Exim configuration file.
6762 In particular, continuation lines are supported.
6764 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6766 15. Added -ti meaning -t -i.
6768 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6769 domains, and warn by logging if others are found.
6771 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6772 returned from GIFCONF when Exim is trying to find the list of interfaces on
6773 a host. The code in os.c has been modified to copy each ifreq to an aligned
6774 structure in all cases.
6776 Also, in some cases, the returned ifreq's were being copied to a 'struct
6777 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6778 means the last couple of bytes of an IPv6 address could be chopped if the
6779 ifreq contained only a normal sockaddr (14 bytes storage).
6781 18. Named domain lists were not supported in the hosts_treat_as_local option.
6782 An entry such as +xxxx was not recognized, and was treated as a literal
6785 19. Ensure that header lines added by a DATA ACL are included in the reject log
6786 if the ACL subsequently rejects the message.
6788 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6789 MD5 (which is deprecated).
6791 21. When testing a filter file using -bf, Exim was writing a message when it
6792 took the sender from a "From " line in the message, but it was not doing so
6793 when it took $return_path from a Return-Path: header line. It now does.
6795 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6796 with a valid header line field name (a series of printing characters
6797 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6799 23. Changed "disc" in the source to "disk" to conform to the documentation and
6800 the book and for uniformity.
6802 24. Ignore Sendmail's -Ooption=value command line item.
6804 25. When execve() failed while trying to run a command in a pipe transport,
6805 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6806 could be confused with a return value of 69 from the command itself. This
6807 has been changed to 127, the value the shell returns if it is asked to run
6808 a non-existent command. The wording for the related log line suggests a
6809 non-existent command as the problem.
6811 26. If received_header_text expands to an empty string, do not add a Received:
6812 header line to the message. (Well, it adds a token one on the spool, but
6813 marks it "old" so that it doesn't get used or transmitted.)
6815 27. Installed eximstats 1.28 (addition of -nt option).
6817 28. There was no check for failure on the call to getsockname() in the daemon
6818 code. This can fail if there is a shortage of resources on the system, with
6819 ENOMEM, for example. A temporary error is now given on failure.
6821 29. Contrary to the C standard, it seems that in some environments, the
6822 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6823 program. Exim now does this explicitly; it affects the formatting of
6824 timestamps using strftime().
6826 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6827 complaints. I've now initialized all the variables, to avoid this.
6829 32. Header lines added by a system filter were not being "seen" during
6830 transport-time rewrites.
6832 33. The info_callback() function passed to OpenSSL is set up with type void
6833 (*)(SSL *, int, int), as described somewhere. However, when calling the
6834 function (actually a macro) that sets it up, the type void(*)() is
6835 expected. I've put in a cast to prevent warnings from picky compilers.
6837 34. If a DNS black list lookup found a CNAME record, but there were no A
6838 records associated with the domain it pointed at, Exim crashed.
6840 35. If a DNS black list lookup returned more than one A record, Exim ignored
6841 all but the first. It now scans all returned addresses if a particular IP
6842 value is being sought. In this situation, the contents of the
6843 $dnslist_value variable are a list of all the addresses, separated by a
6846 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6847 to accept a host name and all its aliases if the forward lookup for any of
6848 them yielded the IP address of the incoming connection. Now it accepts only
6849 those names whose forward lookup yields the correct IP address. Any other
6850 names are discarded. This closes a loophole whereby a rogue DNS
6851 administrator could create reverse DNS records to break through a
6852 wildcarded host restriction in an ACL.
6854 37. If a user filter or a system filter that ran in a subprocess used any of
6855 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6856 the wrong values were passed to the pipe command ($thisaddress had the
6857 value of $0, $0 had the value of $1, etc). This bug was introduced by
6858 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6860 38. Improved the line breaking for long SMTP error messages from ACLs.
6861 Previously, if there was no break point between 40 and 75 characters, Exim
6862 left the rest of the message alone. Two changes have been made: (a) I've
6863 reduced the minimum length to 35 characters; (b) if it can't find a break
6864 point between 35 and 75 characters, it looks ahead and uses the first one
6865 that it finds. This may give the occasional overlong line, but at least the
6866 remaining text gets split now.
6868 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6869 file descriptors might be low, and that setting 1000 would always raise it.
6870 It turns out that in some environments, the limit is already over 1000 and
6871 that lowering it causes trouble. So now Exim takes care not to decrease it.
6873 40. When delivering a message, the value of $return_path is set to $sender_
6874 address at the start of routing (routers may change the value). By an
6875 oversight, this default was not being set up when an address was tested by
6876 -bt or -bv, which affected the outcome if any router or filter referred to
6879 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6880 log only when "message" or "log_message" are set. However, if one of the
6881 conditions was an address verification, or a call to a nested ACL, the
6882 messages generated by the underlying test were being passed through. This
6883 no longer happens. The underlying message is available in $acl_verify_
6884 message for both "message" and "log_message" expansions, so it can be
6885 passed through if needed.
6887 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6888 new expansion $bh_ to give the encoded byte string without charset
6889 translation. Translation happens only if iconv() is available; HAVE_ICONV
6890 indicates this at build time. HEADERS_CHARSET gives the charset to
6891 translate to; headers_charset can change it in the configuration, and
6892 "headers charset" can change it in an individual filter file.
6894 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6895 that creates RFC 2047 encoded "words" labels them as that charset instead
6896 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6897 expansion operator; (ii) when Exim creates a From: line for a local
6898 message; (iii) when a header line is rewritten to include a "phrase" part.
6900 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6901 buggy, causing it to skip the first lines of messages whose message ID
6902 ended in 'D'. This would not have bitten before Exim release 4.14, because
6903 message IDs were unlikely to end in 'D' before then. The effect was to have
6904 incorrect size information for certain domains.
6906 45. #include "config.h" was missing at the start of the crypt16.c module. This
6907 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6910 46. If there was a timeout during a "random" callout check, Exim treated it as
6911 a failure of the random address, and carried on sending RSET and the real
6912 address. If the delay was just some slowness somewhere, the response to the
6913 original RCPT would be taken as a response to RSET and so on, causing
6914 mayhem of various kinds.
6916 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6917 when I implemented it. It didn't allow for the fact that some option values
6918 may legitimately be negative (e.g. size_addition), and it didn't even do
6919 the right test for positive values.
6921 48. Domain names in DNS records are case-independent. Exim always looks them up
6922 in lower case. Some resolvers return domain names in exactly the case they
6923 appear in the zone file, that is, they may contain uppercase letters. Not
6924 all resolvers do this - some return always lower case. Exim was treating a
6925 change of case by a resolver as a change of domain, similar to a widening
6926 of a domain abbreviation. This triggered its re-routing code and so it was
6927 trying to route what was effectively the same domain again. This normally
6928 caused routing to fail (because the router wouldn't handle the domain
6929 twice). Now Exim checks for this case specially, and just changes the
6930 casing of the domain that it ultimately uses when it transmits the message
6933 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6936 50. If a filter generated a file delivery with a non-absolute name (possible if
6937 no home directory exists for the router), the forbid_file option was not
6940 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6941 the existing equality matching.
6943 52. Exim was using ints instead of ino_t variables in some places where it was
6944 dealing with inode numbers.
6946 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6947 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6948 called TMPDIR, and if it finds it is different, it changes its value.
6950 54. The smtp_printf() function is now made available to local_scan() so
6951 additional output lines can be written before returning. There is also an
6952 smtp_fflush() function to enable the detection of a dropped connection.
6953 The variables smtp_input and smtp_batched_input are exported to
6956 55. Changed the default runtime configuration: the message "Unknown user"
6957 has been removed from the ACL, and instead placed on the localuser router,
6958 using the cannot_route_message feature. This means that any verification
6959 failures that generate their own messages won't get overridden. Similarly,
6960 the "Unrouteable address" message that was in the ACL for unverifiable
6961 relay addresses has also been removed.
6963 56. Added hosts_avoid_esmtp to the smtp transport.
6965 57. The exicyclog script was not checking for the esoteric option
6966 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6967 will work only if exicyclog is run under the appropriate euid.
6969 58. Following a discussion on the list, the rules by which Exim recognises line
6970 endings on incoming messages have been changed. The -dropcr and drop_cr
6971 options are now no-ops, retained only for backwards compatibility. The
6972 following line terminators are recognized: LF CRLF CR. However, special
6973 processing applies to CR:
6975 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6976 nor a local message in the state where . is a terminator.
6978 (ii) If a bare CR is encountered in a header line, an extra space is added
6979 after the line terminator so as not to end the header. The reasoning
6980 behind this is that bare CRs in header lines are most likely either
6981 to be mistakes, or people trying to play silly games.
6983 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6984 was being incorrectly given as 18 bytes larger than it should have been.
6985 This is a VOB (very old bug).
6987 60. This may never have affected anything current, but just in case it has:
6988 When the local host is found other than at the start of a list of hosts,
6989 the local host, those with the same MX, and any that follow, are discarded.
6990 When the list in question was part of a longer list of hosts, the following
6991 hosts (not currently being processed) were also being discarded. This no
6992 longer happens. I'm not sure if this situation could ever has previously
6995 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6998 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6999 header. This is something that is recommended in a new Internet Draft, and
7000 is something that is documented as being done by Sendmail. There are two
7001 possible values. For messages generated by the autoreply transport, Exim
7004 Auto-Submitted: auto-replied
7006 whereas for all other generated messages (e.g. bounces) it adds
7008 Auto-Submitted: auto-generated
7010 63. The "personal" condition in filters now includes a test for the
7011 Auto-Submitted: header. If it contains the string "auto-" the message it
7012 not considered personal.
7014 64. Added rcpt_include_affixes as a generic transport option.
7016 65. Added queue_only_override (default true).
7018 66. Added the syslog_duplication option.
7020 67. If what should have been the first header line of a message consisted of
7021 a space followed by a colon, Exim was mis-interpreting it as a header line.
7022 It isn't of course - it is syntactically invalid and should therefore be
7023 treated as the start of the message body. The misbehaviour could have
7024 caused a number of strange effects, including loss of data in subsequent
7025 header lines, and spool format errors.
7027 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7028 client host had authenticated. This control can now be exercised by an ACL
7029 for more flexibility.
7031 69. By default, callouts do not happen when testing with -bh. There is now a
7032 variant, -bhc, which does actually run the callout code, including
7033 consulting and updating the callout cache.
7035 70. Added support for saslauthd authentication, courtesy of Alexander
7038 71. If statvfs() failed on the spool or log directories while checking their
7039 size for availability, Exim confusingly gave the error "space shortage".
7040 Furthermore, in debugging mode it crashed with a floating point exception.
7041 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7042 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7043 serious problem, Exim now writes to the main and panic logs when this
7044 happens, with details of the failure. It then refuses to accept the
7045 incoming message, giving the message "spool directory problem" or "log
7046 directory problem" with a 421 code for SMTP messages.
7048 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7049 0, 1, and 2 exist, because some OS complain for execs without them (see
7050 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7051 descriptors. However, the code omitted to check that the open succeeded,
7052 causing mysterious errors if for some reason the permissions on /dev/null
7053 got screwed. Now Exim writes a message to the main and panic logs, and
7054 bombs out if it can't open /dev/null.
7056 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7057 interact so that it is all more flexible. It is supposed to remain
7058 backwards compatible. Also added extra_local_interfaces.
7060 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7061 to bomb out with an assertion failure - to the client this appears as a
7062 connection drop. This problem occurs in the part of the code that was taken
7063 from the Samba project. Fortunately, the assertion is in a very simple
7064 function, so I have fixed this by reproducing the function inline in the
7065 one place where it is called, and arranging for authentication to fail
7066 instead of killing the process with assert().
7068 75. The SPA client code was not working when the server requested OEM rather
7069 than Unicode encoding.
7071 76. Added code to make require_files with a specific uid setting more usable in
7072 the case where statting the file as root fails - usually a non-root-mounted
7073 NFS file system. When this happens and the failure is EACCES, Exim now
7074 forks a subprocess and does the per-uid checking as the relevant uid.
7076 77. Added process_log_path.
7078 78. If log_file_path was not explicitly set, a setting of check_log_space or
7079 check_log_inodes was ignored.
7081 79. If a space check for the spool or log partitions fails, the incident is now
7082 logged. Of course, in the latter case the data may get lost...
7084 80. Added the %p formatting code to string_format() so that it can be used to
7085 print addresses in debug_print(). Adjusted all the address printing in the
7086 debugging in store.c to use %p rather than %d.
7088 81. There was a concern that a line of code in smtp_in.c could overflow a
7089 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7090 initially expressed, the concern was not well-founded, because trailing
7091 spaces are removed early. However, if the trailing spaces were followed by
7092 a NULL, they did not get removed, so the overflow was possible. Two fixes
7095 (a) I re-wrote the offending code in a cleaner fashion.
7096 (b) If an incoming SMTP command contains a NULL character, it is rejected
7099 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7100 runs initgroups(), so that if the Exim user is in any additional groups,
7101 they will be used during message reception.
7107 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob