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.
148 Do the same for the queue-runner loop, for variables set from spool
149 message files. Do the same for the SMTP per-message loop, for certain
150 variables indirectly set in ACL operations.
156 JH/01 Rework error string handling in TLS interface so that the caller in
157 more cases is responsible for logging. This permits library-sourced
158 string to be attached to addresses during delivery, and collapses
159 pairs of long lines into single ones.
161 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
162 during configuration. Wildcards are allowed and expanded.
164 JH/02 Rework error string handling in DKIM to pass more info back to callers.
165 This permits better logging.
167 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
168 do not close it down and have the child transport start it up again on
169 the passed-on TCP connection. Instead, proxy the child (and any
170 subsequent ones) for TLS via a unix-domain socket channel. Logging is
171 affected: the continued delivery log lines do not have any DNSSEC, TLS
172 Certificate or OCSP information. TLS cipher information is still logged.
174 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
175 identical IP addresses on different listening ports. Will also affect
178 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
179 add noisy ifdef guards to special-case this sillyness.
180 Patch from Bernd Kuhls.
182 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
183 than 255 are no longer allowed.
185 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
186 Disable the session-cache too, which might reduce our load. Since we
187 currrectly use a new context for every connection, both as server and
188 client, there is no benefit for these.
189 GnuTLS appears to not support tickets server-side by default (we don't
190 call gnutls_session_ticket_enable_server()) but client side is enabled
191 by default on recent versions (3.1.3 +) unless the PFS priority string
194 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
195 <https://reproducible-builds.org/specs/source-date-epoch/>.
197 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
198 the check for any unsuccessful recipients did not notice the limit, and
199 erroneously found still-pending ones.
201 JH/08 Pipeline CHUNKING command and data together, on kernels that support
202 MSG_MORE. Only in-clear (not on TLS connections).
204 JH/09 Avoid using a temporary file during transport using dkim. Unless a
205 transport-filter is involved we can buffer the headers in memory for
206 creating the signature, and read the spool data file once for the
207 signature and again for transmission.
209 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
210 4.77 as the kernel support then wasn't solid, having issues in 64bit
211 mode. Now, it's been long enough. Add support for FreeBSD also.
213 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
214 case where the routing stage had gathered several addresses to send to
215 a host before calling the transport for the first, we previously failed
216 to close down TLS in the old transport process before passing the TCP
217 connection to the new process. The new one sent a STARTTLS command
218 which naturally failed, giving a failed delivery and bloating the retry
219 database. Investigation and fix prototype from Wolfgang Breyha.
221 JH/12 Fix check on SMTP command input synchronisation. Previously there were
222 false-negatives in the check that the sender had not preempted a response
223 or prompt from Exim (running as a server), due to that code's lack of
224 awareness of the SMTP input buffering.
226 PP/04 Add commandline_checks_require_admin option.
227 Exim drops privileges sanely, various checks such as -be aren't a
228 security problem, as long as you trust local users with access to their
229 own account. When invoked by services which pass untrusted data to
230 Exim, this might be an issue. Set this option in main configuration
231 AND make fixes to the calling application, such as using `--` to stop
234 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
235 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
236 responses to those, into a single TLS record each way (this usually means
237 a single packet). As a side issue, smtp_enforce_sync now works on TLS
240 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
241 affects you only if you're dancing at the edge of the param size limits.
242 If you are, and this message makes sense to you, then: raise the
243 configured limit or use OpenSSL 1.1. Nothing we can do for older
246 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
247 cmdline element to get the documented default one. Previously it was
250 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
251 are unsuable for non-message-specific reasons - eg. connection timeout,
252 banner-time rejection.
254 JH/16 Fix logging of delivery remote port, when specified by router, under
257 PP/06 Repair manualroute's ability to take options in any order, even if one
258 is the name of a transport.
261 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
263 JH/17 Change the list-building routines interface to use the expanding-string
264 triplet model, for better allocation and copying behaviour.
266 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
267 Previously it was constructed the first time a possibly-matching string
268 was met in the configuration file input during startup; now it is done
271 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
272 compatible one, to avoid the (poorly documented) possibility of a config
273 file in the working directory redirecting the DB files, possibly correpting
274 some existing file. CVE-2017-10140 assigned for BDB.
276 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
277 cache-hot. Previously, although the result was properly cached, the
278 initial verify call returned a defer.
280 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
281 the main verify for receipient in uncached-mode.
283 JH/22 Retire historical build files to an "unsupported" subdir. These are
284 defined as "ones for which we have no current evidence of testing".
286 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
287 if present. Previously it was ignored.
289 JH/24 Start using specified-initialisers in C structure init coding. This is
290 a C99 feature (it's 2017, so now considered safe).
292 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
293 if was a fixed-sized field and bitmask ops via macros; it is now more
296 PP/07 GitHub PR 56: Apply MariaDB build fix.
297 Patch provided by Jaroslav Škarvada.
299 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
300 during Coverity cleanups [4.87 JH/47]
301 Diagnosis and fix provided by Michael Fischer v. Mollard.
303 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
304 the right size to place the terminating semicolon on its own folded
305 line, the header hash was calculated to an incorrect value thanks to
306 the (relaxed) space the fold became.
308 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
309 and confused the parent.
311 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
312 which could crash as a result. This could lead to undeliverable messages.
314 JH/28 Logging: "next input sent too soon" now shows where input was truncated
317 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
318 matters on fast-turnover and PID-randomising systems, which were getting
319 out-of-order delivery.
321 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
322 a possibly-overlapping copy. The symptom was that "Remote host closed
323 connection in response to HELO" was logged instead of the actual 4xx
326 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
327 Previously only that bufferd was discarded, resulting in SYMTP command
330 JH/32 DKIM: when a message has multiple signatures matching an identity given
331 in dkim_verify_signers, run the dkim acl once for each. Previously only
332 one run was done. Bug 2189.
334 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
335 "panic the current process" to "deliberately defer". The panic log is
336 still written with the problem list name; the mail and reject logs now
337 get a temp-reject line for the message that was being handled, saying
338 something like "domains check lookup or other defer". The SMTP 451
339 message is still "Temporary local problem".
341 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
342 A crafted sequence of BDAT commands could result in in-use memory beeing
343 freed. CVE-2017-16943.
345 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
346 from SMTP input. Previously it was always done; now only done for DATA
347 and not BDAT commands. CVE-2017-16944.
349 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
350 to the message (such as an overlong header line). Previously this was
351 not done and we did not exit BDAT mode. Followon from the previous item
352 though a different problem.
358 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
359 than -2003 did; needs libidn2 in addition to libidn.
361 JH/02 The path option on a pipe transport is now expanded before use.
363 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
364 Patch provided by "Björn", documentation fix added too.
366 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
367 missing a wire-to-host endian conversion.
369 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
370 close after a BDAT command line could be taken as a following command,
371 giving a synch failure. Fix by only checking for synch immediately
372 before acknowledging the chunk.
374 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
375 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
376 macro. Patches provided by Josh Soref.
378 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
379 Previously we did not; the RFC seems ambiguous and VRFY is not listed
380 by IANA as a service extension. However, John Klensin suggests that we
383 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
384 the dkim code may be unix-mode line endings rather than smtp wire-format
385 CRLF, so prepend a CR to any bare LF.
387 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
388 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
390 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
391 the first were themselves being wrongly included in the feed into dkim
392 processing; with most chunk sizes in use this resulted in an incorrect
393 body hash calculated value.
395 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
396 DKIM signature block, for verification. Although advised against by
397 standards it is specifically not ruled illegal.
399 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
401 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
402 missing a body hash (the bh= tag).
404 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
405 It seems that HAProxy sends the Proxy Protocol information in clear and
406 only then does a TLS startup, so do the same.
408 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
409 TCP connections (such as for Spamd) unless the daemon successfully set
410 Fast Open mode on its listening sockets. This fixes breakage seen on
411 too-old kernels or those not configured for Fast Open, at the cost of
412 requiring both directions being enabled for TFO, and TFO never being used
413 by non-daemon-related Exim processes.
415 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
416 endings, at least on the first header line. Try to canonify any that get
417 past that check, despite the cost.
419 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
420 now limited to an arbitrary five deep, while parsing addresses with the
421 strip_excess_angle_brackets option enabled.
423 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
424 instead leave the unprompted TLS handshake in socket buffer for the
425 TLS library to consume.
427 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
429 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
431 JH/16 Drop variables when they go out of scope. Memory management drops a whole
432 region in one operation, for speed, and this leaves assigned pointers
433 dangling. Add checks run only under the testsuite which checks all
434 variables at a store-reset and panics on a dangling pointer; add code
435 explicitly nulling out all the variables discovered. Fixes one known
436 bug: a transport crash, where a dangling pointer for $sending_ip_address
437 originally assigned in a verify callout, is re-used.
439 PP/06 Drop '.' from @INC in various Perl scripts.
441 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
443 PP/08 Reduce a number of compilation warnings under clang; building with
444 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
445 should be warning-free.
447 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
449 HS/01 Fix portability problems introduced by PP/08 for platforms where
450 realloc(NULL) is not equivalent to malloc() [SunOS et al].
452 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
453 chunk. This allows us to accept broken chunked messages. We need a more
454 general solution here.
456 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
457 already-broken messages in the queue.
459 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
461 JH/19 Fix reference counting bug in routing-generated-address tracking.
467 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
468 supports it and a size is available (ie. the sending peer gave us one).
470 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
471 years of being deprecated). The replacements are the ACLs
472 acl_smtp_mime and acl_not_smtp_mime.
474 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
475 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
476 or even an in-clear connection were permitted. Now, if the host lookup
477 was dnssec and dane was requested then the host is only used if the
478 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
479 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
480 if one fails this test.
481 This means that a poorly-configured remote DNS will make it incommunicado;
482 but it protects against a DNS-interception attack on it.
484 JH/04 Bug 1810: make continued-use of an open smtp transport connection
485 non-noisy when a race steals the message being considered.
487 JH/05 If main configuration option tls_certificate is unset, generate a
488 self-signed certificate for inbound TLS connections.
490 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
491 in rewrites and routers.
493 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
494 and logged a warning sing 4.83; now they are a configuration file error.
496 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
497 (lacking @domain). Apply the same qualification processing as RCPT.
499 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
501 JH/10 Support ${sha256:} applied to a string (as well as the previous
504 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
505 a cutthrough deliver is pending, as we always want to make a connection.
506 This also avoids re-routing the message when later placing the cutthrough
507 connection after a verify cache hit.
508 Do not update it with the verify result either.
510 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
511 when routing results in more than one destination address.
513 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
514 signing (which inhibits the cutthrough capability). Previously only
515 the presence of an option was tested; now an expansion evaluating as
516 empty is permissible (obviously it should depend only on data available
517 when the cutthrough connection is made).
519 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
520 the relevant preceding SMTP command did not note the pipelining mode.
522 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
523 Previously they were not counted.
525 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
526 as one having no matching records. Previously we deferred the message
527 that needed the lookup.
529 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
530 distinguished as "(=".
532 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
533 for missing MX records. Previously it only worked for missing A records.
535 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
537 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
538 after the data-go-ahead and data-ack. Patch from Jason Betts.
540 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
541 even for a "none" policy. Patch from Tony Meyer.
543 JH/22 Fix continued use of a connection for further deliveries. If a port was
544 specified by a router, it must also match for the delivery to be
547 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
548 When one of the recipients of a message was unsuitable for the connection
549 (has no matching addresses), we lost track of needing to mark it
550 deferred. As a result mail would be lost.
552 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
554 JH/25 Decoding ACL controls is now done using a binary search; the source code
555 takes up less space and should be simpler to maintain. Merge the ACL
556 condition decode tables also, with similar effect.
558 JH/26 Fix problem with one_time used on a redirect router which returned the
559 parent address unchanged. A retry would see the parent address marked as
560 delivered, so not attempt the (identical) child. As a result mail would
563 JH/27 Fix a possible security hole, wherein a process operating with the Exim
564 UID can gain a root shell. Credit to http://www.halfdog.net/ for
565 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
568 JH/28 Enable {spool,log} filesystem space and inode checks as default.
569 Main config options check_{log,spool}_{inodes,space} are now
570 100 inodes, 10MB unless set otherwise in the configuration.
572 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
573 Previously it only applied to the main-section connection policy
576 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
578 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
579 by me. Added RFC7919 DH primes as an alternative.
581 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
582 are not in the system include path.
584 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
585 GnuTLS, when a session startup failed (eg because the client disconnected)
586 Exim did stdio operations after fclose. This was exposed by a recent
587 change which nulled out the file handle after the fclose.
589 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
590 signed directly by the cert-signing cert, rather than an intermediate
591 OCSP-signing cert. This is the model used by LetsEncrypt.
593 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
595 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
596 an incoming connection.
598 HS/02 Bug 1802: Do not half-close the connection after sending a request
601 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
602 fallback to "prime256v1".
604 JH/34 SECURITY: Use proper copy of DATA command in error message.
605 Could leak key material. Remotely exploitable. CVE-2016-9963.
611 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
612 and 3.4.4 - once the server is enabled to respond to an OCSP request
613 it does even when not requested, resulting in a stapling non-aware
614 client dropping the TLS connection.
616 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
617 support variable-length bit vectors. No functional change.
619 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
620 The I= interface field on outgoing lines is now after the H= remote
621 host field, same as incoming lines. There is a separate
622 outgoing_interface log selector which allows you to disable the
625 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
626 If not running log_selector +smtp_connection the mainlog would be held
627 open indefinitely after a "too many connections" event, including to a
628 deleted file after a log rotate. Leave the per net connection logging
629 leaving it open for efficiency as that will be quickly detected by the
630 check on the next write.
632 HS/01 Bug 1671: Fix post transport crash.
633 Processing the wait-<transport> messages could crash the delivery
634 process if the message IDs didn't exist for some reason. When
635 using 'split_spool_directory=yes' the construction of the spool
636 file name failed already, exposing the same netto behaviour.
638 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
639 mime_regex ACL conditions.
641 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
642 to DSN fail messages (bounces): remote IP, remote greeting, remote response
643 to HELO, local diagnostic string.
645 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
646 log line to debug. Even when configured with a tls authenticator many
647 client connections are expected to not authenticate in this way, so
648 an authenticate fail is not an error.
650 HS/02 Add the Exim version string to the process info. This way exiwhat
651 gives some more detail about the running daemon.
653 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
654 matter for fast-change records such as DNSBLs.
656 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
657 retry record, even if constant. There may be multiple transports with
658 different interface settings and the retry behaviour needs to be kept
661 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
663 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
665 JH/10 Bug 840: fix log_defer_output option of pipe transport
667 JH/11 Bug 830: use same host for all RCPTS of a message, even under
668 hosts_randomize. This matters a lot when combined with mua_wrapper.
670 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
671 ${quote_pgsql:<string>} operator.
673 JH/13 Bug 1708: avoid misaligned access in cached lookup.
675 JH/14 Change header file name for freeradius-client. Relevant if compiling
676 with Radius support; from the Gentoo tree and checked under Fedora.
678 JH/15 Bug 1712: Introduce $prdr_requested flag variable
680 JH/16 Bug 1714: Permit an empty string as expansion result for transport
681 option transport_filter, meaning no filtering.
683 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
685 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
686 defaults to "*" (all hosts). The variable is now available when not built
687 with TLS, default unset, mainly to enable keeping the testsuite sane.
688 If a server certificate is not supplied (via tls_certificate) an error is
689 logged, and clients will find TLS connections fail on startup. Presumably
690 they will retry in-clear.
691 Packagers of Exim are strongly encouraged to create a server certificate
692 at installation time.
694 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
695 with the $config_file variable.
697 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
698 in transport context, after the attempt, and per-recipient. The latter type
699 is per host attempted. The event data is the error message, and the errno
700 information encodes the lookup type (A vs. MX) used for the (first) host,
701 and the trailing two digits of the smtp 4xx response.
703 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
704 to write to mainlog (or rejectlog, paniclog) in the window between file
705 creation and permissions/ownership being changed. Particularly affects
706 installations where exicyclog is run as root, rather than exim user;
707 result is that the running daemon panics and dies.
709 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
711 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
712 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
713 "pri" and "weight". Note that the previous implicit priority given by the
714 list order is no longer honoured.
716 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
719 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
720 by defining SUPPORT_SOCKS.
722 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
723 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
724 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
725 variables are renamed to proxy_{local,external}_{address,port}.
727 JH/27 Move Internationalisation support from Experimental to mainline, enabled
728 for a build by defining SUPPORT_I18N
730 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
731 of the query string, and make ${quote_redis:} do that quoting.
733 JH/29 Move Events support from Experimental to mainline, enabled by default
734 and removable for a build by defining DISABLE_EVENT.
736 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
738 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
739 cached by the daemon.
741 JH/32 Move Redis support from Experimental to mainline, enabled for a build
742 by defining LOOKUP_REDIS. The libhiredis library is required.
744 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
745 keys are given for lookup.
747 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
748 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
749 only supported when built with TLS support. The PolarSSL SHA routines
750 are still used when the TLS library is too old for convenient support.
752 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
753 openssl_options), for security. OpenSSL forces this from version 1.1.0
754 server-side so match that on older versions.
756 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
757 allocation for $value could be released as the expansion processing
758 concluded, but leaving the global pointer active for it.
760 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
761 and to use the domains and local_parts ACL conditions.
763 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
764 incorrectly not doubled on cutthrough transmission, hence seen as a
765 body-termination at the receiving system - resulting in truncated mails.
766 Commonly the sender saw a TCP-level error, and retransmitted the message
767 via the normal store-and-forward channel. This could result in duplicates
768 received - but deduplicating mailstores were liable to retain only the
769 initial truncated version.
771 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
773 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
775 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
776 we're in there, support oversigning also; bug 1309.
778 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
780 HS/04 Add support for keep_environment and add_environment options.
782 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
783 either intentional arithmetic overflow during PRNG, or testing config-
786 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
787 delivery resulted in actual delivery. Cancel cutthrough before DATA
790 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
791 rejects a recipient: pass the reject to the originator.
793 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
794 Many were false-positives and ignorable, but it's worth fixing the
797 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
798 for the new environment-manipulation done at startup. Move the routines
799 from being local to tls.c to being global via the os.c file.
801 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
802 an extract embedded as result-arg for a map, the first arg for extract
803 is unavailable so we cannot tell if this is a numbered or keyed
804 extraction. Accept either.
810 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
813 JH/02 The smtp transport option "multi_domain" is now expanded.
815 JH/03 The smtp transport now requests PRDR by default, if the server offers
818 JH/04 Certificate name checking on server certificates, when exim is a client,
819 is now done by default. The transport option tls_verify_cert_hostnames
820 can be used to disable this per-host. The build option
821 EXPERIMENTAL_CERTNAMES is withdrawn.
823 JH/05 The value of the tls_verify_certificates smtp transport and main options
824 default to the word "system" to access the system default CA bundle.
825 For GnuTLS, only version 3.0.20 or later.
827 JH/06 Verification of the server certificate for a TLS connection is now tried
828 (but not required) by default. The verification status is now logged by
829 default, for both outbound TLS and client-certificate supplying inbound
832 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
835 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
836 Status Notification (bounce) messages are now MIME format per RFC 3464.
837 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
838 under the control of the dsn_advertise_hosts option, and routers may
839 have a dsn_lasthop option.
841 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
842 default, modifiable by a malware= option. The list separator for
843 the options can now be changed in the usual way. Bug 68.
845 JH/10 The smtp_receive_timeout main option is now expanded before use.
847 JH/11 The incoming_interface log option now also enables logging of the
848 local interface on delivery outgoing connections.
850 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
851 if the interface and destination host and port all match.
853 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
856 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
857 Patch from Andrew Lewis.
859 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
860 now supports optional time-restrictions, weighting, and priority
861 modifiers per server. Patch originally by <rommer@active.by>.
863 JH/16 The spamd_address main option now supports a mixed list of local
864 and remote servers. Remote servers can be IPv6 addresses, and
865 specify a port-range.
867 JH/17 Bug 68: The spamd_address main option now supports an optional
868 timeout value per server.
870 JH/18 Bug 1581: Router and transport options headers_add/remove can
871 now have the list separator specified.
873 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
876 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
879 JH/21 Support for the A6 type of dns record is withdrawn.
881 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
882 rather than the verbs used.
884 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
885 from 255 to 1024 chars.
887 JH/24 Verification callouts now attempt to use TLS by default.
889 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
890 are generic router options now. The defaults didn't change.
892 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
893 Original patch from Alexander Shikoff, worked over by JH.
895 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
896 files if ZCAT_COMMAND is not executable.
898 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
900 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
902 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
903 Normally benign, it bites when the pair was led to by a CNAME;
904 modern usage is to not canonicalize the domain to a CNAME target
905 (and we were inconsistent anyway for A-only vs AAAA+A).
907 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
909 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
910 when evaluating $sender_host_dnssec.
912 JH/31 Check the HELO verification lookup for DNSSEC, adding new
913 $sender_helo_dnssec variable.
915 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
917 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
919 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
921 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
922 documented as working, but never had. Support all but $spam_report.
924 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
925 added for tls authenticator.
927 HS/03 Add perl_taintmode main config option
933 TL/01 When running the test suite, the README says that variables such as
934 no_msglog_check are global and can be placed anywhere in a specific
935 test's script, however it was observed that placement needed to be near
936 the beginning for it to behave that way. Changed the runtest perl
937 script to read through the entire script once to detect and set these
938 variables, reset to the beginning of the script, and then run through
939 the script parsing/test process like normal.
941 TL/02 The BSD's have an arc4random API. One of the functions to induce
942 adding randomness was arc4random_stir(), but it has been removed in
943 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
944 function when detected.
946 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
947 cause callback expansion.
949 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
950 syntax errors in an expansion can be treated as a string instead of
951 logging or causing an error, due to the internal use of bool_lax
952 instead of bool when processing it.
954 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
955 server certificates when making smtp deliveries.
957 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
959 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
961 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
963 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
964 Merged patch from Sebastian Wiedenroth.
966 JH/05 Fix results-pipe from transport process. Several recipients, combined
967 with certificate use, exposed issues where response data items split
968 over buffer boundaries were not parsed properly. This eventually
969 resulted in duplicates being sent. This issue only became common enough
970 to notice due to the introduction of connection certificate information,
971 the item size being so much larger. Found and fixed by Wolfgang Breyha.
973 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
974 size buffer was used, resulting in syntax errors when an expansion
977 JH/07 Add support for directories of certificates when compiled with a GnuTLS
978 version 3.3.6 or later.
980 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
981 is EXPERIMENTAL_EVENT, the main-configuration and transport options
982 both become "event_action", the variables become $event_name, $event_data
983 and $event_defer_errno. There is a new variable $verify_mode, usable in
984 routers, transports and related events. The tls:cert event is now also
985 raised for inbound connections, if the main configuration event_action
988 TL/06 In test suite, disable OCSP for old versions of openssl which contained
989 early OCSP support, but no stapling (appears to be less than 1.0.0).
991 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
992 server certificate names available under the smtp transport option
993 "tls_verify_cert_hostname" now do not permit multi-component wildcard
996 JH/10 Time-related extraction expansions from certificates now use the main
997 option "timezone" setting for output formatting, and are consistent
998 between OpenSSL and GnuTLS compilations. Bug 1541.
1000 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1001 encoded parameter in the incoming message. Bug 1558.
1003 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1004 include certificate info, eximon was claiming there were spoolfile
1007 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1009 JH/14 Log delivery-related information more consistently, using the sequence
1010 "H=<name> [<ip>]" wherever possible.
1012 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1013 are problematic for Debian distribution, omit them from the release
1016 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1018 JH/16 Fix string representation of time values on 64bit time_t architectures.
1021 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1022 output list separator was used.
1027 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1028 checkers that were complaining about end of non-void function with no
1031 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1032 This was a regression introduced in 4.83 by another bugfix.
1034 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1036 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1037 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1043 TF/01 Correctly close the server side of TLS when forking for delivery.
1045 When a message was received over SMTP with TLS, Exim failed to clear up
1046 the incoming connection properly after forking off the child process to
1047 deliver the message. In some situations the subsequent outgoing
1048 delivery connection happened to have the same fd number as the incoming
1049 connection previously had. Exim would try to use TLS and fail, logging
1050 a "Bad file descriptor" error.
1052 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1053 utilities have not been installed.
1055 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1056 temporary space as the ACL may create new global variables.
1058 TL/01 LDAP support uses per connection or global context settings, depending
1059 upon the detected version of the libraries at build time.
1061 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1062 to extract and use the src ip:port in logging and expansions as if it
1063 were a direct connection from the outside internet. PPv2 support was
1064 updated based on HAProxy spec change in May 2014.
1066 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1068 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1069 Properly escape header and check for NULL return.
1071 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1074 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1076 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1077 characters in header names, implemented as a verify condition.
1078 Contributed by Michael Fischer v. Mollard.
1080 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1081 results permerror and temperror. Previous values are deprecated but
1082 still accepted. In a future release, err_perm and err_temp will be
1083 completely removed, which will be a backward incompatibility if the
1084 ACL tests for either of these two old results. Patch contributed by
1085 user bes-internal on the mailing list.
1087 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1089 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1090 selectors, in both main and reject logs.
1092 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1095 JH/07 Add malware type "sock" for talking to simple daemon.
1097 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1099 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1100 routers/transports under cutthrough routing.
1102 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1103 numbers. Touch up "bool" conditional to keep the same definition.
1105 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1107 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1109 JH/12 Expand items in router/transport headers_add or headers_remove lists
1110 individually rather than the list as a whole. Bug 1452.
1112 Required for reasonable handling of multiple headers_ options when
1113 they may be empty; requires that headers_remove items with embedded
1114 colons must have them doubled (or the list-separator changed).
1116 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1117 view the policy declared in the DMARC record. Currently, $dmarc_status
1118 is a combined value of both the record presence and the result of the
1121 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1123 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1124 dnslookup router and the smtp transport (applying to the forward
1127 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1128 of ldap servers used for a specific lookup. Patch provided by Heiko
1131 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1132 New variable $lookup_dnssec_authenticated for observability.
1134 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1135 Patch submitted by Lars Timman.
1137 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1139 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1140 Requires trusted mode and valid format message id, aborts otherwise.
1141 Patch contributed by Heiko Schlichting.
1143 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1144 certextract with support for various fields. Bug 1358.
1146 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1147 is requested by default, modifiable by smtp transport option
1150 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1151 operate on certificate variables to give certificate fingerprints
1152 Also new ${sha256:cert_variable}.
1154 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1156 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1157 Christian Aistleitner.
1159 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1161 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1162 file. Patch from Wolfgang Breyha.
1164 JH/25 Expand the coverage of the delivery $host and $host_address to
1165 client authenticators run in verify callout. Bug 1476.
1167 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1168 align with daemon_smtp_ports. Bug 72.
1170 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1171 support and error reporting did not work properly.
1173 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1174 and is readable. Patch from Andrew Colin Kissa.
1176 TL/14 Enhance documentation of ${run expansion and how it parses the
1177 commandline after expansion, particularly in the case when an
1178 unquoted variable expansion results in an empty value.
1180 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1182 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1183 by renaming away from it. Addresses GH issue 15, reported by
1186 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1188 TL/15 SECURITY: prevent double expansion in math comparison functions
1189 (can expand unsanitized data). Not remotely exploitable.
1196 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1198 PP/02 Make -n do something, by making it not do something.
1199 When combined with -bP, the name of an option is not output.
1201 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1204 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1205 $sender_host_name and config options to manage this, and basic check
1208 PP/05 DSCP support for outbound connections and control modifier for inbound.
1210 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1211 (Only plugin which currently uses this is kerberos4, which nobody should
1212 be using, but we should make it available and other future plugins might
1213 conceivably use it, even though it would break NAT; stuff *should* be
1214 using channel bindings instead).
1216 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1217 name; added for Sendmail compatibility; requires admin caller.
1218 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1219 just ignore it); requires trusted caller.
1220 Also parse but ignore: -Ac -Am -X<logfile>
1223 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1225 TL/02 Add +smtp_confirmation as a default logging option.
1227 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1228 Patch by Magnus Holmgren from 2007-02-20.
1230 TL/04 Bugzilla 1281 - Spec typo.
1231 Bugzilla 1283 - Spec typo.
1232 Bugzilla 1290 - Spec grammar fixes.
1234 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1236 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1238 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1239 mailing list by by Dmitry Isaikin.
1241 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1243 JH/02 Support "G" suffix to numbers in ${if comparisons.
1245 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1247 NM/01 Bugzilla 1197 - Spec typo
1248 Bugzilla 1196 - Spec examples corrections
1250 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1252 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1253 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1256 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1257 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1259 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1260 "acl {{name}{arg}...}", and optional args on acl condition
1263 JH/05 Permit multiple router/transport headers_add/remove lines.
1265 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1267 JH/07 Avoid using a waiting database for a single-message-only transport.
1268 Performance patch from Paul Fisher. Bugzilla 1262.
1270 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1273 JH/09 Add $headers_added variable, with content from use of ACL modifier
1274 add_header (but not yet added to the message). Bugzilla 199.
1276 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1277 Pulled from Bugzilla 817 by Wolfgang Breyha.
1279 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1281 (nb: this is the same fix as in Exim 4.80.1)
1283 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1286 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1289 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1291 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1292 advertise SMTP AUTH mechanism to us, instead of a generic
1293 protocol violation error. Also, make Exim more robust to bad
1294 data from the Dovecot auth socket.
1296 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1298 When a queue runner is handling a message, Exim first routes the
1299 recipient addresses, during which it prunes them based on the retry
1300 hints database. After that it attempts to deliver the message to
1301 any remaining recipients. It then updates the hints database using
1304 So if a recipient address works intermittently, it can get repeatedly
1305 deferred at routing time. The retry hints record remains fresh so the
1306 address never reaches the final cutoff time.
1308 This is a fairly common occurrence when a user is bumping up against
1309 their storage quota. Exim had some logic in its local delivery code
1310 to deal with this. However it did not apply to per-recipient defers
1311 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1313 This change adds a proper retry rule check during routing so that the
1314 final cutoff time is checked against the message's age. We only do
1315 this check if there is an address retry record and there is not a
1316 domain retry record; this implies that previous attempts to handle
1317 the address had the retry_use_local_parts option turned on. We use
1318 this as an approximation for the destination being like a local
1319 delivery, as in LMTP.
1321 I suspect this new check makes the old local delivery cutoff check
1322 redundant, but I have not verified this so I left the code in place.
1324 TF/02 Correct gecos expansion when From: is a prefix of the username.
1326 Test 0254 submits a message to Exim with the header
1330 When I ran the test suite under the user fanf2, Exim expanded
1331 the header to contain my full name, whereas it should have added
1332 a Resent-Sender: header. It erroneously treats any prefix of the
1333 username as equal to the username.
1335 This change corrects that bug.
1337 GF/01 DCC debug and logging tidyup
1338 Error conditions log to paniclog rather than rejectlog.
1339 Debug lines prefixed by "DCC: " to remove any ambiguity.
1341 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1343 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1344 Bug spotted by Jeremy Harris; was flawed since initial commit.
1345 Would have resulted in OCSP responses post-SNI triggering an Exim
1346 NULL dereference and crash.
1348 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1350 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1351 Bug detection, analysis and fix by Samuel Thibault.
1352 Bugzilla 1331, Debian bug #698092.
1354 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1356 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1357 Server implementation by Todd Lyons, client by JH.
1358 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1359 config variable "prdr_enable" controls whether the server
1360 advertises the facility. If the client requests PRDR a new
1361 acl_data_smtp_prdr ACL is called once for each recipient, after
1362 the body content is received and before the acl_smtp_data ACL.
1363 The client is controlled by both of: a hosts_try_prdr option
1364 on the smtp transport, and the server advertisement.
1365 Default client logging of deliveries and rejections involving
1366 PRDR are flagged with the string "PRDR".
1368 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1369 fclose(). Diagnosis by Todd Lyons.
1371 PP/17 Update configure.default to handle IPv6 localhost better.
1372 Patch by Alain Williams (plus minor tweaks).
1375 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1376 This is now consistent with GnuTLS, and is now documented: the
1377 previous undocumented portable approach to treating the option as
1378 unset was to force an expansion failure. That still works, and
1379 an empty string is now equivalent.
1381 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1382 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1383 not performing validation itself.
1385 PP/20 Added force_command boolean option to pipe transport.
1386 Patch from Nick Koston, of cPanel Inc.
1388 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1391 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1393 PP/21 Fix eximon continuous updating with timestamped log-files.
1394 Broken in a format-string cleanup in 4.80, missed when I repaired the
1395 other false fix of the same issue.
1396 Report and fix from Heiko Schlichting.
1399 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1400 Report from Prashanth Katuri.
1402 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1403 It's SecureTransport, so affects any MacOS clients which use the
1404 system-integrated TLS libraries, including email clients.
1406 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1407 using a MIME ACL for non-SMTP local injection.
1408 Report and assistance in diagnosis by Warren Baker.
1410 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1412 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1414 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1415 last id that failed so it may be referenced in subsequent ACL's.
1417 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1420 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1421 ldap library initialization, allowing self-signed CA's to be
1422 used. Also properly sets require_cert option later in code by
1423 using NULL (global ldap config) instead of ldap handle (per
1424 session). Bug diagnosis and testing by alxgomz.
1426 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1427 the src/util/ subdirectory.
1429 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1430 renamed to Transport Post Delivery Action by Jeremy Harris, as
1433 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1434 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1435 redis_servers = needs to be configured which will be used by the redis
1436 lookup. Patch from Warren Baker, of The Packet Hub.
1438 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1440 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1441 hostname or reverse DNS when processing a host list. Used suggestions
1442 from multiple comments on this bug.
1444 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1446 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1447 Missed a few lines, added it to make the runtest require no keyboard
1450 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1451 contains upper case chars. Make router use caseful_local_part.
1453 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1454 support when GnuTLS has been built with p11-kit.
1460 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1462 This, or similar/improved, will also be change PP/11 of 4.82.
1468 PP/01 Handle short writes when writing local log-files.
1469 In practice, only affects FreeBSD (8 onwards).
1470 Bugzilla 1053, with thanks to Dmitry Isaikin.
1472 NM/01 Bugzilla 949 - Documentation tweak
1474 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1477 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1479 PP/02 Implemented gsasl authenticator.
1481 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1483 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1484 `pkg-config foo` for cflags/libs.
1486 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1487 with rest of GSASL and with heimdal_gssapi.
1489 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1490 `pkg-config foo` for cflags/libs for the TLS implementation.
1492 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1493 properties get this fed in as external SSF. A number of robustness
1494 and debugging improvements to the cyrus_sasl authenticator.
1496 PP/08 cyrus_sasl server now expands the server_realm option.
1498 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1499 Patch by Jeremy Harris.
1501 PP/10 Added dbmjz lookup type.
1503 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1505 PP/12 MAIL args handles TAB as well as SP, for better interop with
1506 non-compliant senders.
1507 Analysis and variant patch by Todd Lyons.
1509 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1510 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1511 Patch from Dirk Mueller <dmueller@suse.com>
1513 PP/13 tls_peerdn now print-escaped for spool files.
1514 Observed some $tls_peerdn in wild which contained \n, which resulted
1515 in spool file corruption.
1517 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1518 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1519 or write after TLS renegotiation, which otherwise led to messages
1522 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1523 as a tracking header (ie: a signed header comes before the signature).
1524 Patch from Wolfgang Breyha.
1526 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1527 comma-sep list; embedded commas doubled.
1529 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1531 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1533 Report and patch from Dmitry Banschikov.
1535 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1536 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1537 failures. We appear to now support TLS1.1+ with Exim.
1539 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1540 lets Exim select keys and certificates based upon TLS SNI from client.
1541 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1542 before an outbound SMTP session. New log_selector, +tls_sni.
1544 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1545 NULL dereference. Report and patch from Alun Jones.
1547 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1548 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1549 Not seeing resolver debug output on NetBSD, but suspect this is a
1550 resolver implementation change.
1552 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1553 Left warnings. Added "eximon gdb" invocation mode.
1555 PP/21 Defaulting "accept_8bitmime" to true, not false.
1557 PP/22 Added -bw for inetd wait mode support.
1559 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1560 locate the relevant includes and libraries. Made this the default.
1562 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1563 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1565 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1566 This may cause build issues on older platforms.
1568 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1569 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1570 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1571 Added SNI support via GnuTLS too.
1572 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1574 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1576 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1577 Applied second patch from Janne, implementing suggestion to default
1578 multiple-strings-in-record handling to match SPF spec.
1580 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1582 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1583 read-only, out of scope).
1584 Patch from Wolfgang Breyha, report from Stuart Northfield.
1586 PP/29 Fix three issues highlighted by clang analyser static analysis.
1587 Only crash-plausible issue would require the Cambridge-specific
1588 iplookup router and a misconfiguration.
1589 Report from Marcin Mirosław.
1591 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1593 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1594 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1595 As part of this, removing so much warning spew let me fix some minor
1596 real issues in debug logging.
1598 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1599 assignment on my part. Fixed.
1601 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1602 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1603 Janne Snabb (who went above and beyond: thank you).
1605 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1606 string otherwise requires a connection and a bunch more work and it's
1607 relatively easy to get wrong. Should also expose TLS library linkage
1610 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1611 64-bit ${eval} (JH/03).
1613 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1614 GNU libc to support some of the 64-bit stuff, should not lead to
1615 conflicts. Defined before os.h is pulled in, so if a given platform
1616 needs to override this, it can.
1618 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1619 protection layer was required, which is not implemented.
1620 Bugzilla 1254, patch from Wolfgang Breyha.
1622 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1623 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1624 tls_dhparam take prime identifiers. Also unbreak combination of
1625 OpenSSL+DH_params+TLSSNI.
1627 PP/39 Disable SSLv2 by default in OpenSSL support.
1633 PP/01 Solaris build fix for Oracle's LDAP libraries.
1634 Bugzilla 1109, patch from Stephen Usher.
1636 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1638 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1641 TF/02 Fix a couple more cases where we did not log the error message
1642 when unlink() failed. See also change 4.74-TF/03.
1644 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1645 lock up or crash if it happened to be inside a call to libc when it
1646 got a SIGUSR1 from exiwhat.
1648 The SIGUSR1 handler appends the current process status to the process
1649 log which is later printed by exiwhat. It used to use the general
1650 purpose logging code to do this, but several functions it calls are
1651 not safe for signals.
1653 The new output code in the SIGUSR1 handler is specific to the process
1654 log, and simple enough that it's easy to inspect for signal safety.
1655 Removing some special cases also simplifies the general logging code.
1656 Removing the spurious timestamps from the process log simplifies
1659 TF/04 Improved ratelimit ACL condition.
1661 The /noupdate option has been deprecated in favour of /readonly which
1662 has clearer semantics. The /leaky, /strict, and /readonly update modes
1663 are mutually exclusive. The update mode is no longer included in the
1664 database key; it just determines when the database is updated. (This
1665 means that when you upgrade Exim will forget old rate measurements.)
1667 Exim now checks that the per_* options are used with an update mode that
1668 makes sense for the current ACL. For example, when Exim is processing a
1669 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1670 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1671 must specify per_mail/readonly. If you omit the update mode it defaults to
1672 /leaky where that makes sense (as before) or /readonly where required.
1674 The /noupdate option is now undocumented but still supported for
1675 backwards compatibility. It is equivalent to /readonly except that in
1676 ACLs where /readonly is required you may specify /leaky/noupdate or
1677 /strict/noupdate which are treated the same as /readonly.
1679 A useful new feature is the /count= option. This is a generalization
1680 of the per_byte option, so that you can measure the throughput of other
1681 aggregate values. For example, the per_byte option is now equivalent
1682 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1684 The per_rcpt option has been generalized using the /count= mechanism
1685 (though it's more complicated than the per_byte equivalence). When it is
1686 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1687 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1688 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1689 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1690 Note that using per_rcpt with a non-readonly update mode in more than
1691 one ACL will cause the recipients to be double-counted. (The per_mail
1692 and per_byte options don't have this problem.)
1694 The handling of very low rates has changed slightly. If the computed rate
1695 is less than the event's count (usually one) then this event is the first
1696 after a long gap. In this case the rate is set to the same as this event's
1697 count, so that the first message of a spam run is counted properly.
1699 The major new feature is a mechanism for counting the rate of unique
1700 events. The new per_addr option counts the number of different
1701 recipients that someone has sent messages to in the last time period. It
1702 behaves like per_rcpt if all the recipient addresses are different, but
1703 duplicate recipient addresses do not increase the measured rate. Like
1704 the /count= option this is a general mechanism, so the per_addr option
1705 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1706 example, measure the rate that a client uses different sender addresses
1707 with the options per_mail/unique=$sender_address. There are further
1708 details in the main documentation.
1710 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1712 TF/06 Removed a few PCRE remnants.
1714 TF/07 Automatically extract Exim's version number from tags in the git
1715 repository when doing development or release builds.
1717 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1718 Bugzilla 879. Patch from Paul Fisher.
1720 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1721 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1724 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1726 PP/05 Variable $av_failed, true if the AV scanner deferred.
1727 Bugzilla 1078. Patch from John Horne.
1729 PP/06 Stop make process more reliably on build failure.
1730 Bugzilla 1087. Patch from Heiko Schlittermann.
1732 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1733 Bugzilla 1089. Patch from Heiko Schlittermann.
1735 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1736 Bugzilla 1131. Patch from Holger Weiß.
1738 PP/09 Handle IPv6 addresses with SPF.
1739 Bugzilla 860. Patch from Wolfgang Breyha.
1741 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1743 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1746 PP/11 match_* no longer expand right-hand-side by default.
1747 New compile-time build option, EXPAND_LISTMATCH_RHS.
1748 New expansion conditions, "inlist", "inlisti".
1750 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1752 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1754 PP/14 fix log_write() format string regression from TF/03.
1755 Bugzilla 1152. Patch from Dmitry Isaikin.
1761 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1763 PP/02 Harmonised TLS library version reporting; only show if debugging.
1764 Layout now matches that introduced for other libraries in 4.74 PP/03.
1766 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1768 PP/04 New "dns_use_edns0" global option.
1770 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1773 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1774 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1776 TK/01 Updated PolarSSL code to 0.14.2.
1777 Bugzilla 1097. Patch from Andreas Metzler.
1779 PP/07 Catch divide-by-zero in ${eval:...}.
1780 Fixes bugzilla 1102.
1782 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1785 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1786 format-string attack -- SECURITY: remote arbitrary code execution.
1788 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1789 time unintentionally subject to list matching rules, letting the header
1790 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1791 arbitrary string expansion). This allowed for information disclosure.
1793 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1794 INT_MIN/-1 -- value coerced to INT_MAX.
1800 NM/01 Workaround for PCRE version dependency in version reporting
1803 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1804 This fixes portability to compilers other than gcc, notably
1805 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1807 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1808 makefiles for portability to HP-UX and POSIX correctness.
1810 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1811 Also via indented variable definition in the Makefile.
1812 (Debugging by Oliver Heesakkers).
1814 PP/02 Restore caching of spamd results with expanded spamd_address.
1815 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1817 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1818 Improves build reliability. Fix from: Frank Elsner
1820 NM/02 Fix wide character breakage in the rfc2047 coding
1821 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1823 NM/03 Allow underscore in dnslist lookups
1824 Fixes bug 1026. Patch from Graeme Fowler
1826 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1827 Code patches from Adam Ciarcinski of NetBSD.
1829 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1832 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1833 is logged, to avoid truncation. Patch from John Horne.
1835 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1836 Patch from Jakob Hirsch.
1838 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1839 SQL string expansion failure details.
1840 Patch from Andrey Oktyabrski.
1842 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1843 Patch from Simon Arlott.
1845 PP/09 New lookups functionality failed to compile on old gcc which rejects
1846 extern declarations in function scope.
1847 Patch from Oliver Fleischmann
1849 PP/10 Use sig_atomic_t for flags set from signal handlers.
1850 Check getgroups() return and improve debugging.
1851 Fixed developed for diagnosis in bug 927 (which turned out to be
1854 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1855 Patch from Mark Zealey.
1857 PP/12 Bugzilla 1056: Improved spamd server selection.
1858 Patch from Mark Zealey.
1860 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1861 Based on patch from Heiko Schlittermann.
1863 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1864 Patch from Uwe Doering, sign-off by Michael Haardt.
1866 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1867 variable declaration deep within a block. Bug and patch from
1870 PP/15 lookups-Makefile IRIX compatibility coercion.
1872 PP/16 Make DISABLE_DKIM build knob functional.
1874 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1875 Patch by Simon Arlott
1877 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1878 variable argument macros. Our copy now differs from upstream.
1884 TF/01 Failure to get a lock on a hints database can have serious
1885 consequences so log it to the panic log.
1887 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1888 controlled using the smtp_confirmation log selector.
1890 TF/03 Include the error message when we fail to unlink a spool file.
1892 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1893 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1894 for maintaining out-of-tree patches for some time.
1896 PP/01 Bugzilla 139: Documentation and portability issues.
1897 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1898 Handle per-OS dynamic-module compilation flags.
1900 PP/02 Let /dev/null have normal permissions.
1901 The 4.73 fixes were a little too stringent and complained about the
1902 permissions on /dev/null. Exempt it from some checks.
1903 Reported by Andreas M. Kirchwitz.
1905 PP/03 Report version information for many libraries, including
1906 Exim version information for dynamically loaded libraries. Created
1907 version.h, now support a version extension string for distributors
1908 who patch heavily. Dynamic module ABI change.
1910 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1911 privilege escalation vulnerability whereby the Exim run-time user
1912 can cause root to append content of the attacker's choosing to
1915 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1918 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1919 If dropping privileges for untrusted macros, we disabled normal logging
1920 on the basis that it would fail; for the Exim run-time user, this is not
1921 the case, and it resulted in successful deliveries going unlogged.
1922 Fixed. Reported by Andreas Metzler.
1928 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1929 only prepend for the Resent-* case. Fixes regression introduced in
1930 Exim 4.70 by NM/22 for Bugzilla 607.
1932 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1933 increasing numbers of administrators be bitten by this.
1935 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1937 PP/03 Bugzilla 994: added openssl_options main configuration option.
1939 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1941 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1943 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1945 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1946 without a peer certificate, leading to a segfault because of an
1947 assumption that peers always have certificates. Be a little more
1948 paranoid. Problem reported by Martin Tscholak.
1950 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1951 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1952 NB: ClamAV planning to remove STREAM in "middle of 2010".
1953 CL also introduces -bmalware, various -d+acl logging additions and
1954 more caution in buffer sizes.
1956 PP/09 Implemented reverse_ip expansion operator.
1958 PP/10 Bugzilla 937: provide a "debug" ACL control.
1960 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1962 PP/12 Bugzilla 973: Implement --version.
1964 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1966 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1968 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1970 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1971 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1972 ignore trailing whitespace.
1974 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1976 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1979 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1980 Notification from Dr Andrew Aitchison.
1982 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1983 ExtendedDetectionInfo response format.
1984 Notification from John Horne.
1986 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1989 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1990 XSL and documented dependency on system catalogs, with examples of how
1993 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1996 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1997 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1998 configuration file which is writeable by the Exim user or group.
2000 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2001 of configuration files to cover files specified with the -C option if
2002 they are going to be used with root privileges, not just the default
2005 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2006 option (effectively making it always true).
2008 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2009 files to be used while preserving root privileges.
2011 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2012 that rogue child processes cannot use them.
2014 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2015 run-time user, instead of root.
2017 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2018 Exim run-time user without dropping privileges.
2020 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2021 result string, instead of calling string_vformat() twice with the same
2024 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2025 for other users. Others should always drop root privileges if they use
2026 -C on the command line, even for a whitelisted configure file.
2028 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2030 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2036 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2037 $data_path, and $header_path variables; fixed documentation bugs and
2040 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2041 exipick to access non-standard spools, including the "frozen" queue
2044 NM/01 Bugzilla 965: Support mysql stored procedures.
2045 Patch from Alain Williams
2047 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2049 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2050 Patch from Andreas Metzler
2052 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2053 Patch from Kirill Miazine
2055 NM/05 Bugzilla 671: Added umask to procmail example.
2057 JJ/03 installed exipick 20100323.0, fixing doc bug
2059 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2060 directory. Notification and patch from Dan Rosenberg.
2062 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2064 TK/02 Improve log output when DKIM signing operation fails.
2066 MH/01 Treat the transport option dkim_domain as a colon separated
2067 list, not as a single string, and sign the message with each element,
2068 omitting multiple occurences of the same signer.
2070 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2071 Bugzilla 985, 986. Patch by Simon Arlott
2073 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2074 Patch by Simon Arlott
2076 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2077 MBX locking. Notification from Dan Rosenberg.
2083 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2085 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2087 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2089 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2091 NM/04 Bugzilla 915: Flag broken perl installation during build.
2097 TK/01 Added patch by Johannes Berg that expands the main option
2098 "spamd_address" if it starts with a dollar sign.
2100 TK/02 Write list of recipients to X-Envelope-Sender header when building
2101 the mbox-format spool file for content scanning (suggested by Jakob
2104 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2105 (http://www.dcc-servers.net/) support via dccifd. Activated by
2106 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2108 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2109 by Mark Daniel Reidel <mr@df.eu>.
2111 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2112 When building exim an external PCRE library is now needed -
2113 PCRE is a system library on the majority of modern systems.
2114 See entry on PCRE_LIBS in EDITME file.
2116 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2117 conversation. Added nologin parameter to request.
2118 Patch contributed by Kirill Miazine.
2120 TF/01 Do not log submission mode rewrites if they do not change the address.
2122 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2124 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2125 log files in place. Contributed by Roberto Lima.
2127 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2129 TF/03 Bugzilla 615: When checking the local_parts router precondition
2130 after a local_part_suffix or local_part_prefix option, Exim now
2131 does not use the address's named list lookup cache, since this
2132 contains cached lookups for the whole local part.
2134 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2135 Robert Millan. Documentation is in experimental-spec.txt.
2137 TF/04 Bugzilla 668: Fix parallel build (make -j).
2139 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2141 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2142 Patch provided by Jan Srzednicki.
2144 TF/05 Leading white space used to be stripped from $spam_report which
2145 wrecked the formatting. Now it is preserved.
2147 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2148 that they are available at delivery time.
2150 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2152 TF/08 TLS error reporting now respects the incoming_interface and
2153 incoming_port log selectors.
2155 TF/09 Produce a more useful error message if an SMTP transport's hosts
2156 setting expands to an empty string.
2158 NM/06 Bugzilla 744: EXPN did not work under TLS.
2159 Patch provided by Phil Pennock.
2161 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2162 Patch provided by Richard Godbee.
2164 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2165 acl_smtp_notquit, added index entry.
2167 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2168 Patch provided by Eugene Bujak.
2170 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2171 accept(). Patch provided by Maxim Dounin.
2173 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2174 Patch provided by Phil Pennock.
2176 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2178 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2179 Patch provided by Brad "anomie" Jorsch.
2181 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2182 Patch provided by Dean Brooks.
2184 TK/05 Add native DKIM support (does not depend on external libraries).
2186 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2187 Patch provided by Graeme Fowler.
2189 NM/16 Bugzilla 851: Documentation example syntax fix.
2191 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2193 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2196 NM/19 Bugzilla 745: TLS version reporting.
2197 Patch provided by Phil Pennock.
2199 NM/20 Bugzilla 167: bool: condition support.
2200 Patch provided by Phil Pennock.
2202 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2203 clients. Patch provided by Phil Pennock.
2205 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2206 Patch provided by Brad "anomie" Jorsch.
2208 NM/23 Bugzilla 687: Fix misparses in eximstats.
2209 Patch provided by Heiko Schlittermann.
2211 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2212 Patch provided by Heiko Schlittermann.
2214 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2215 plus update to original patch.
2217 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2219 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2220 Patch provided by David Brownlee.
2222 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2224 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2226 NM/30 Bugzilla 888: TLS documentation bugfixes.
2228 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2230 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2231 Unlike the original bugzilla I have changed all shell scripts in src tree.
2233 NM/33 Bugzilla 898: Transport filter timeout fix.
2234 Patch by Todd Rinaldo.
2236 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2237 Patch by Serge Demonchaux.
2239 NM/35 Bugzilla 39: Base64 decode bug fixes.
2240 Patch by Jakob Hirsch.
2242 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2244 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2246 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2248 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2254 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2255 ALT-N's libdkim that I have put here:
2256 http://duncanthrax.net/exim-experimental/
2258 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2259 were called 'true' and it seems that C99 defines that as a reserved
2260 keyword to be used with 'bool' variable types. That means you could
2261 not include C99-style headers which use bools without triggering
2262 build errors in sieve.c.
2264 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2265 as mailq or other aliases. Changed the --help handling significantly
2266 to do whats expected. exim_usage() emits usage/help information.
2268 SC/01 Added the -bylocaldomain option to eximstats.
2270 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2272 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2274 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2280 PH/01 Another patch from the Sieve maintainer.
2282 PH/02 When an IPv6 address is converted to a string for single-key lookup
2283 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2284 dots are used instead of colons so that keys in lsearch files need not
2285 contain colons. This was done some time before quoting was made available
2286 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2287 (notated using the quote facility) so as to distinguish them from IPv4
2288 keys. This meant that lookups for IP addresses in host lists did not work
2289 for iplsearch lookups.
2291 This has been fixed by arranging for IPv6 addresses to be expressed with
2292 colons if the lookup type is iplsearch. This is not incompatible, because
2293 previously such lookups could never work.
2295 The situation is now rather anomalous, since one *can* have colons in
2296 ordinary lsearch keys. However, making the change in all cases is
2297 incompatible and would probably break a number of configurations.
2299 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2302 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2303 conversion specification without a maximum field width, thereby enabling
2304 a rogue spamd server to cause a buffer overflow. While nobody in their
2305 right mind would setup Exim to query an untrusted spamd server, an
2306 attacker that gains access to a server running spamd could potentially
2307 exploit this vulnerability to run arbitrary code as the Exim user.
2309 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2310 $primary_hostname instead of what libspf2 thinks the hosts name is.
2312 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2313 a directory entry by the name of the lookup key. Previously, if a
2314 symlink pointed to a non-existing file or a file in a directory that
2315 Exim lacked permissions to read, a lookup for a key matching that
2316 symlink would fail. Now it is enough that a matching directory entry
2317 exists, symlink or not. (Bugzilla 503.)
2319 PH/03 The body_linecount and body_zerocount variables are now exported in the
2322 PH/04 Added the $dnslist_matched variable.
2324 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2325 This means they are set thereafter only if the connection becomes
2328 PH/06 Added the client_condition to authenticators so that some can be skipped
2329 by clients under certain conditions.
2331 PH/07 The error message for a badly-placed control=no_multiline_responses left
2332 "_responses" off the end of the name.
2334 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2336 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2337 (without spaces) instead of just copying the configuration text.
2339 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2341 PH/11 Added $max_received_linelength.
2343 PH/12 Added +ignore_defer and +include_defer to host lists.
2345 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2346 way in which PCRE > 7.0 is built.
2348 PH/14 Implemented queue_only_load_latch.
2350 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2351 MAIL command. The effect was to mangle the value on 64-bit systems.
2353 PH/16 Another patch from the Sieve maintainer.
2355 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2357 PH/18 If a system quota error occurred while trying to create the file for
2358 a maildir delivery, the message "Mailbox is full" was not appended to the
2359 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2360 only to a quota excession during the actual writing of the file.
2362 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2363 characters?) which causes problems in log lines. The DN values are now
2364 passed through string_printing() before being added to log lines.
2366 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2367 and InterBase are left for another time.)
2369 PH/21 Added message_body_newlines option.
2371 PH/22 Guard against possible overflow in moan_check_errorcopy().
2373 PH/23 POSIX allows open() to be a macro; guard against that.
2375 PH/24 If the recipient of an error message contained an @ in the local part
2376 (suitably quoted, of course), incorrect values were put in $domain and
2377 $local_part during the evaluation of errors_copy.
2383 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2384 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2387 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2388 issue a MAIL command.
2390 PH/02 In an ACL statement such as
2392 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2394 if a client was not listed at all, or was listed with a value other than
2395 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2396 the condition was not true (as it should be), so access was not denied.
2397 The bug was that the ! inversion was incorrectly passed on to the second
2398 item. This has been fixed.
2400 PH/03 Added additional dnslists conditions == and =& which are different from
2401 = and & when the dns lookup returns more than one IP address.
2403 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2404 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2406 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2407 FSYNC, which compiles an option called disable_fsync that allows for
2408 bypassing fsync(). The documentation is heavily laced with warnings.
2410 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2412 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2413 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2414 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2415 including adding "make clean"; (3) Added -fPIC when compiling the test
2416 dynamically loaded module, to get rid of a warning.
2418 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2419 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2420 = 0s. The bug is otherwise harmless.
2422 PH/07 There was a bug in the dovecot authenticator such that the value of
2423 $auth1 could be overwritten, and so not correctly preserved, after a
2424 successful authentication. This usually meant that the value preserved by
2425 the server_setid option was incorrect.
2427 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2429 PH/09 Installed PCRE release 7.0.
2431 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2432 run for batched SMTP input. It is now run at the start of every message
2433 in the batch. While fixing this I discovered that the process information
2434 (output by running exiwhat) was not always getting set for -bs and -bS
2435 input. This is fixed, and it now also says "batched" for BSMTP.
2437 PH/11 Added control=no_pipelining.
2439 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2440 patch, slightly modified), and move the expansion of helo_data till after
2441 the connection is made in the smtp transport (so it can use these
2444 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2446 PH/14 Added log_selector = +pid.
2448 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2450 PH/16 Add ${if forany and ${if forall.
2452 PH/17 Added dsn_from option to vary the From: line in DSNs.
2454 PH/18 Flush SMTP output before performing a callout, unless control =
2455 no_callout_flush is set.
2457 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2458 was true (the default) a successful delivery failed to delete the retry
2459 item, thus causing premature timeout of the address. The bug is now
2462 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2464 PH/21 Long custom messages for fakedefer and fakereject are now split up
2465 into multiline responses in the same way that messages for "deny" and
2466 other ACL rejections are.
2468 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2469 with slight modification.
2471 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2472 draft, changing the syntax and factoring some duplicate code".
2474 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2475 for deliveries of the second and subsequent messages over the same SMTP
2478 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2479 ${reduce, with only minor "tidies".
2481 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2483 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2484 expansion side effects.
2486 PH/27 When a message times out after an over-quota error from an Exim-imposed
2487 quota, the bounce message says "mailbox is full". This message was not
2488 being given when it was a system quota that was exceeded. It now should
2491 MH/03 Made $recipients available in local_scan(). local_scan() already has
2492 better access to the recipient list through recipients_list[], but
2493 $recipients can be useful in postmaster-provided expansion strings.
2495 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2496 in the case of a MAIL command with additional options following the
2497 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2498 were accidentally chopped off.
2500 PH/29 SMTP synchronization checks are implemented when a command is read -
2501 there is a check that no more input is waiting when there shouldn't be
2502 any. However, for some commands, a delay in an ACL can mean that it is
2503 some time before the response is written. In this time, more input might
2504 arrive, invalidly. So now there are extra checks after an ACL has run for
2505 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2506 pipelining has not been advertised.
2508 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2510 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2511 returned in the SMTP response when smtp_return_error_details was false.
2512 This has been fixed.
2514 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2515 instead of the C I/O that was originally supplied, because problems were
2516 reported on Solaris.
2518 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2519 Exim which did not show up earlier: it was assuming that a call to
2520 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2521 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2522 was a macro that became an assignment, so it seemed to work. This has
2523 changed to a proper function call with a void return, hence the compile
2524 error. Exim's code has been fixed.
2526 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2529 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2531 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2533 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2534 because it turns out to be too big or there is a timeout) there is no
2535 "Completed" line in the log. When some messages of this type were
2536 selected by exigrep, they were listed as "not completed". Others were
2537 picked up by some special patterns. I have improved the selection
2538 criteria to be more general.
2540 PH/38 The host_find_failed option in the manualroute router can now be set
2541 to "ignore", to completely ignore a host whose IP address cannot be
2542 found. If all hosts are ignored, the behaviour is controlled by the new
2543 host_all_ignored option.
2545 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2546 homing or because of multiple MX records with /mx) generated more than
2547 one IP address, and the following item turned out to be the local host,
2548 all the secondary addresses of the first item were incorrectly removed
2549 from the list, along with the local host and any following hosts (which
2550 is what is supposed to happen).
2552 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2553 whoever called Exim into the -H file. In the case of the daemon it was
2554 behaving confusingly. When first started, it used values for whoever
2555 started the daemon, but after a SIGHUP it used the Exim user (because it
2556 calls itself on a restart). I have changed the code so that it now always
2559 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2560 message are rejected with the same error (e.g. no authentication or bad
2561 sender address), and a DATA command is nevertheless sent (as can happen
2562 with PIPELINING or a stupid MUA), the error message that was given to the
2563 RCPT commands is included in the rejection of the DATA command. This is
2564 intended to be helpful for MUAs that show only the final error to their
2567 PH/42 Another patch from the Sieve maintainer.
2569 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2570 Eximstats - Fixed some broken HTML links and added missing column headers
2572 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2573 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2575 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2577 PH/43 Yet another patch from the Sieve maintainer.
2579 PH/44 I found a way to check for a TCP/IP connection going away before sending
2580 the response to the final '.' that terminates a message, but only in the
2581 case where the client has not sent further data following the '.'
2582 (unfortunately, this is allowed). However, in many cases there won't be
2583 any further data because there won't be any more messages to send. A call
2584 to select() can be used: if it shows that the input is "ready", there is
2585 either input waiting, or the socket has been closed. An attempt to read
2586 the next input character can distinguish the two cases. Previously, Exim
2587 would have sent an OK response which the client would never have see.
2588 This could lead to message repetition. This fix should cure that, at
2589 least in a lot of common cases.
2591 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2592 advertised in response to EHLO.
2598 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2599 fixed by 4.65/MH/01 (is this a record?) are fixed:
2601 (i) An empty string was always treated as zero by the numeric comparison
2602 operators. This behaviour has been restored.
2604 (ii) It is documented that the numeric comparison operators always treat
2605 their arguments as decimal numbers. This was broken in that numbers
2606 starting with 0 were being interpreted as octal.
2608 While fixing these problems I realized that there was another issue that
2609 hadn't been noticed. Values of message_size_limit (both the global option
2610 and the transport option) were treated as octal if they started with 0.
2611 The documentation was vague. These values are now always treated as
2612 decimal, and I will make that clear in the documentation.
2618 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2619 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2622 MH/01 Don't check that the operands of numeric comparison operators are
2623 integers when their expansion is in "skipping" mode (fixes bug
2624 introduced by 4.64-PH/07).
2626 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2627 child addresses, Exim now panics and dies. Previously, because the count
2628 is held in a short int, deliveries were likely to be lost. As such a
2629 large number of recipients for a single message is ridiculous
2630 (performance will be very, very poor), I have chosen to impose a limit
2631 rather than extend the field.
2637 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2638 leftover -K file (the existence of which was triggered by #402).
2639 While we were at it, introduced process PID as part of the -K
2640 filename. This should rule out race conditions when creating
2643 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2644 processing considerably. Previous code took too long for large mails,
2645 triggering a timeout which in turn triggers #401.
2647 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2648 in the DK code in transports.c. sendfile() is not really portable,
2649 hence the _LINUX specificness.
2651 TF/01 In the add_headers option to the mail command in an Exim filter,
2652 there was a bug that Exim would claim a syntax error in any
2653 header after the first one which had an odd number of characters
2656 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2657 callout verification, Exim cached a "reject" for the entire domain. This
2658 is correct for most verifications, but it is not correct for a recipient
2659 verification with use_sender or use_postmaster set, because in that case
2660 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2661 case of MAIL FROM:<> rejection from other early rejections (e.g.
2662 rejection of HELO). When verifying a recipient using a non-null MAIL
2663 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2664 Whatever the result of the callout, the value of the domain cache is
2665 left unchanged (for any other kind of callout, getting as far as trying
2666 RCPT means that the domain itself is ok).
2668 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2671 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2672 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2673 session, the daemon ignores ECONNECT errors and logs others; it now
2674 ignores EPIPE as well.
2676 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2677 (quoted-printable decoding).
2679 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2680 later the small subsequent patch to fix an introduced bug.
2682 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2684 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2686 PH/08 An error is now given if message_size_limit is specified negative.
2688 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2689 to be given (somewhat) arbitrary names.
2691 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2694 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2695 miscellaneous code fixes
2697 PH/10 Added the log_reject_target ACL modifier to specify where to log
2700 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2701 hostname. This is wrong, because it relates to the incoming message (and
2702 probably the interface on which it is arriving) and not to the outgoing
2703 callout (which could be using a different interface). This has been
2704 changed to use the value of the helo_data option from the smtp transport
2705 instead - this is what is used when a message is actually being sent. If
2706 there is no remote transport (possible with a router that sets up host
2707 addresses), $smtp_active_hostname is used.
2709 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2710 tweaks were necessary in order to get it to work (see also 21 below):
2711 (a) The code assumed that strncpy() returns a negative number on buffer
2712 overflow, which isn't the case. Replaced with Exim's string_format()
2714 (b) There were several signed/unsigned issues. I just did the minimum
2715 hacking in of casts. There is scope for a larger refactoring.
2716 (c) The code used strcasecmp() which is not a standard C function.
2717 Replaced with Exim's strcmpic() function.
2718 (d) The code set only $1; it now sets $auth1 as well.
2719 (e) A simple test gave the error "authentication client didn't specify
2720 service in request". It would seem that Dovecot has changed its
2721 interface. Fortunately there's a specification; I followed it and
2722 changed what the client sends and it appears to be working now.
2724 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2727 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2728 address A is aliased to B and C, where B exists and C does not. Without
2729 -v the output is "A verified" because verification stops after a
2730 successful redirection if more than one address is generated. However,
2731 with -v the child addresses are also verified. Exim was outputting "A
2732 failed to verify" and then showing the successful verification for C,
2733 with its parentage. It now outputs "B failed to verify", showing B's
2734 parentage before showing the successful verification of C.
2736 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2737 look up a TXT record in a specific list after matching in a combined
2740 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2741 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2742 they consult the DNS. I had assumed they would set it the way they
2743 wanted; and indeed my experiments on Linux seem to show that in some
2744 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2745 To be on the safe side, however, I have now made the interface to
2746 host_find_byname() similar to host_find_bydns(), with an argument
2747 containing the DNS resolver options. The host_find_byname() function now
2748 sets these options at its start, just as host_find_bydns() does. The smtp
2749 transport options dns_qualify_single and dns_search_parents are passed to
2750 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2751 of host_find_byname() use the default settings of RES_DEFNAMES
2752 (qualify_single) but not RES_DNSRCH (search_parents).
2754 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2755 spool_read_header() do less string testing, by means of a preliminary
2756 switch on the second character of optional "-foo" lines. (This is
2757 overdue, caused by the large number of possibilities that now exist.
2758 Originally there were few.) While I was there, I also converted the
2759 str(n)cmp tests so they don't re-test the leading "-" and the first
2760 character, in the hope this might squeeze out yet more improvement.
2762 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2763 flag allowing group syntax was set by the header_syntax check but not
2764 turned off, possible causing trouble later; (2) The flag was not being
2765 set at all for the header_verify test, causing "group"-style headers to
2766 be rejected. I have now set it in this case, and also caused header_
2767 verify to ignore an empty address taken from a group. While doing this, I
2768 came across some other cases where the code for allowing group syntax
2769 while scanning a header line wasn't quite right (mostly, not resetting
2770 the flag correctly in the right place). These bugs could have caused
2771 trouble for malformed header lines. I hope it is now all correct.
2773 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2774 with the "reply" argument non-NULL. The code, however (which originally
2775 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2776 but it didn't always do it. This confused somebody who was copying the
2777 code for some other use. I have removed all the tests.
2779 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2780 feature that was used to support insecure browsers during the U.S. crypto
2781 embargo. It requires special client support, and Exim is probably the
2782 only MTA that supported it -- and would never use it because real RSA is
2783 always available. This code has been removed, because it had the bad
2784 effect of slowing Exim down by computing (never used) parameters for the
2785 RSA_EXPORT functionality.
2787 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2788 authenticator to fail if there's a tab character in the incoming data
2789 (there should never be unless someone is messing about, as it's supposed
2790 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2791 if the connection is using TLS or if the remote IP is the same as the
2792 local IP, and the "valid-client-cert option" if a client certificate has
2795 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2796 authenticators. This can be used for authorization after authentication
2797 succeeds. (In the case of plaintext, it servers for both authentication
2800 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2801 if any retry times were supplied.
2803 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2804 connection, where there is no client IP address to check. In this
2805 situation, the verify now always succeeds.
2807 PH/25 Applied John Jetmore's -Mset patch.
2809 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2811 PH/27 In a string expansion for a processed (not raw) header when multiple
2812 headers of the same name were present, leading whitespace was being
2813 removed from all of them, but trailing whitespace was being removed only
2814 from the last one. Now trailing whitespace is removed from each header
2815 before concatenation. Completely empty headers in a concatenation (as
2816 before) are ignored.
2818 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2819 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2821 PH/29 [Removed. This was a change that I later backed out, and forgot to
2822 correct the ChangeLog entry (that I had efficiently created) before
2823 committing the later change.]
2825 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2826 address errors (4xx response to RCPT) over the same connection as other
2827 messages routed to the same hosts. Such deliveries are always "forced",
2828 so retry times are not inspected. This resulted in far too many retries
2829 for the affected addresses. The effect occurred only when there were more
2830 hosts than the hosts_max_try setting in the smtp transport when it had
2831 the 4xx errors. Those hosts that it had tried were not added to the list
2832 of hosts for which the message was waiting, so if all were tried, there
2833 was no problem. Two fixes have been applied:
2835 (i) If there are any address or message errors in an SMTP delivery, none
2836 of the hosts (tried or untried) are now added to the list of hosts
2837 for which the message is waiting, so the message should not be a
2838 candidate for sending over the same connection that was used for a
2839 successful delivery of some other message. This seems entirely
2840 reasonable: after all the message is NOT "waiting for some host".
2841 This is so "obvious" that I'm not sure why it wasn't done
2842 previously. Hope I haven't missed anything, but it can't do any
2843 harm, as the worst effect is to miss an optimization.
2845 (ii) If, despite (i), such a delivery is accidentally attempted, the
2846 routing retry time is respected, so at least it doesn't keep
2847 hammering the server.
2849 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2850 in ${readsocket because some servers need this prod.
2852 PH/32 Added some extra debug output when updating a wait-xxx database.
2854 PH/33 The hint "could be header name not terminated by colon", which has been
2855 given for certain expansion errors for a long time, was not being given
2856 for the ${if def:h_colon_omitted{... case.
2858 PH/34 The spec says: "With one important exception, whenever a domain list is
2859 being scanned, $domain contains the subject domain." There was at least
2860 one case where this was not true.
2862 PH/35 The error "getsockname() failed: connection reset by peer" was being
2863 written to the panic log as well as the main log, but it isn't really
2864 panic-worthy as it just means the connection died rather early on. I have
2865 removed the panic log writing for the ECONNRESET error when getsockname()
2868 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2869 runs only) independently of the message's sender address. This meant
2870 that, if the 4xx error was in fact related to the sender, a different
2871 message to the same recipient with a different sender could confuse
2872 things. In particular, this can happen when sending to a greylisting
2873 server, but other circumstances could also provoke similar problems.
2874 I have changed the default so that the retry time for these errors is now
2875 based a combination of the sender and recipient addresses. This change
2876 can be overridden by setting address_retry_include_sender=false in the
2879 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2880 remote server are returned as part of bounce messages. This was not
2881 happening for LMTP over a pipe (the lmtp transport), but now it is the
2882 same for both kinds of LMTP.
2884 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2885 in header lines that were in fact CNAMEs. This is no longer the case.
2887 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2888 and queue runs started by the daemon processed all messages. This has
2889 been fixed so that -R and -S can now usefully be given with -q<time>.
2891 PH/40 Import PCRE release 6.7 (fixes some bugs).
2893 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2895 PH/42 Give an error if -q is specified more than once.
2897 PH/43 Renamed the variables $interface_address and $interface_port as
2898 $received_ip_address and $received_port, to make it clear that these
2899 values apply to message reception, and not to the outgoing interface when
2900 a message is delivered. (The old names remain recognized, of course.)
2902 PH/44 There was no timeout on the connect() call when using a Unix domain
2903 socket in the ${readsocket expansion. There now is.
2905 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2906 be meaningful with "accept".
2908 SC/01 Eximstats V1.43
2909 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2911 SC/02 Eximstats V1.44
2912 Use a glob alias rather than an array ref in the generated
2913 parser. This improves both readability and performance.
2915 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2916 Collect SpamAssassin and rejection statistics.
2917 Don't display local sender or destination tables unless
2918 there is data to show.
2919 Added average volumes into the top table text output.
2921 SC/04 Eximstats V1.46
2922 Collect data on the number of addresses (recipients)
2923 as well as the number of messages.
2925 SC/05 Eximstats V1.47
2926 Added 'Message too big' to the list of mail rejection
2927 reasons (thanks to Marco Gaiarin).
2929 SC/06 Eximstats V1.48
2930 Mainlog lines which have GMT offsets and are too short to
2931 have a flag are now skipped.
2933 SC/07 Eximstats V1.49 (Alain Williams)
2934 Added the -emptyok flag.
2936 SC/08 Eximstats V1.50
2937 Fixes for obtaining the IP address from reject messages.
2939 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2940 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2941 whitespace changes from 4.64-PH/27
2943 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2946 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2947 are found, allow negative numbers in numeric criteria)
2949 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2951 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2952 to match changes made in 4.64-PH/43
2954 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2956 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2957 "domain" is actually an IP address, and force a failure. This locks out
2958 those revolvers/nameservers that support "A-for-A" lookups, in
2959 contravention of the specifications.
2961 PH/48 When a host name was looked up from an IP address, and the subsequent
2962 forward lookup of the name timed out, the host name was left in
2963 $sender_host_name, contrary to the specification.
2965 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2966 restricted to single-key lookups, Exim was not diagnosing an error if
2967 * or *@ was used with a query-style lookup.
2969 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2971 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2972 long ago, but noone interested enough thought of it. Let's just say that
2973 the "1.1" means that there are some new functions that weren't there at
2974 some point in the past.
2976 PH/51 Error processing for expansion failure of helo_data from an smtp
2977 transport during callout processing was broken.
2979 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2980 tested/used via the -bh/-bhc/-bs options.
2982 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2983 bug, fixed in subsequent PCRE releases).
2985 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2986 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2988 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2994 SC/01 Use a glob alias rather than an array ref in eximstats generated
2995 parser. This improves both readability and performance.
2997 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2998 Don't display local sender or destination tables in eximstats unless
2999 there is data to show.
3000 Added average volumes into the eximstats top table text output.
3002 SC/03 Collect data on the number of addresses (recipients) as well
3003 as the number of messages in eximstats.
3005 TF/01 Correct an error in the documentation for the redirect router. Exim
3006 does (usually) call initgroups() when daemonizing.
3008 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3009 with consistent privilege compared to when running as a daemon.
3011 TF/03 Note in the spec that $authenticated_id is not set for local
3012 submissions from trusted users.
3014 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3015 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3017 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3018 by adding some example configuration directives to the default
3019 configuration file. A little bit of work is required to uncomment the
3020 directives and define how usernames and passwords are checked, but
3021 there is now a framework to start from.
3023 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3024 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3025 without this. I don't know how relevant this is to other LDAP libraries.
3027 PH/02 Add the verb name to the "unknown ACL verb" error.
3029 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3031 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3033 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3034 directory not expanded when it should be if an expanded home directory
3035 was set for the address (which is overridden by the transport).
3037 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3040 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3041 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3042 because it is too late at that time, and has no effect.
3044 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3045 security issue with \' (bugzilla #107). I could not use the
3046 PQescapeStringConn() function, because it needs a PGconn value as one of
3049 PH/08 When testing addresses using -bt, indicate those final addresses that
3050 are duplicates that would not cause an additional delivery. At least one
3051 person was confused, thinking that -bt output corresponded to deliveries.
3052 (Suppressing duplicates isn't a good idea as you lose the information
3053 about possibly different redirections that led to the duplicates.)
3055 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3056 systems where poll() doesn't work, in particular OS X.
3058 PH/10 Added more information to debugging output for retry time not reached.
3060 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3061 operations in malware.c.
3063 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3066 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3067 syslog_duplication set false, log lines that would normally be written
3068 both the the main log and to the reject log were not written to syslog at
3071 PH/14 In the default configuration, change the use of "message" in ACL warn
3072 statements to "add_header".
3074 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3075 not followed by a command (e.g. "seen endif").
3077 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3078 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3081 PH/17 Added extra conditions to the default value of delay_warning_condition
3085 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3086 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3087 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3090 The Auto-Submitted: and various List- headers are standardised, whereas I
3091 don't think Precedence: ever was.
3093 PH/18 Refactored debugging code in route_finduser() to show more information,
3094 in particular, the error code if getpwnam() issues one.
3096 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3097 This is apparently needed in addition to the PH/07 change above to avoid
3098 any possible encoding problems.
3100 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3101 but not after initializing Perl.
3103 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3104 output them only if debugging. By default they are written stderr,
3105 apparently, which is not desirable.
3107 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3110 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3113 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3115 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3116 authenticated or an ident call has been made. Suppress the default
3117 values for $authenticated_id and $authenticated_sender (but permit -oMai
3118 and -oMas) when testing with -bh.
3120 PH/24 Re-jigged the order of the tests in the default configuration so that the
3121 tests for valid domains and recipients precede the DNS black list and CSA
3122 tests, on the grounds that those ones are more expensive.
3124 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3125 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3126 This bug exists in every version of Exim that I still have, right back to
3129 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3130 However, an attempt to turn on case-sensitivity in a regex key by
3131 including (?-i) didn't work because the subject string was already
3132 lowercased, and the effects were non-intuitive. It turns out that a
3133 one-line patch can be used to allow (?-i) to work as expected.
3139 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3140 other effects) broke the use of negated acl sub-conditions.
3142 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3145 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3146 "Deny" causes Exim to reject the incoming connection with a 554 error.
3147 Unfortunately, if there is a major crisis, such as a disk failure,
3148 tcp-wrappers gives "deny", whereas what one would like would be some
3149 kind of temporary error. A kludge has been added to help with this.
3150 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3151 554 error is used if errno is still zero or contains ENOENT (which occurs
3152 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3155 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3157 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3158 errors. Otherwise a message that provokes a temporary error (when other
3159 messages do not) can cause a whole host to time out.
3161 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3162 addresses were routed directly to files or pipes from a redirect router.
3163 File deliveries just didn't batch; pipe deliveries might have suffered
3166 PH/06 A failure to get a lock for a hints database would erroneously always say
3167 "Failed to get write lock", even when it was really a read lock.
3169 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3170 of 0600. This has been changed to use the value of the lockfile_mode
3171 option (which defaults to 0600).
3173 PH/08 Applied small patch from the Sieve maintainer.
3175 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3176 folder from quota calculations, a direct delivery into this folder messed
3177 up the contents of the maildirsize file. This was because the regex was
3178 used only to exclude .Trash (or whatever) when the size of the mailbox
3179 was calculated. There was no check that a delivery was happening into an
3180 excluded directory. This bug has been fixed by ignoring all quota
3181 processing for deliveries into excluded directories.
3183 PH/10 Added the maildirfolder_create_regex option to appendfile.
3189 PH/01 The code for finding all the local interface addresses on a FreeBSD
3190 system running IPv6 was broken. This may well have applied to all BSD
3191 systems, as well as to others that have similar system calls. The broken
3192 code found IPv4 interfaces correctly, but gave incorrect values for the
3193 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3194 that it would not match correctly against @[] and not recognize the IPv6
3197 PH/02 The ipliteral router was not recognizing addresses of the form user@
3198 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3200 PH/03 Added disable_ipv6.
3202 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3203 decoded form, because it is most often used to construct To: headers
3204 lines in autoreplies, and the decoded form may well be syntactically
3205 invalid. However, $reply_address has leading white space removed, and all
3206 newlines turned into spaces so that the autoreply transport does not
3209 PH/05 If group was specified without a user on a router, and no group or user
3210 was specified on a transport, the group from the router was ignored.
3212 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3213 for visible compile-time settings that can be used to change these
3214 numbers, for those that want even more. Backwards compatibility with old
3215 spool files has been maintained. However, going back to a previous Exim
3216 release will lost any variables that are in spool files.
3218 PH/07 Two small changes when running in the test harness: increase delay when
3219 passing a TCP/IP connection to a new process, in case the original
3220 process has to generate a bounce, and remove special handling of
3221 127.0.0.2 (sic), which is no longer necessary.
3223 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3224 be the same on different OS.
3226 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3229 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3230 whether --show-vars was specified or not
3232 JJ/02 exipick: Added support for new ACL variable spool format introduced
3235 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3236 syntactically invalid From: or Reply-to: line, and a filter used this to
3237 generate an autoreply, and therefore failed to obtain an address for the
3238 autoreply, Exim could try to deliver to a non-existent relative file
3239 name, causing unrelated and misleading errors. What now happens is that
3240 it logs this as a hard delivery error, but does not attempt to create a
3243 PH/11 The exinext utility has a -C option for testing purposes, but although
3244 the given file was scanned by exinext itself; it wasn't being passed on
3245 when Exim was called.
3247 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3248 an end-of-file indication when reading a command response.
3250 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3251 compiled. In many other places in Exim, IPv6 addresses are always
3252 recognized, so I have changed this. It also means that IPv4 domain
3253 literals of the form [IPV4:n.n.n.n] are now always recognized.
3255 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3256 used if the router is not running as root, for example, when verifying at
3257 ACL time, or when using -bh. The debugging output from this situation was
3258 non-existent - all you got was a failure to exec. I have made two
3261 (a) Failures to set uid/gid, the current directory, or a process leader
3262 in a subprocess such as that created by queryprogram now generate
3263 suitable debugging output when -d is set.
3265 (b) The queryprogram router detects when it is not running as root,
3266 outputs suitable debugging information if -d is set, and then runs
3267 the subprocess without attempting to change uid/gid.
3269 PH/15 Minor change to Makefile for building test_host (undocumented testing
3272 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3273 additional section of a DNS packet that returns MX or SRV records.
3274 Instead, it always explicitly searches for A/AAAA records. This avoids
3275 major problems that occur when a DNS server includes only records of one
3276 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3277 fixed another bug: if SRV records were looked up and the corresponding
3278 address records were *not* found in the additional section, the port
3279 values from the SRV records were lost.
3281 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3282 using the correct key (the original address) when searching the retry
3283 rules in order to find which one to use for generating the retry hint.
3285 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3286 adding the default one. Similarly, if it contains a Reply-To: header, the
3287 errors_reply_to option, if set, is not used.
3289 PH/19 When calculating a retry time, Exim used to measure the "time since
3290 failure" by looking at the "first failed" field in the retry record. Now
3291 it does not use this if it is later than than the arrival time of the
3292 message. Instead it uses the arrival time. This makes for better
3293 behaviour in cases where some deliveries succeed, thus re-setting the
3294 "first failed" field. An example is a quota failure for a huge message
3295 when small messages continue to be delivered. Without this change, the
3296 "time since failure" will always be short, possible causing more frequent
3297 delivery attempts for the huge message than are intended.
3298 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3300 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3301 $1, $2, $3) because the numerical variables can be reset during some
3302 expansion items (e.g. "match"), thereby losing the authentication data.
3304 PH/21 Make -bV show the size of off_t variables so that the test suite can
3305 decide whether to run tests for quotas > 2G.
3307 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3308 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3309 filecount value is greater than 2G or if a quota value is greater than 2G
3310 on a system where the size of off_t is not greater than 4, a panic error
3313 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3314 never match. The debug and -bh output now contains an explicit error
3315 message indicating a malformed IPv4 address or mask.
3317 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3318 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3319 PH/23 above applies.
3321 PH/25 Do not write to syslog when running in the test harness. The only
3322 occasion when this arises is a failure to open the main or panic logs
3323 (for which there is an explicit test).
3325 PH/26 Added the /no_tell option to "control=freeze".
3327 PH/27 If a host name lookup failed very early in a connection, for example, if
3328 the IP address matched host_lookup and the reverse lookup yielded a name
3329 that did not have a forward lookup, an error message of the form "no IP
3330 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3331 could be logged. Now it outputs the IP address instead of "NULL".
3333 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3334 allows the sender and the authenticated sender to be set when
3335 submitting a message from within Exim. Since child_open_exim() is
3336 documented for local_scan(), the new function should be too.
3338 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3339 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3340 results in an empty string is now treated as unset.
3342 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3344 PH/31 Added sender_verify_fail logging option.
3346 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3347 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3348 tidied the source and removed it altogether.
3350 PH/33 When a queue run was abandoned because the load average was too high, a
3351 log line was always written; now it is written only if the queue_run log
3352 selector is set. In addition, the log line for abandonment now contains
3353 information about the queue run such as the pid. This is always present
3354 in "start" and "stop" lines but was omitted from the "abandon" line.
3356 PH/34 Omit spaces between a header name and the colon in the error message that
3357 is given when verify = headers_syntax fails (if there are lots of them,
3358 the message gets confusing).
3360 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3361 names, as there are now some PTR records that contain slashes. This check
3362 is only to protect against broken name servers that fall over on strange
3363 characters, so the fact that it applies to all lookups doesn't matter.
3365 PH/36 Now that the new test suite is complete, we can remove some of the
3366 special code in Exim that was needed for the old test suite. For example,
3367 sorting DNS records because real resolvers return them in an arbitrary
3368 order. The new test suite's fake resolver always returns records in the
3371 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3372 bounces) except when queue_only is set, to avoid logging races between
3373 the different processes.
3375 PH/38 Panic-die if .include specifies a non-absolute path.
3377 PH/39 A tweak to the "H" retry rule from its user.
3379 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3380 a label. They prevented compilation on older perls.
3382 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3383 a warning to be raised on newish perls.
3385 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3386 on queue. Changes to match documented behaviour of showing count of
3387 messages matching specified criteria.
3389 PH/40 Changed the default ident timeout from 30s to 5s.
3391 PH/41 Added support for the use of login_cap features, on those BSD systems
3392 that have them, for controlling the resources used by pipe deliveries.
3394 PH/42 The content-scanning code uses fopen() to create files in which to put
3395 message data. Previously it was not paying any attention to the mode of
3396 the files. Exim runs with umask(0) because the rest of the code creates
3397 files with open(), and sets the required mode explicitly. Thus, these
3398 files were ending up world-writeable. This was not a big issue, because,
3399 being within the spool directory, they were not world-accessible. I have
3400 created a function called modefopen, which takes an additional mode
3401 argument. It sets umask(777), creates the file, chmods it to the required
3402 mode, then resets the umask. All the relevant calls to fopen() in the
3403 content scanning code have been changed to use this function.
3405 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3406 to 24 hours. This avoids potential overflow problems when processing G
3407 and H retry rules. I suspect nobody ever tinkers with this value.
3409 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3411 PH/45 When the plaintext authenticator is running as a client, the server's
3412 challenges are checked to ensure they are valid base64 strings. By
3413 default, the authentication attempt is cancelled if an invalid string is
3414 received. Setting client_ignore_invalid_base64 true ignores these errors.
3415 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3416 they are received. Thus, the responses can be made to depend on the
3417 challenges. If an invalid string is ignored, an empty string is placed in
3420 PH/46 Messages that are created by the autoreply transport now contains a
3421 References: header, in accordance with RFCs 2822 and 3834.
3423 PH/47 Added authenticated_sender_force to the smtp transport.
3425 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3427 PH/49 Installed latest patch from the Sieve maintainer.
3429 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3430 was also set, the appendfile transport was unnecessarily scanning a
3431 directory of message files (e.g. for maildir delivery) to find the count
3432 of files (along with the size), even though it did not need this
3433 information. It now does the scan only if it needs to find either the
3434 size of the count of files.
3436 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3438 PH/52 Two bugs concerned with error handling when the smtp transport is
3441 (i) Exim was not creating retry information for temporary errors given
3442 for individual recipients after the DATA command when the smtp transport
3443 was used in LMTP mode. This meant that they could be retried too
3444 frequently, and not timed out correctly.
3446 (ii) Exim was setting the flag that allows error details to be returned
3447 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3448 recipients that were returned after the DATA command.
3450 PH/53 This is related to PH/52, but is more general: for any failing address,
3451 when detailed error information was permitted to be returned to the
3452 sender, but the error was temporary, then after the final timeout, only
3453 "retry timeout exceeded" was returned. Now it returns the full error as
3454 well as "retry timeout exceeded".
3456 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3457 do this, and (what is worse) MTAs that accept it.
3459 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3460 will now be deprecated.
3462 PH/56 New os.c-cygwin from the Cygwin maintainer.
3464 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3465 formats (previously only available in exim formats via -bpr, -bpru,
3466 and -bpra. Now also available in native and exiqgrep formats)
3468 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3469 with very large, slow to parse queues
3471 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3473 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3475 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3476 responses to authentication challenges, though it was showing the
3477 challenges; (ii) I've removed the CR characters from the debug output for
3480 PH/58 Allow for the insertion of a newline as well as a space when a string
3481 is turned into more than one encoded-word during RFC 2047 encoding. The
3482 Sieve code now uses this.
3484 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3485 data_4xx, lost_connection, tls_required.
3487 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3488 message was being sent as an SMTP response.
3490 PH/61 Add -l and -k options to exicyclog.
3492 PH/62 When verifying, if an address was redirected to one new address, so that
3493 verification continued, and the new address failed or deferred after
3494 having set something in $address_data, the value of $address_data was not
3495 passed back to the ACL. This was different to the case when no
3496 redirection occurred. The value is now passed back in both cases.
3498 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3499 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3500 use login_cap.h, so on its own it isn't the distinguishing feature. The
3501 new name refers directly to the setclassresources() function.
3503 PH/65 Added configuration files for NetBSD3.
3505 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3507 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3508 is preferred over IPv4.
3510 PH/68 The bounce_return_message and bounce_return_body options were not being
3511 honoured for bounces generated during the reception of non-SMTP messages.
3512 In particular, this applied to messages rejected by the ACL. This bug has
3513 been fixed. However, if bounce_return_message is true and bounce_return_
3514 body is false, the headers that are returned for a non-SMTP message
3515 include only those that have been read before the error was detected.
3516 (In the case of an ACL rejection, they have all been read.)
3518 PH/69 The HTML version of the specification is now built in a directory called
3519 spec_html instead of spec.html, because the latter looks like a path with
3520 a MIME-type, and this confuses some software.
3522 PH/70 Catch two compiler warnings in sieve.c.
3524 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3525 function verify_get_ident() calls ip_connect() to connect a socket, but
3526 if the "connect()" function timed out, ip_connect() used to close the
3527 socket. However, verify_get_ident() also closes the socket later, and in
3528 between Exim writes to the log, which may get opened at this point. When
3529 the socket was closed in ip_connect(), the log could get the same file
3530 descriptor number as the socket. This naturally causes chaos. The fix is
3531 not to close the socket in ip_connect(); the socket should be closed by
3532 the function that creates it. There was only one place in the code where
3533 this was missing, in the iplookup router, which I don't think anybody now
3534 uses, but I've fixed it anyway.
3536 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3537 well as to direct DNS lookups. Otherwise the handling of names in host
3538 lists is inconsistent and therefore confusing.
3544 PH/01 Two changes to the default runtime configuration:
3546 (1) Move the checks for relay_from_hosts and authenticated clients from
3547 after to before the (commented out) DNS black list checks.
3549 (2) Add control=submission to the relay_from_hosts and authenticated
3550 clients checks, on the grounds that messages accepted by these
3551 statements are most likely to be submissions.
3553 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3555 (1) Generate an error if the third argument for the ${prvs expansion is
3558 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3561 (3) Reset the variables that are obtained from the first argument of
3562 ${prvscheck and used in the second argument before leaving the code,
3563 because their memory is reclaimed, so using them afterwards may do
3566 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3567 one (it's much easier than Tom thought :-).
3569 (5) Because of (4), we can now allow for the use of $prvscheck_result
3570 inside the third argument.
3572 PH/03 For some reason, the default setting of PATH when running a command from
3573 a pipe transport was just "/usr/bin". I have changed it to
3576 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3577 anything to be listed in the output from -bV.
3579 PH/05 When a filter generated an autoreply, the entire To: header line was
3580 quoted in the delivery log line, like this:
3582 => >A.N.Other <ano@some.domain> <original@ddress> ...
3584 This has been changed so that it extracts the operative address. There
3585 may be more than one such address. If so, they are comma-separated, like
3588 => >ano@some.domain,ona@other.domain <original@ddress> ...
3590 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3591 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3592 not being looked up in the rDNS to get a host name, Exim was showing the
3593 IP address twice in Received: lines, even though the IP addresses were
3594 identical. For example:
3596 Received: from [1.2.3.4] (helo=[1.2.3.4])
3598 However, if the real host name was known, it was omitting the HELO data
3599 if it matched the actual IP address. This has been tidied up so that it
3600 doesn't show the same IP address twice.
3602 PH/07 When both +timestamp and +memory debugging was on, the value given by
3603 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3604 called by the debug printing, thereby overwriting the timestamp buffer.
3605 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3607 PH/08 When the original message was included in an autoreply transport, it
3608 always said "this is a copy of the message, including all the headers",
3609 even if body_only or headers_only was set. It now gives an appropriate
3612 PH/09 Applied a patch from the Sieve maintainer which:
3614 o fixes some comments
3615 o adds the (disabled) notify extension core
3616 o adds some debug output for the result of if/elsif tests
3617 o points to the current vacation draft in the documentation
3618 and documents the missing references header update
3622 o fixes a bug in processing the envelope test (when testing
3623 multiple envelope elements, the last element determined the
3626 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3627 Electronic Mail") by including:
3629 Auto-submitted: auto-generated
3631 in the messages that it generates (bounce messages and others, such as
3632 warnings). In the case of bounce messages for non-SMTP messages, there was
3633 also a typo: it was using "Auto_submitted" (underscore instead of
3634 hyphen). Since every message generated by Exim is necessarily in response
3635 to another message, thes have all been changed to:
3637 Auto-Submitted: auto-replied
3639 in accordance with these statements in the RFC:
3641 The auto-replied keyword:
3643 - SHOULD be used on messages sent in direct response to another
3644 message by an automatic process,
3646 - MUST NOT be used on manually-generated messages,
3648 - MAY be used on Delivery Status Notifications (DSNs) and Message
3649 Disposition Notifications (MDNs),
3651 - MUST NOT be used on messages generated by automatic or periodic
3652 processes, except for messages which are automatic responses to
3655 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3656 to the default Received: header definition.
3658 PH/12 Added log selector acl_warn_skipped (default on).
3660 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3661 variables because (a) they are in the wrong storage pool and (b) even if
3662 they were copied, it wouldn't work properly because of the caching.
3664 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3665 checking when decoding. Apparently there are clients that generate
3666 overlong encoded strings. Why am I not surprised?
3668 PH/15 If the first argument of "${if match_address" was not empty, but did not
3669 contain an "@" character, Exim crashed. Now it writes a panic log message
3670 and treats the condition as false.
3672 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3674 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3675 extension "envelope-auth". The code is finished and in agreement with
3676 other implementations, but there is no documentation so far and in fact,
3677 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3678 not changing the active code.
3680 Print executed "if" and "elsif" statements when debugging is used. This
3681 helps a great deal to understand what a filter does.
3683 Document more things not specified clearly in RFC3028. I had all this
3684 sorted out, when out of a sudden new issues came to my mind. Oops."
3686 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3689 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3690 canonical form (as documented). However, after a host name lookup from
3691 the IP address, check_host() was doing a simple string comparison with
3692 addresses acquired from the DNS when checking that the found name did
3693 have the original IP as one of its addresses. Since any found IPv6
3694 addresses are likely to be in abbreviated form, the comparison could
3695 fail. Luckily, there already exists a function for doing the comparison
3696 by converting both addresses to binary, so now that is used instead of
3697 the text comparison.
3699 PH/20 There was another similar case to PH/19, when a complete host name was
3700 given in a host list; looking up its IP address could give an abbreviated
3701 form, whereas the current host's name might or might not be abbreviated.
3702 The same fix has been applied.
3708 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3709 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3712 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3713 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3715 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3717 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3718 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3719 Also, for "Auto-Submitted", treat anything other than "no" as
3720 non-personal, in accordance with RFC 3834. (Previously it treated
3721 anything starting "auto-" as non-personal.)
3723 TF/01 The control=submission/name=... option had a problem with syntax
3724 errors if the name included a slash character. The /name= option
3725 now slurps the rest of the string, so it can include any characters
3726 but it must come last in the list of options (after /sender_retain
3729 PH/05 Some modifications to the interface to the fake nameserver for the new
3737 TK/01 Added the "success_on_redirect" address verification option. See
3738 NewStuff for rationale and an example.
3740 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3742 PH/02 Patch to exigrep to allow it to work on syslog lines.
3744 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3745 fread() to skip over the body file's header line, because in Cygwin the
3746 header line is locked and is inaccessible.
3748 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3749 co-exist for some time) to make it clear that it is the Exim ID that is
3750 referenced, not the Message-ID: header line.
3752 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3753 string_format() function, because snprintf() does not exist on all
3756 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3757 ${stat: expansion item.
3759 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3760 protocol synchronization error", to keep the pedants happy.
3762 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3763 well as for IRIX systems, when gcc is being used. See the host.c source
3766 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3768 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3771 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3772 transport and to the smtp transport in LMTP mode.
3774 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3776 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3777 run a filter in a subprocess. This could lead to confusion in subsequent
3778 lookups in the parent process. There should also be a search_tidyup() at
3779 the end of the subprocess.
3781 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3782 only if the host matched helo_try_verify_hosts, which caused the
3783 verification to occur when the EHLO/HELO command was issued. The ACL just
3784 tested the remembered result. Now, if a previous verification attempt has
3785 not happened, "verify = helo" does it there and then.
3787 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3789 TK/03 Fix log output including CR from clamd.
3791 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3792 exist provoked a memory error which could cause a segfault.
3794 PH/15 Installed PCRE 6.2
3796 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3798 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3799 of the problem. Specifically, suggested +O2 rather than +O1 for the
3802 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3804 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3805 record, the port was not correctly logged when the outgoing_port log
3806 selector was set (it logged the transort's default port).
3808 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3809 fallback_hosts, and "hosts" in the smtp transport.
3811 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3812 host errors such as "Connection refused".
3814 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3815 authentication with radiusclient 0.4.9:
3817 - Error returned from rc_read_config was caught wrongly
3818 - Username/password not passed on to radius server due to wrong length.
3820 The presumption is that some radiusclient API changes for 4.51/PH/17
3821 were not taken care of correctly. The code is still untested by me (my
3822 Linux distribution still has 0.3.2 of radiusclient), but it was
3823 contributed by a Radius user.
3825 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3826 expanding the "port" option of the smtp transport.
3828 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3829 while reading a MIME header. Thanks to Tom Hughes for a patch.
3831 PH/24 Include config.h inside local_scan.h so that configuration settings are
3834 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3835 that in an ACL for RCPT (for example), you can examine exactly what was
3838 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3839 commands, but it was not correctly comparing the address with the actual
3840 client host address. Thus, it would show the EHLO address in Received:
3841 header lines when this was not necessary.
3843 PH/27 Added the % operator to ${eval:}.
3845 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3846 it should be ignoring failures (because with -C, for example, it has lost
3847 privilege). It wasn't ignoring creation failures other than "already
3850 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3853 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3854 because an input line was too long, either on its own, or by virtue of
3855 too many continuations, the temporary file was not being removed, and the
3856 return code was incorrect.
3858 PH/31 Missing "BOOL" in function definition in filtertest.c.
3860 PH/32 Applied Sieve patches from the maintainer.
3862 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3864 PH/33 Added "verify = not_blind".
3866 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3867 Local/Makefile (with some defaults set). These are used in built scripts
3868 such as exicyclog, but they have never been used in the exim_install
3869 script (though there are many overriding facilities there). I have
3870 arranged that the exim_install script now takes note of these two
3873 PH/35 Installed configuration files for Dragonfly.
3875 PH/36 When a locally submitted message by a trusted user did not contain a
3876 From: header, and the sender address was obtained from -f or from an SMTP
3877 MAIL command, and the trusted user did not use -F to supply a sender
3878 name, $originator_name was incorrectly used when constructing a From:
3879 header. Furthermore, $originator_name was used for submission mode
3880 messages from external hosts without From: headers in a similar way,
3881 which is clearly wrong.
3883 PH/37 Added control=suppress_local_fixups.
3885 PH/38 When log_selector = +received_sender was set, and the addition of the
3886 sender made the log line's construction buffer exactly full, or one byte
3887 less than full, an overflow happened when the terminating "\n" was
3890 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3891 when the result of a list match is failure because a DNS lookup failed.
3893 PH/40 RM_COMMAND is now used in the building process.
3895 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3896 the "build-* directories that it finds.
3898 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3899 address in a domain literal was a prefix of an interface address.
3901 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3902 when verifying a sender address, unless rewrite_headers is false.
3904 PH/44 Wrote a long comment about why errors_to addresses are verified as
3905 recipients, not senders.
3907 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3908 the ratelimit ACL was added.
3910 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3912 PH/46 Added extra information about PostgreSQL errors to the error string.
3914 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3915 avoiding the need to install special zones in a real server. This is
3916 backwards compatible; if it can't find the fake resolver, it drops back.
3917 Thus, both old and new test suites can be run.
3919 TF/02 Added util/ratelimit.pl
3921 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3922 clock is set back in time.
3924 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3925 Candler <B.Candler@pobox.com>.
3927 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3928 OK for addresses that are the result of redirections.
3930 PH/48 A number of further additions for the benefit of the new test suite,
3931 including a fake gethostbyname() that interfaces to the fake DNS resolver
3934 TF/06 The fix for widen_domains has also been applied to qualify_single and
3935 search_parents which are the other dnslookup options that can cause
3938 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3941 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3943 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3944 service" if it encountered an MX record with an empty target hostname.
3945 The message is now "an MX or SRV record indicated no SMTP service".
3947 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3948 if the DNS of the sending site is misconfigured. This is quite a
3949 common situation. This change restores the behaviour of treating a
3950 helo verification defer as a failure.
3952 PH/51 If self=fail was set on a router, the bounce message did not include the
3953 actual error message.
3959 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3961 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3962 command itself ended in error, the underlying message about the transport
3963 filter timeout was being overwritten with the pipe command error. Now the
3964 underlying error message should be appended to the second error message.
3966 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3968 PH/02 Reduce the amount of output that "make" produces by default. Full output
3969 can still be requested.
3971 PH/03 The warning log line about a condition test deferring for a "warn" verb
3972 was being output only once per connection, rather than after each
3973 occurrence (because it was using the same function as for successful
3974 "warn" verbs). This seems wrong, so I have changed it.
3976 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3977 it should not have, which might have caused a crash in the right
3978 circumstances, but probably never did.
3980 PH/04 Installed a modified version of Tony Finch's patch to make submission
3981 mode fix the return path as well as the Sender: header line, and to
3982 add a /name= option so that you can make the user's friendly name appear
3985 TF/03 Added the control = fakedefer ACL modifier.
3987 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3988 Mark Lowes for thorough testing.
3990 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3992 TK/03 Merged latest SRS patch from Miles Wilton.
3994 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3995 with the definition in sysexits.h (which is #included earlier).
3996 Fortunately, Exim does not actually use EX_OK. The code used to try to
3997 preserve the sysexits.h value, by assuming that macro definitions were
3998 scanned for macro replacements. I have been disabused of this notion,
3999 so now the code just undefines EX_OK before #including unistd.h.
4001 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4002 in the smtp transport. When a block could not be written in a single
4003 write() function, the timeout was being re-applied to each part-write.
4004 This seems wrong - if the receiver was accepting one byte at a time it
4005 would take for ever. The timeout is now adjusted when this happens. It
4006 doesn't have to be particularly precise.
4008 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4009 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4011 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4012 without a domain if the check to <postmaster@domain> fails.
4014 SC/01 Eximstats: added -xls and the ability to specify output files
4015 (patch written by Frank Heydlauf).
4017 SC/02 Eximstats: use FileHandles for outputting results.
4019 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4021 SC/04 Eximstats: fixed display of large numbers with -nvr option
4023 SC/05 Eximstats: fixed merging of reports with empty tables.
4025 SC/06 Eximstats: added the -include_original_destination flag
4027 SC/07 Eximstats: removed tabs and trailing whitespace.
4029 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4031 TK/06 MBOX spool code: Add real "From " MBOX separator line
4032 so the .eml file is really in mbox format (even though
4033 most programs do not really care). Patch from Alex Miller.
4035 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4036 The latter is generated from $received_to and is only set if the
4037 message has one envelope recipient. SA can use these headers,
4038 obviously out-of-the-box. Patch from Alex Miller.
4040 PH/08 The ${def test on a variable was returning false if the variable's
4041 value was "0", contrary to what the specification has always said!
4042 The result should be true unless the variable is empty.
4044 PH/09 The syntax error of a character other than { following "${if
4045 def:variable_name" (after optional whitespace) was not being diagnosed.
4046 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4047 accidental colon was present, for example, could give incorrect results.
4049 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4050 result is used (not including appendfile, where other changes are about
4053 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4054 This involved changing a lot of size variables from int to off_t. It
4055 should work with maildirs and everything.
4057 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4058 spamd dying while we are connected to it.
4060 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4063 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4064 function for BDB 4.3.
4066 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4068 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4069 the sources for PCRE have been re-arranged and more files are now
4072 PH/15 The code I had for printing potentially long long variables in PH/11
4073 above was not the best (it lost precision). The length of off_t variables
4074 is now inspected at build time, and an appropriate printing format (%ld
4075 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4076 to be "long long int" or "long int". This is needed for the internal
4077 formatting function string_vformat().
4079 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4080 the configuration file to be ":syslog", then the script "guesses" where
4081 the logs files are, rather than using the compiled in default. In our
4082 case the guess is not the same as the compiled default, so the script
4083 suddenly stopped working when I started to use syslog. The patch checks
4084 to see if log_file_path is "". If so, it attempts to read it from exim
4085 with no configuration file to get the compiled in version, before it
4086 falls back to the previous guessing code."
4088 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4089 implementing BATV in an Exim configuration. See NewStuff for the gory
4092 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4093 Makefile that are specific to HP-UX.
4095 PH/18 If the "use_postmaster" option was set for a recipient callout together
4096 with the "random" option, the postmaster address was used as the MAIL
4097 FROM address for the random test, but not for the subsequent recipient
4098 test. It is now used for both.
4100 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4101 patch removes a few documentation additions to RFC 3028, because the
4102 latest draft now contains them. It adds the new en;ascii-case comparator
4103 and a new error check for 8bit text in MIME parts. Comparator and
4104 require names are now matched exactly. I enabled the subaddress
4105 extension, but it is not well tested yet (read: it works for me)."
4107 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4108 rework some of the code of TK/09 above to avoid the hardwired use of
4109 "%lld" and "long long". Replaced the call to snprintf() with a call to
4112 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4113 records point to non-existent hosts", "retry timeout exceeded", and
4114 "retry time not reached for any host after a long failure period".
4116 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4117 experimental DomainKeys support:
4119 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4120 (2) On an error such as an illegally used "control", the wrong name for
4121 the control was given.
4123 These problems did NOT occur unless DomainKeys support was compiled.
4125 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4127 PH/24 Added ${if match_ip condition.
4129 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4130 databases so that it will be absolutely obvious if a crash occurs in the
4131 DB library. This is a regular occurrence (often caused by mis-matched
4134 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4135 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4136 picked up on a user's system that detects such things. There doesn't seem
4137 to be a gcc warning option for this - only an attribute that has to be
4138 put on the function's prototype. It seems that in Fedora Core 4 they have
4139 set this on a number of new functions. No doubt there will be more in due
4142 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4143 specify a transport. However, if an address that was verified by such a
4144 router was the subject of a callout, Exim crashed because it tried to
4145 read the rcpt_include_affixes from the non-existent transport. Now it
4146 just assumes that the setting of that option is false. This bug was
4147 introduced by 4.51/PH/31.
4149 PH/28 Changed -d+all to exclude +memory, because that information is very
4150 rarely of interest, but it makes the output a lot bigger. People tend to
4151 do -d+all out of habit.
4153 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4154 code in os-type was giving problems when libc.so lives in lib64, like on
4157 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4158 aren't the modern standard, and it seems that some systems' include files
4159 don't always have them. Exim was already checking for some of the newer
4160 ones like T_AAAA, and defining it itself. I've added checks for all the
4161 record types that Exim uses.
4163 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4164 not automatically generating a new one, as it is supposed to. This
4165 prevented TLS from working. If the file did exist, but contained invalid
4166 data, a new version was generated, as expected. It was only the case of a
4167 non-existent file that was broken.
4169 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4170 with a change in libdomainkeys > 0.64.
4172 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4173 from DNS. If the selector record carries the flag, it now has
4174 precedence over the domain-wide flag.
4176 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4178 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4179 the use of a port name that isn't defined in /etc/services) occurred, the
4180 message was deferred as in a normal delivery, and thus remained on the
4181 spool, instead of being failed because of the mua_wrapper setting. This
4182 is now fixed, and I tidied up some of the mua_wrapper messages at the
4185 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4186 the messages in a hash of arrays rather than using individual hashes.
4187 This is a bit cleaner and results in dramatic memory savings, albeit
4188 at a slight CPU cost.
4190 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4191 as requested by Marc Sherman.
4193 SC/10 Eximstats: added histograms for user specified patterns as requested
4196 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4198 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4199 fopen() in the content-scanning modules that did not already have it.
4205 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4206 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4208 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4210 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4212 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4213 to test Sieve filters that use "vacation".
4215 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4216 that changes the way the GnuTLS parameters are stored in the cache file.
4217 The new format can be generated externally. For backward compatibility,
4218 if the data in the cache doesn't make sense, Exim assumes it has read an
4219 old-format file, and it generates new data and writes a new file. This
4220 means that you can't go back to an older release without removing the
4223 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4224 work if there are any delivery delays because "one_time" forces the
4225 parent to be marked "delivered", so its unseen clone is never tried
4226 again. For this reason, Exim now forbids the simultaneous setting of
4229 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4230 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4231 entry if you want to know the details. The fix, however, neglected to
4232 consider the case when local delivery batching is involved. The test for
4233 "previously delivered" was not happening when checking to see if an
4234 address could be batched with a previous (undelivered) one; under
4235 certain circumstances this could lead to multiple deliveries to the same
4238 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4239 in its include files, and this causes problems building Exim.
4241 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4242 header_syntax) but Exim was just ignoring anything given after a slash.
4243 In particular, this caused confusion with an attempt to use "verify =
4244 reverse_host_lookup/defer_ok". An error is now given when options are
4245 supplied for verify items that do not have them. (Maybe reverse_host_
4246 lookup should have a defer_ok option, but that's a different point.)
4248 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4249 defined by RFC 821) to 2048, because there were problems with some AUTH
4250 commands, and RFC 1869 says the size should be increased for extended
4251 SMTP commands that take arguments.
4253 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4256 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4257 "unknown" error; now it says that the functionality isn't in the binary.
4259 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4260 an address' error message when a string expansion fails (syntax or
4261 whatever). Otherwise the password may appear in the log. Following change
4262 PH/42 below, there is no longer a chance of it appearing in a bounce
4265 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4267 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4268 of its addresses was ever tried. (Bugzilla bug #2.)
4270 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4271 the result incorrectly in the debug output. (It correctly added a newline
4272 to what was transported.)
4274 TF/01 Added $received_time.
4276 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4277 commented out examples of how to interface to a virus scanner and to
4278 SpamAssassin. Also added commented examples of av_scanner and
4279 spamd_address settings.
4281 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4282 and controls are allowed in which ACLs. There were a couple of minor
4283 errors. Some of the entries in the conditions table (which is a table of
4284 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4285 negation of where the condition IS allowed.
4287 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4289 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4290 header file does not have a version number, so I've had to invent a new
4291 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4292 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4293 radiusclient), but it was contributed by a Radius user.
4295 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4296 files or directories, for OpenSSL.
4298 PH/19 When an Exim process that is running as root has to create an Exim log
4299 file, it does so in a subprocess that runs as exim:exim so as to get the
4300 ownership right at creation (otherwise, other Exim processes might see
4301 the file with the wrong ownership). There was no test for failure of this
4302 fork() call, which would lead to the process getting stuck as it waited
4303 for a non-existent subprocess. Forks do occasionally fail when resources
4304 run out. I reviewed all the other calls to fork(); they all seem to check
4307 PH/20 When checking for unexpected SMTP input at connect time (before writing
4308 the banner), Exim was not dealing correctly with a non-positive return
4309 from the read() function. If the client had disconnected by this time,
4310 the result was a log entry for a synchronization error with an empty
4311 string after "input=" when read() returned zero. If read() returned -1
4312 (an event I could not check), uninitialized data bytes were printed.
4313 There were reports of junk text (parts of files, etc) appearing after
4316 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4318 PH/22 Added support for macro redefinition, and (re)definition in between
4319 driver and ACL definitions.
4321 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4322 forgetting to use the resulting value; it was using the unexpanded value.
4324 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4325 hadn't been configured. The fix is from Juergen Kreileder, who
4326 understands it better than I do:
4328 "Here's what I see happening with three configured cyrus_sasl
4329 authenticators configured (plain, login, cram-md5):
4331 On startup auth_cyrus_sasl_init() gets called for each of these.
4332 This means three calls to sasl_listmech() without a specified mech_list.
4333 => SASL tests which mechs of all available mechs actually work
4334 => three warnings about OTP not working
4335 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4337 With the patch, sasl_listmech() also gets called three times. But now
4338 SASL's mech_list option is set to the server_mech specified in the the
4339 authenticator. Or in other words, the answer from sasl_listmech()
4340 gets limited to just the mech you're testing for (which is different
4342 => the return list contains just 'plain' or 'login', 'cram-md5' or
4343 nothing depending on the value of ob->server_mech.
4345 I've just tested the patch: Authentication still works fine,
4346 unavailable mechs specified in the exim configuration are still
4347 caught, and the auth.log warnings about OTP are gone."
4349 PH/25 When debugging is enabled, the contents of the command line are added
4350 to the debugging output, even when log_selector=+arguments is not
4353 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4354 answer is "GNU", and only if the return is "GNU/something" is the answer
4357 PH/27 $acl_verify_message is now set immediately after the failure of a
4358 verification in an ACL, and so is available in subsequent modifiers. In
4359 particular, the message can be preserved by coding like this:
4361 warn !verify = sender
4362 set acl_m0 = $acl_verify_message
4364 Previously, $acl_verify_message was set only while expanding "message"
4365 and "log_message" when a very denied access.
4367 PH/28 Modified OS/os.c-Linux with
4369 -#ifndef OS_LOAD_AVERAGE
4370 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4372 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4373 nomenclature these days.)
4375 PH/29 Installed patch from the Sieve maintainer that adds the options
4376 sieve_useraddress and sieve_subaddress to the redirect router.
4378 PH/30 In these circumstances:
4379 . Two addresses routed to the same list of hosts;
4380 . First host does not offer TLS;
4381 . First host accepts first address;
4382 . First host gives temporary error to second address;
4383 . Second host offers TLS and a TLS session is established;
4384 . Second host accepts second address.
4385 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4386 and peerdn, if requested) that were in fact used only for the second
4389 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4390 attention to any local part prefix or suffix that was matched by the
4391 router that accepted the address. It now behaves in the same way as it
4392 does for delivery: the affixes are removed from the local part unless
4393 rcpt_include_affixes is set on the transport.
4395 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4396 timeout during the DATA phase of an incoming message.
4398 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4399 applied a patch sanctioned by the Sieve maintainer.
4401 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4402 the uid or gid is negative. A case of a negative gid caused this to be
4403 noticed. The fix allows for either to be negative.
4405 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4406 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4409 PH/36 The change PH/12 above was broken. Fixed it.
4411 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4412 the grounds that routing the same address twice would always produce the
4413 same answer. This might have been true once, but it is certainly no
4414 longer true now. Routing a child address may depend on the previous
4415 routing that produced that child. Some complicated redirection strategies
4416 went wrong when messages had multiple recipients, and made Exim's
4417 behaviour dependent on the order in which the addresses were given.
4419 I have moved the duplicate checking until after the routing is complete.
4420 Exim scans the addresses that are assigned to local and remote
4421 transports, and removes any duplicates. This means that more work will be
4422 done, as duplicates will always all be routed, but duplicates are
4423 presumably rare, so I don't expect this is of any significance.
4425 For deliveries to pipes, files, and autoreplies, the duplicate checking
4426 still happens during the routing process, since they are not going to be
4429 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4430 It corrects a timeout issue with spamd. This is Ian's comment: "The
4431 background is that sometimes spamd either never reads data from a
4432 connection it has accepted, or it never writes response data. The exiscan
4433 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4434 blindly assumes that writes won't block so it may never time out."
4436 PH/39 Allow G after quota size as well as K and M.
4438 PH/40 The value set for $authenticated_id in an authenticator may not contain
4439 binary zeroes or newlines because the value is written to log lines and
4440 to spool files. There was no check on this. Now the value is run through
4441 the string_printing() function so that such characters are converted to
4442 printable escape sequences.
4444 PH/41 $message_linecount is a new variable that contains the total number of
4445 lines in the message. Compare $body_linecount, which is the count for the
4448 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4449 bounce and delay warning messages, except in certain special cases, which
4452 (a) An SMTP error message from a remote host;
4453 (b) A message specified in a :fail: redirection;
4454 (c) A message specified in a "fail" command in a system filter;
4455 (d) A message specified in a FAIL return from the queryprogram router;
4456 (e) A message specified by the cannot_route_message router option.
4458 In these cases only, Exim does include the error details in bounce and
4459 warning messages. There are also a few cases where bland messages such
4460 as "unrouteable address" or "local delivery error" are given.
4462 PH/43 $value is now also set for the "else" part of a ${run expansion.
4464 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4465 being worked on, but at least Exim now implements the latest version to
4468 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4469 process to complete was treated as a delivery failure, a timeout while
4470 writing the message to the pipe was logged, but erroneously treated as a
4471 successful delivery. Such timeouts include transport filter timeouts. For
4472 consistency with the overall process timeout, these timeouts are now
4473 treated as errors, giving rise to delivery failures by default. However,
4474 there is now a new Boolean option for the pipe transport called
4475 timeout_defer, which, if set TRUE, converts the failures into defers for
4476 both kinds of timeout. A transport filter timeout is now identified in
4479 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4480 systems where "make" and "gmake" are different, calling "gmake" at top
4481 level broke things. I've arranged for the value of $(MAKE) to be passed
4482 from the Makefile to this script so that it can call the same version of
4486 A note about Exim versions 4.44 and 4.50
4487 ----------------------------------------
4489 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4490 changes of various kinds. As a consequence, a big documentation update was
4491 needed. This delayed the release for rather longer than seemed good, especially
4492 in the light of a couple of (minor) security issues. Therefore, the changes
4493 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4494 release. So 4.44 and 4.50 are in effect two different branches that both start
4497 I have left the 4.50 change log unchanged; it contains all the changes since
4498 4.43. The change log for 4.44 is below; many of its items are identical to
4499 those for 4.50. This seems to be the most sensible way to preserve the
4500 historical information.
4506 1. Minor wording change to the doc/README.SIEVE file.
4508 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4509 computation of the current number of files was incorrect.
4511 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4512 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4515 4. Give more explanation in the error message when the command for a transport
4516 filter fails to execute.
4518 5. There are several places where Exim runs a non-Exim command in a
4519 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4520 was being done only for the command run by the queryprogram router. It is
4521 now done for all such subprocesses. The other cases are: ${run, transport
4522 filters, and the commands run by the lmtp and pipe transports.
4524 6. Added CONFIGURE_GROUP build-time option.
4526 7. Some older OS have a limit of 256 on the maximum number of file
4527 descriptors. Exim was using setrlimit() to set 1000 as a large value
4528 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4529 systems. I've change it so that if it can't get 1000, it tries for 256.
4531 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4532 was an oversight, and furthermore, ever since the addition of extra
4533 controls (e.g. 4.43/32), the checks on when to allow different forms of
4534 "control" were broken. There should now be diagnostics for all cases when a
4535 control that does not make sense is encountered.
4537 9. Added the /retain_sender option to "control=submission".
4539 10. $recipients is now available in the predata ACL (oversight).
4541 11. Tidy the search cache before the fork to do a delivery from a message
4542 received from the command line. Otherwise the child will trigger a lookup
4543 failure and thereby defer the delivery if it tries to use (for example) a
4544 cached ldap connection that the parent has called unbind on.
4546 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4547 of $address_data from the recipient verification was clobbered by the
4548 sender verification.
4550 13. The value of address_data from a sender verification is now available in
4551 $sender_address_data in subsequent conditions in the ACL statement.
4553 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4555 15. Added a new option "connect=<time>" to callout options, to set a different
4558 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4559 was its contents. (It was OK if the option was not defined at all.)
4561 17. A "Completed" log line is now written for messages that are removed from
4562 the spool by the -Mrm option.
4564 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4565 information about exactly what failed.
4567 19. Added -dd to debug only the daemon process.
4569 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4570 handles timeouts, both on the server side and network timeouts. Renamed the
4571 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4573 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4574 It is now set to "smtps".
4576 22. $host_address is now set to the target address during the checking of
4577 ignore_target_hosts.
4579 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4580 being passed; this would have caused $sender_host_name to have been used if
4581 matching the list had actually called for a host name (not very likely,
4582 since this list is usually IP addresses). A host name is now passed as
4585 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4586 code that specifies a non-restarting handler (typically sigaction() in
4587 modern systems) in an attempt to fix a rare and obscure crash bug.
4589 25. Narrowed the window for a race in the daemon that could cause it to ignore
4590 SIGCHLD signals. This is not a major problem, because they are used only to
4591 wake it up if nothing else does.
4593 26. A malformed maildirsize file could cause Exim to calculate negative values
4594 for the mailbox size or file count. Odd effects could occur as a result.
4595 The maildirsize information is now recalculated if the size or filecount
4598 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4599 support for a long time. Removed HAVE_SYS_VFS_H.
4601 28. Installed the latest version of exipick from John Jetmore.
4603 29. In an address list, if the pattern was not a regular expression, an empty
4604 subject address (from a bounce message) matched only if the pattern was an
4605 empty string. Non-empty patterns were not even tested. This was the wrong
4606 because it is perfectly reasonable to use an empty address as part of a
4607 database query. An empty address is now tested by patterns that are
4608 lookups. However, all the other forms of pattern expect the subject to
4609 contain a local part and a domain, and therefore, for them, an empty
4610 address still always fails if the pattern is not itself empty.
4612 30. Exim went into a mad DNS loop when attempting to do a callout where the
4613 host was specified on an smtp transport, and looking it up yielded more
4614 than one IP address.
4616 31. Re-factored the code for checking spool and log partition space into a
4617 function that finds that data and another that does the check. The former
4618 is then used to implement four new variables: $spool_space, $log_space,
4619 $spool_inodes, and $log_inodes.
4621 32. The RFC2047 encoding function was originally intended for short strings
4622 such as real names; it was not keeping to the 75-character limit for
4623 encoded words that the RFC imposes. It now respects the limit, and
4624 generates multiple encoded words if necessary. To be on the safe side, I
4625 have increased the buffer size for the ${rfc2047: expansion operator from
4628 33. It is now permitted to omit both strings after an "if" condition; if the
4629 condition is true, the result is "true". As before, when the second string
4630 is omitted, a false condition yields an empty string. This makes it less
4631 cumbersome to write custom ACL and router conditions.
4633 34. Failure to deliver a bounce message always caused it to be frozen, even if
4634 there was an errors_to setting on the router. The errors_to setting is now
4637 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4638 canonical form (fully expanded) before being placed in
4639 $sender_host_address.
4641 36. The table in the code that translates DNS record types into text (T_A to
4642 "A" for instance) was missing entries for NS and CNAME. It is just possible
4643 that this could have caused confusion if both these types were looked up
4644 for the same domain, because the text type is used as part of Exim's
4645 per-process caching. But the chance of anyone hitting this buglet seems
4648 37. The dnsdb lookup has been extended in a number of ways.
4650 (1) There is a new type, "zns", which walks up the domain tree until it
4651 finds some nameserver records. It should be used with care.
4653 (2) There is a new type, "mxh", which is like "mx" except that it returns
4654 just the host names, not the priorities.
4656 (3) It is now possible to give a list of domains (or IP addresses) to be
4657 looked up. The behaviour when one of the lookups defers can be
4658 controlled by a keyword.
4660 (4) It is now possible to specify the separator character for use when
4661 multiple records are returned.
4663 38. The dnslists ACL condition has been extended: it is now possible to supply
4664 a list of IP addresses and/or domains to be looked up in a particular DNS
4667 39. Added log_selector=+queue_time_overall.
4669 40. When running the queue in the test harness, wait just a tad after forking a
4670 delivery process, to get repeatability of debugging output.
4672 41. Include certificate and key file names in error message when GnuTLS fails
4673 to set them up, because the GnuTLS error message doesn't include the name
4674 of the failing file when there is a problem reading it.
4676 42. Allow both -bf and -bF in the same test run.
4678 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4680 44. The "Exiscan patch" is now merged into the mainline Exim source.
4682 45. Sometimes the final signoff response after QUIT could fail to get
4683 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4684 < 0 before doing a fflush(). This bug looks as though it goes back to the
4685 introduction of TLS in release 3.20, but "sometimes" must have been rare
4686 because the tests only now provoked it.
4688 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4689 (this can affect the format of dates).
4691 47. exim_tidydb, when checking for the continued existence of a message for
4692 which it has found a message-specific retry record, was not finding
4693 messages that were in split spool directories. Consequently, it was
4694 deleting retry records that should have stayed in existence.
4696 48. Steve fixed some bugs in eximstats.
4698 49. The SPA authentication driver was not abandoning authentication and moving
4699 on to the next authenticator when an expansion was forced to fail,
4700 contradicting the general specification for all authenticators. Instead it
4701 was generating a temporary error. It now behaves as specified.
4703 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4704 (the order specifies the preference for clients). The order is now AES256,
4705 AES128, 3DES, ARCFOUR128.
4707 51. Small patch to Sieve code - explicitly set From: when generating an
4710 52. Exim crashed if a remote delivery caused a very long error message to be
4711 recorded - for instance if somebody sent an entire SpamAssassin report back
4712 as a large number of 550 error lines. This bug was coincidentally fixed by
4713 increasing the size of one of Exim's internal buffers (big_buffer) that
4714 happened as part of the Exiscan merge. However, to be on the safe side, I
4715 have made the code more robust (and fixed the comments that describe what
4718 53. Now that there can be additional text after "Completed" in log lines (if
4719 the queue_time_overall log selector is set), a one-byte patch to exigrep
4720 was needed to allow it to recognize "Completed" as not the last thing in
4723 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4724 patch that reportedly fixes this has been added. I am not expert enough to
4725 create a test for it. This is what the patch creator wrote:
4727 "I found a little strange behaviour of ldap code when working with
4728 Windows 2003 AD Domain, where users was placed in more than one
4729 Organization Units. When I tried to give exim partial DN, the exit code
4730 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4731 But simultaneously result of request was absolutely normal ldap result,
4732 so I produce this patch..."
4734 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4735 REFERENCE, so I have modified the code to exclude the patch when that macro
4738 55. Some experimental protocols are using DNS PTR records for new purposes. The
4739 keys for these records are domain names, not reversed IP addresses. The
4740 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4741 leaves it alone. Component reversal etc. now happens only for IP addresses.
4744 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4746 57. Double the size of the debug message buffer (to 2048) so that more of very
4747 long debug lines gets shown.
4749 58. The exicyclog utility now does better if the number of log files to keep
4750 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4752 59. Two changes related to the smtp_active_hostname option:
4754 (1) $smtp_active_hostname is now available as a variable.
4755 (2) The default for smtp_banner uses $smtp_active_hostname instead
4756 of $primary_hostname.
4758 60. The host_aton() function is supposed to be passed a string that is known
4759 to be a valid IP address. However, in the case of IPv6 addresses, it was
4760 not checking this. This is a hostage to fortune. Exim now panics and dies
4761 if the condition is not met. A case was found where this could be provoked
4762 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4763 components; fortuitously, this particular loophole had already been fixed
4764 by change 4.50/55 above.
4766 If there are any other similar loopholes, the new check in host_aton()
4767 itself should stop them being exploited. The report I received stated that
4768 data on the command line could provoke the exploit when Exim was running as
4769 exim, but did not say which command line option was involved. All I could
4770 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4771 running as the user.
4774 61. There was a buffer overflow vulnerability in the SPA authentication code
4775 (which came originally from the Samba project). I have added a test to the
4776 spa_base64_to_bits() function which I hope fixes it.
4779 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4780 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4782 63. The daemon start-up calls getloadavg() while still root for those OS that
4783 need the first call to be done as root, but it missed one case: when
4784 deliver_queue_load_max is set with deliver_drop_privilege. This is
4785 necessary for the benefit of the queue runner, because there is no re-exec
4786 when deliver_drop_privilege is set.
4788 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4789 This has been fixed.
4791 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4792 was in use, was not putting the data itself into the right store pool;
4793 consequently, it could be overwritten for a subsequent message in the same
4794 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4797 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4799 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4800 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4801 the function were treating the return as a boolean value, which happened to
4802 work because 0=false and not-0=true, but is not correct code.
4804 68. The host_aton() function was not handling scoped IPv6 addresses (those
4805 with, for example, "%eth0" on the end) correctly.
4807 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4808 negated items (that is, ~something) in unsigned ints. Some compilers
4809 apparently mutter when there is no cast.
4811 70. If an address verification called from an ACL failed, and did not produce a
4812 user-specific message (i.e. there was only a "system" message), nothing was
4813 put in $acl_verify_message. In this situation, it now puts the system
4816 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4817 session; change 4.31/43 added the unwanted input to the log line - except
4818 that it did not do this in the start of session case. It now does.
4820 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4821 This is wrong and can cause the other end to generate a synchronization
4822 error if it is another Exim or anything else that does the synchronization
4823 check. A QUIT command is no longer sent after a timeout.
4825 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4826 during host lookups.
4828 74. The defer_ok option of callout verification was not working if it was used
4829 when verifying addresses in header lines, that is, for this case:
4831 verify = header_sender/callout=defer_ok
4833 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4834 those file descriptors could be used for SMTP connections. If anything
4835 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4836 could be sent to the SMTP client, causing chaos. The daemon now opens
4837 stdin, stdout, and stderr to /dev/null when it puts itself into the
4840 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4841 log by default. The user can override this with suitable Perl magic.
4843 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4844 the log message when discard triggers, was not working for the DATA ACL or
4845 for the non-SMTP ACL.
4847 78. Error message wording change in sieve.c.
4849 79. If smtp_accept_max_per_host was set, the number of connections could be
4850 restricted to fewer than expected, because the daemon was trying to set up
4851 a new connection before checking whether the processes handling previous
4852 connections had finished. The check for completed processes is now done
4853 earlier. On busy systems, this bug wouldn't be noticed because something
4854 else would have woken the daemon, and it would have reaped the completed
4857 80. If a message was submitted locally by a user whose login name contained one
4858 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4859 It caused a spool format error. I have fixed the spool reading code. A
4860 related problem was that the "from" clause in the Received: line became
4861 illegal because of the space(s). It is now covered by ${quote_local_part.
4863 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4866 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4868 83. Patch from Sieve maintainer to fix unterminated string problem in
4869 "vacation" handling.
4871 84. Some minor changes to the Linux configuration files to help with other
4872 OS variants using glibc.
4874 85. One more patch for Sieve to update vacation handling to latest spec.
4877 ----------------------------------------------------
4878 See the note above about the 4.44 and 4.50 releases.
4879 ----------------------------------------------------
4885 1. Change 4.43/35 introduced a bug that caused file counts to be
4886 incorrectly computed when quota_filecount was set in an appendfile
4889 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4890 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4893 3. Give more explanation in the error message when the command for a transport
4894 filter fails to execute.
4896 4. There are several places where Exim runs a non-Exim command in a
4897 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4898 was being done only for the command run by the queryprogram router. It is
4899 now done for all such subprocesses. The other cases are: ${run, transport
4900 filters, and the commands run by the lmtp and pipe transports.
4902 5. Some older OS have a limit of 256 on the maximum number of file
4903 descriptors. Exim was using setrlimit() to set 1000 as a large value
4904 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4905 systems. I've change it so that if it can't get 1000, it tries for 256.
4907 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4908 was an oversight, and furthermore, ever since the addition of extra
4909 controls (e.g. 4.43/32), the checks on when to allow different forms of
4910 "control" were broken. There should now be diagnostics for all cases when a
4911 control that does not make sense is encountered.
4913 7. $recipients is now available in the predata ACL (oversight).
4915 8. Tidy the search cache before the fork to do a delivery from a message
4916 received from the command line. Otherwise the child will trigger a lookup
4917 failure and thereby defer the delivery if it tries to use (for example) a
4918 cached ldap connection that the parent has called unbind on.
4920 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4921 of $address_data from the recipient verification was clobbered by the
4922 sender verification.
4924 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4925 was its contents. (It was OK if the option was not defined at all.)
4927 11. A "Completed" log line is now written for messages that are removed from
4928 the spool by the -Mrm option.
4930 12. $host_address is now set to the target address during the checking of
4931 ignore_target_hosts.
4933 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4934 being passed; this would have caused $sender_host_name to have been used if
4935 matching the list had actually called for a host name (not very likely,
4936 since this list is usually IP addresses). A host name is now passed as
4939 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4940 code that specifies a non-restarting handler (typically sigaction() in
4941 modern systems) in an attempt to fix a rare and obscure crash bug.
4943 15. Narrowed the window for a race in the daemon that could cause it to ignore
4944 SIGCHLD signals. This is not a major problem, because they are used only to
4945 wake it up if nothing else does.
4947 16. A malformed maildirsize file could cause Exim to calculate negative values
4948 for the mailbox size or file count. Odd effects could occur as a result.
4949 The maildirsize information is now recalculated if the size or filecount
4952 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4953 support for a long time. Removed HAVE_SYS_VFS_H.
4955 18. Updated exipick to current release from John Jetmore.
4957 19. Allow an empty sender to be matched against a lookup in an address list.
4958 Previously the only cases considered were a regular expression, or an
4961 20. Exim went into a mad DNS lookup loop when doing a callout where the
4962 host was specified on the transport, if the DNS lookup yielded more than
4965 21. The RFC2047 encoding function was originally intended for short strings
4966 such as real names; it was not keeping to the 75-character limit for
4967 encoded words that the RFC imposes. It now respects the limit, and
4968 generates multiple encoded words if necessary. To be on the safe side, I
4969 have increased the buffer size for the ${rfc2047: expansion operator from
4972 22. Failure to deliver a bounce message always caused it to be frozen, even if
4973 there was an errors_to setting on the router. The errors_to setting is now
4976 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4977 canonical form (fully expanded) before being placed in
4978 $sender_host_address.
4980 24. Updated eximstats to version 1.33
4982 25. Include certificate and key file names in error message when GnuTLS fails
4983 to set them up, because the GnuTLS error message doesn't include the name
4984 of the failing file when there is a problem reading it.
4986 26. Expand error message when OpenSSL has problems setting up cert/key files.
4989 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4990 (this can affect the format of dates).
4992 28. exim_tidydb, when checking for the continued existence of a message for
4993 which it has found a message-specific retry record, was not finding
4994 messages that were in split spool directories. Consequently, it was
4995 deleting retry records that should have stayed in existence.
4997 29. eximstats updated to version 1.35
4998 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4999 1.35 - bugfix such that pie charts by volume are generated correctly
5001 30. The SPA authentication driver was not abandoning authentication and moving
5002 on to the next authenticator when an expansion was forced to fail,
5003 contradicting the general specification for all authenticators. Instead it
5004 was generating a temporary error. It now behaves as specified.
5006 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5007 (the order specifies the preference for clients). The order is now AES256,
5008 AES128, 3DES, ARCFOUR128.
5010 31. Small patch to Sieve code - explicitly set From: when generating an
5013 32. Exim crashed if a remote delivery caused a very long error message to be
5014 recorded - for instance if somebody sent an entire SpamAssassin report back
5015 as a large number of 550 error lines. This bug was coincidentally fixed by
5016 increasing the size of one of Exim's internal buffers (big_buffer) that
5017 happened as part of the Exiscan merge. However, to be on the safe side, I
5018 have made the code more robust (and fixed the comments that describe what
5021 33. Some experimental protocols are using DNS PTR records for new purposes. The
5022 keys for these records are domain names, not reversed IP addresses. The
5023 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5024 leaves it alone. Component reversal etc. now happens only for IP addresses.
5027 34. The host_aton() function is supposed to be passed a string that is known
5028 to be a valid IP address. However, in the case of IPv6 addresses, it was
5029 not checking this. This is a hostage to fortune. Exim now panics and dies
5030 if the condition is not met. A case was found where this could be provoked
5031 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5032 components; fortuitously, this particular loophole had already been fixed
5033 by change 4.50/55 or 4.44/33 above.
5035 If there are any other similar loopholes, the new check in host_aton()
5036 itself should stop them being exploited. The report I received stated that
5037 data on the command line could provoke the exploit when Exim was running as
5038 exim, but did not say which command line option was involved. All I could
5039 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5040 running as the user.
5043 35. There was a buffer overflow vulnerability in the SPA authentication code
5044 (which came originally from the Samba project). I have added a test to the
5045 spa_base64_to_bits() function which I hope fixes it.
5048 36. The daemon start-up calls getloadavg() while still root for those OS that
5049 need the first call to be done as root, but it missed one case: when
5050 deliver_queue_load_max is set with deliver_drop_privilege. This is
5051 necessary for the benefit of the queue runner, because there is no re-exec
5052 when deliver_drop_privilege is set.
5054 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5055 was in use, was not putting the data itself into the right store pool;
5056 consequently, it could be overwritten for a subsequent message in the same
5057 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5060 38. Sometimes the final signoff response after QUIT could fail to get
5061 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5062 < 0 before doing a fflush(). This bug looks as though it goes back to the
5063 introduction of TLS in release 3.20, but "sometimes" must have been rare
5064 because the tests only now provoked it.
5070 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5071 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5072 Now it returns an integer. A number of calls were still expecting a T/F
5073 return. Fortuitously, in all cases, the tests worked in OK situations,
5074 which is the norm. However, things would have gone wrong on any write
5075 failures on the smtp file descriptor. This function is used when sending
5076 messages over SMTP and also when doing verify callouts.
5078 2. When Exim is called to do synchronous delivery of a locally submitted
5079 message (the -odf or -odi options), it no longer closes stderr before doing
5082 3. Implemented the mua_wrapper option.
5084 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5086 5. Implemented the functions header_remove(), header_testname(),
5087 header_add_at_position(), and receive_remove_recipient(), and exported them
5090 6. If an ACL "warn" statement specified the addition of headers, Exim already
5091 inserted X-ACL-Warn: at the start if there was no header name. However, it
5092 was not making this test for the second and subsequent header lines if
5093 there were newlines in the string. This meant that an invalid header could
5094 be inserted if Exim was badly configured.
5096 7. Allow an ACL "warn" statement to add header lines at the start or after all
5097 the Received: headers, as well as at the end.
5099 8. Added the rcpt_4xx retry error code.
5101 9. Added postmaster_mailfrom=xxx to callout verification option.
5103 10. Added mailfrom=xxxx to the callout verification option, for verify=
5106 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5107 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5109 12. Inserted some casts to stop certain compilers warning when using pointer
5110 differences as field lengths or precisions in printf-type calls (mostly
5111 affecting debugging statements).
5113 13. Added optional readline() support for -be (dynamically loaded).
5115 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5116 same clock tick as a message's arrival, so that its received time was the
5117 same as the "first fail" time on the retry record, and that message
5118 remained on the queue past the ultimate address timeout, every queue runner
5119 would try a delivery (because it was past the ultimate address timeout) but
5120 after another failure, the ultimate address timeout, which should have then
5121 bounced the address, did not kick in. This was a "< instead of <=" error;
5122 in most cases the first failure would have been in the next clock tick
5123 after the received time, and all would be well.
5125 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5126 being recognized when the domain list was tested by the match_domain
5127 condition in an expansion string.
5129 16. Added the ${str2b64: operator.
5131 17. Exim was always calling setrlimit() to set a large limit for the number of
5132 processes, without checking whether the existing limit was already
5133 adequate. (It did check for the limit on file descriptors.) Furthermore,
5134 errors from getrlimit() and setrlimit() were being ignored. Now they are
5135 logged to the main and panic logs, but Exim does carry on, to try to do its
5136 job under whatever limits there are.
5138 18. Imported PCRE 5.0.
5140 19. Trivial typo in log message " temporarily refused connection" (the leading
5143 20. If the log selector return_path_on_delivery was set and an address was
5144 redirected to /dev/null, the delivery process crashed because it assumed
5145 that a return path would always be set for a "successful" delivery. In this
5146 case, the whole delivery is bypassed as an optimization, and therefore no
5149 21. Internal re-arrangement: the function for sending a challenge and reading
5150 a response while authentication was assuming a zero-terminated challenge
5151 string. It's now changed to take a pointer and a length, to allow for
5152 binary data in such strings.
5154 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5156 23. Exim was not respecting finduser_retries when seeking the login of the
5157 uid under which it was called; it was always trying 10 times. (The default
5158 setting of finduser_retries is zero.) Also, it was sleeping after the final
5159 failure, which is pointless.
5161 24. Implemented tls_on_connect_ports.
5163 25. Implemented acl_smtp_predata.
5165 26. If the domain in control=submission is set empty, Exim assumes that the
5166 authenticated id is a complete email address when it generates From: or
5167 Sender: header lines.
5169 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5170 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5171 chown and chgrp in /bin and hostname in /usr/bin.
5173 28. Exim was keeping the "process log" file open after each use, just as it
5174 does for the main log. This opens the possibility of it remaining open for
5175 long periods when the USR1 signal hits a daemon. Occasional processlog
5176 errors were reported, that could have been caused by this. Anyway, it seems
5177 much more sensible not to leave this file open at all, so that is what now
5180 29. The long-running daemon process does not normally write to the log once it
5181 has entered its main loop, and it closes the log before doing so. This is
5182 so that log files can straightforwardly be renamed and moved. However,
5183 there are a couple of unusual error situations where the daemon does write
5184 log entries, and I had neglected to close the log afterwards.
5186 30. The text of an SMTP error response that was received during a remote
5187 delivery was being truncated at 512 bytes. This is too short for some of
5188 the long messages that one sometimes sees. I've increased the limit to
5191 31. It is now possible to make retry rules that apply only when a message has a
5192 specific sender, in particular, an empty sender.
5194 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5195 it possible to be selective about when SMTP synchronization is enforced.
5197 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5199 32. Implemented hosts_connection_nolog.
5201 33. Added an ACL for QUIT.
5203 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5206 35. Added mailbox_size and mailbox_filecount to appendfile.
5208 36. Added control = no_multiline_responses to ACLs.
5210 37. There was a bug in the logic of the code that waits for the clock to tick
5211 in the case where the clock went backwards by a substantial amount such
5212 that the microsecond fraction of "now" was more than the microsecond
5213 fraction of "then" (but the whole seconds number was less).
5215 38. Added support for the libradius Radius client library this is found on
5216 FreeBSD (previously only the radiusclient library was supported).
5222 1. When certain lookups returned multiple values in the form name=value, the
5223 quoting of the values was not always being done properly. Specifically:
5224 (a) If the value started with a double quote, but contained no whitespace,
5226 (b) If the value contained whitespace other than a space character (i.e.
5227 tabs or newlines or carriage returns) it was not quoted.
5228 This fix has been applied to the mysql and pgsql lookups by writing a
5229 separate quoting function and calling it from the lookup code. The fix
5230 should probably also be applied to nisplus, ibase and oracle lookups, but
5231 since I cannot test any of those, I have not disturbed their existing code.
5233 2. A hit in the callout cache for a specific address caused a log line with no
5234 reason for rejecting RCPT. Now it says "Previous (cached) callout
5235 verification failure".
5237 3. There was an off-by-one bug in the queryprogram router. An over-long
5238 return line was truncated at 256 instead of 255 characters, thereby
5239 overflowing its buffer with the terminating zero. As well as fixing this, I
5240 have increased the buffer size to 1024 (and made a note to document this).
5242 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5243 when Exim is waiting for an SMTP response from a remote server, Exim
5244 restarts its select() call on the socket, thereby resetting its timeout.
5245 This is not a problem when such interrupts are rare. Somebody set up a cron
5246 job to run exiwhat every 2 minutes, which is less than the normal select()
5247 timeout (5 or 10 minutes). This meant that the select() timeout never
5248 kicked in because it was always reset. I have fixed this by comparing the
5249 time when an interrupt arrives with the time at the start of the first call
5250 to select(). If more time than the timeout has elapsed, the interrupt is
5251 treated as a timeout.
5253 5. Some internal re-factoring in preparation for the addition of Sieve
5254 extensions (by MH). In particular, the "personal" test is moved to a
5255 separate function, and given an option for scanning Cc: and Bcc: (which is
5256 not set for Exim filters).
5258 6. When Exim created an email address using the login of the caller as the
5259 local part (e.g. when creating a From: or Sender: header line), it was not
5260 quoting the local part when it contained special characters such as @.
5262 7. Installed new OpenBSD configuration files.
5264 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5265 try to make them clearer.
5267 9. Callout options, other than the timeout value, were being ignored when
5268 verifying sender addresses in header lines. For example, when using
5270 verify = header_sender/callout=no_cache
5272 the cache was (incorrectly) being used.
5274 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5275 only the Cygwin environment.
5277 11. When return_path_on_delivery was set as a log selector, if different remote
5278 addresses in the same message used different return paths and parallel
5279 remote delivery occurred, the wrong values would sometimes be logged.
5280 (Whenever a remote delivery process finished, the return path value from
5281 the most recently started remote delivery process was logged.)
5283 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5284 lines when AUTH and/or TLS are in use. This is the "received protocol"
5285 field. Exim used to use "asmtp" for authenticated SMTP, without any
5286 indication (in the protocol name) for TLS use. Now it follows the RFC and
5287 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5288 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5289 names appear in log lines as well as in Received: header lines.
5291 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5292 extensions, and comparison tests, and to fix some bugs.
5294 14. Changes to the "personal" filter test:
5296 (1) The test was buggy in that it was just doing the equivalent of
5297 "contains" tests on header lines. For example, if a user's address was
5298 anne@some.where, the "personal" test would incorrectly be true for
5300 To: susanne@some.where
5302 This test is now done by extracting each address from the header in turn,
5303 and checking the entire address. Other tests that are part of "personal"
5304 are now done using regular expressions (for example, to check local parts
5305 of addresses in From: header lines).
5307 (2) The list of non-personal local parts in From: addresses has been
5308 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5309 taken from the Sieve specification recommendations.
5311 (3) If the message contains any header line starting with "List-" it is
5312 treated as non-personal.
5314 (4) The test for "circular" in the Subject: header line has been removed
5315 because it now seems ill-conceived.
5317 15. Minor typos in src/EDITME comments corrected.
5319 16. Installed latest exipick from John Jetmore.
5321 17. If headers_add on a router specified a text string that was too long for
5322 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5323 of string_sprintf() is now avoided.
5325 18. $message_body_size was not set (it was always zero) when running the DATA
5326 ACL and the local_scan() function.
5328 19. For the "mail" command in an Exim filter, no default was being set for
5329 the once_repeat time, causing a random time value to be used if "once" was
5330 specified. (If the value happened to be <= 0, no repeat happened.) The
5331 default is now 0s, meaning "never repeat". The "vacation" command was OK
5332 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5333 (I found it when inspecting the code).
5335 20. There is now an overall timeout for performing a callout verification. It
5336 defaults to 4 times the callout timeout, which applies to individual SMTP
5337 commands during the callout. The overall timeout applies when there is more
5338 than one host that can be tried. The timeout is checked before trying the
5339 next host. This prevents very long delays if there are a large number of
5340 hosts and all are timing out (e.g. when the network connections are timing
5341 out). The value of the overall timeout can be changed by specifying an
5342 additional sub-option for "callout", called "maxwait". For example:
5344 verify = sender/callout=5s,maxwait=20s
5346 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5347 to the end before writing, but this should make it even safer).
5349 22. Exim was forgetting that it had advertised PIPELINING for the second and
5350 subsequent messages on an SMTP connection. It was also not resetting its
5351 memory on STARTTLS and an internal HELO.
5353 23. When Exim logs an SMTP synchronization error within a session, it now
5354 records whether PIPELINING has been advertised or not.
5356 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5357 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5358 rather than long int.
5360 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5362 26. Added the never_mail option to autoreply.
5368 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5369 crash if the getsockname() call failed; this can happen if a connection is
5370 closed very soon after it is established. The problem was simply in the
5371 order in which certain operations were done, causing Exim to try to write
5372 to the SMTP stream before it had set up the file descriptor. The bug has
5373 been fixed by making things happen in the correct order.
5379 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5380 before the connection was closed, thus losing the rejection response.
5382 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5383 some early Solaris releases, but causes trouble in current releases where
5384 socklen_t is defined.
5386 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5389 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5392 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5393 that releasing the top of it at the end releases what was used for sub-
5394 expansions (unless the block got too big). However, discard this block if
5395 the first thing is a variable or header, so that we can use its block when
5396 it is dynamic (useful for very large $message_headers, for example).
5398 6. Lookups now cache *every* query, not just the most recent. A new, separate
5399 store pool is used for this. It can be recovered when all lookup caches are
5400 flushed. Lookups now release memory at the end of their result strings.
5401 This has involved some general refactoring of the lookup sources.
5403 7. Some code has been added to the store_xxx() functions to reduce the amount
5404 of flapping under certain conditions.
5406 8. log_incoming_interface used to affect only the <= reception log lines. Now
5407 it causes the local interface and port to be added to several more SMTP log
5408 lines, for example "SMTP connection from", and rejection lines.
5410 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5412 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5414 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5415 could be overwritten at the end of the current message (or the start of a
5416 new message if it was set in a HELO ACL). The value is now preserved for
5417 the duration of the SMTP connection.
5419 12. If a transport had a headers_rewrite setting, and a matching header line
5420 contained an unqualified address, that address was qualified, even if it
5421 did not match any rewriting rules. The underlying bug was that the values
5422 of the flags that permit the existence of unqualified sender and recipient
5423 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5424 non-local messages, and by -bnq for local messages) were not being
5425 preserved with the message after it was received.
5427 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5428 "next input=" as part of the text comprising the host identity instead of
5429 the correct text. The code was using the same buffer for two different
5430 strings. However, depending on which order the printing function evaluated
5431 its arguments, the bug did not always show up. Under Linux, for example, my
5432 test suite worked just fine.
5434 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5435 doesn't work with some older versions of Perl. It has been changed to "my",
5436 which in any case is probably the better facility to use.
5438 15. A really picky compiler found some instances of statements for creating
5439 error messages that either had too many or two few arguments for the format
5442 16. The size of the buffer for calls to the DNS resolver has been increased
5443 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5444 for addresses that have a lot of PTR records. This alleviates a problem; it
5445 does not fully solve it.
5447 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5448 buffer now truncates the list and logs the incident, which is the same
5449 action as happens when Exim is looking up a host name and its aliases.
5450 Previously in this situation something unpredictable would happen;
5451 sometimes it was "internal error: store_reset failed".
5453 18. If a server dropped the connection unexpectedly when an Exim client was
5454 using GnuTLS and trying to read a response, the client delivery process
5455 crashed while trying to generate an error log message.
5457 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5458 string, for example:
5460 warn message = H1: something\nH2: something
5462 the text was added as a single header line from Exim's point of view
5463 though it ended up OK in the delivered message. However, searching for the
5464 second and subsequent header lines using $h_h2: did not work. This has been
5465 fixed. Similarly, if a system filter added multiple headers in this way,
5466 the routers could not see them.
5468 20. Expanded the error message when iplsearch is called with an invalid key to
5469 suggest using net-iplsearch in a host list.
5471 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5472 ACLs are no longer actually imposed (and a message to that effect is
5475 22. If a "gecos" field in a passwd entry contained escaped characters, in
5476 particular, if it contained a \" sequence, Exim got it wrong when building
5477 a From: or a Sender: header from that name. A second bug also caused
5478 incorrect handling when an unquoted " was present following a character
5479 that needed quoting.
5481 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5482 was not being matched caselessly.
5484 24. Arranged for all hyphens in the exim.8 source to be escaped with
5487 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5488 properly. Recipient callout cache records were still being keyed to include
5489 the sender, even when use_sender was set false. This led to far more
5490 callouts that were necessary. The sender is no longer included in the key
5491 when use_sender is false.
5493 26. Added "control = submission" modifier to ACLs.
5495 27. Added the ${base62d: operator to decode base 62 numbers.
5497 28. dnsdb lookups can now access SRV records.
5499 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5500 the configuration file.
5502 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5503 (-v) mode. This makes the output for a verbose queue run more intelligible.
5505 31. Added a use_postmaster feature to recipient callouts.
5507 32. Added the $body_zerocount variable, containing the number of binary zero
5508 bytes in the message body.
5510 33. The time of last modification of the "new" subdirectory is now used as the
5511 "mailbox time last read" when there is a quota error for a maildir
5514 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5516 35. Added +ignore_unknown as a special item in host lists.
5518 36. Code for decoding IPv6 addresses in host lists is now included, even if
5519 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5520 address was recognized as an IP address, but was then not correctly decoded
5521 into binary, causing unexpected and incorrect effects when compared with
5528 1. Very minor rewording of debugging text in manualroute to say "list of
5529 hosts" instead of "hostlist".
5531 2. If verify=header_syntax was set, and a header line with an unqualified
5532 address (no domain) and a large number of spaces between the end of the
5533 name and the colon was received, the reception process suffered a buffer
5534 overflow, and (when I tested it) crashed. This was caused by some obsolete
5535 code that should have been removed. The fix is to remove it!
5537 3. When running in the test harness, delay a bit after writing a bounce
5538 message to get a bit more predictability in the log output.
5540 4. Added a call to search_tidyup() just before forking a reception process. In
5541 theory, someone could use a lookup in the expansion of smtp_accept_max_
5542 per_host which, without the tidyup, could leave open a database connection.
5544 5. Added the variables $recipient_data and $sender_data which get set from a
5545 lookup success in an ACL "recipients" or "senders" condition, or a router
5546 "senders" option, similar to $domain_data and $local_part_data.
5548 6. Moved the writing of debug_print from before to after the "senders" test
5551 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5552 problems for message scanning, either using a data ACL, or using
5553 local_scan() because the Received: header was not generated till after they
5554 were called (in order to set the time as the time of reception completion).
5555 I have revised the way this works. The header is now generated after the
5556 body is received, but before the ACL or local_scan() are called. After they
5557 are run, the timestamp in the header is updated.
5563 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5564 before starting a queue runner without re-exec. This happened only when
5565 deliver_drop_privilege was set or when the Exim user was set to root. The
5566 effect of the bug was that timeouts during subsequent deliveries caused
5567 crashes instead of being properly handled. The handler is now left at its
5568 default (and expected) setting.
5570 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5571 message, again when deliver_drop_privilege is set or Exim is run as root.
5572 The bug described in (1) was not present in this case, but the tidying up
5573 of the other signals was missing. I have made the two cases consistent.
5575 3. The ignore_target_hosts setting on a manualroute router was being ignored
5576 for hosts that were looked up using the /MX notation.
5578 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5581 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5582 operated on the sender address. After changing the $sender_address to <>
5583 for the sender address verify, Exim was re-instated it as the original
5584 (before rewriting) address, but remembering that it had rewritten it, so it
5585 wasn't rewriting it again. This bug also had the effect of breaking the
5586 sender address verification caching when the sender address was rewritten.
5588 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5589 This has been changed so that if the ip literal address matches
5590 ignore_target_hosts, the router declines.
5592 7. Added expansion conditions match_domain, match_address, and match_local_
5593 part (NOT match_host).
5595 8. The placeholder for the Received: header didn't have a length field set.
5597 9. Added code to Exim itself and to exim_lock to test for a specific race
5598 condition that could lead to file corruption when using MBX delivery. The
5599 issue is with the lockfile that is created in /tmp. If this file is removed
5600 after a process has opened it but before that process has acquired a lock,
5601 there is the potential for a second process to recreate the file and also
5602 acquire a lock. This could lead to two Exim processes writing to the file
5603 at the same time. The added code performs the same test as UW imapd; it
5604 checks after acquiring the lock that its file descriptor still refers to
5605 the same named file.
5607 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5608 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5611 11. Added the smtp_active_hostname option. If used, this will typically be made
5612 to depend on the incoming interface address. Because $interface_address is
5613 not set up until the daemon has forked a reception process, error responses
5614 that can happen earlier (such as "too many connections") no longer contain
5617 12. If an expansion in a condition on a "warn" statement fails because a lookup
5618 defers, the "warn" statement is abandoned, and the next ACL statement is
5619 processed. Previously this caused the whole ACL to be aborted.
5621 13. Added the iplsearch lookup type.
5623 14. Added ident_timeout as a log selector.
5625 15. Added tls_certificate_verified as a log selector.
5627 16. Added a global option tls_require_ciphers (compare the smtp transport
5628 option of the same name). This controls incoming TLS connections.
5630 17. I finally figured out how to make tls_require_ciphers do a similar thing
5631 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5632 before starting the TLS session.
5634 18. Tabs are now shown as \t in -bP output.
5636 19. If the log selector return_path_on_delivery was set, Exim crashed when
5637 bouncing a message because it had too many Received: header lines.
5639 20. If two routers both had headers_remove settings, and the first one included
5640 a superfluous trailing colon, the final name in the first list and the
5641 first name in the second list were incorrectly joined into one item (with a
5642 colon in the middle).
5648 1. Added -C and -D options to the exinext utility, mainly to make it easier
5649 to include in the automated testing, but these could be helpful when
5650 multiple configurations are in use.
5652 2. The exinext utility was not formatting the output nicely when there was
5653 an alternate port involved in the retry record key, nor when there was a
5654 message id as well (for retries that were specific to a specific message
5655 and a specific host). It was also confused by IPv6 addresses, because of
5656 the additional colons they contain. I have fixed the IPv4 problem, and
5657 patched it up to do a reasonable job for IPv6.
5659 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5660 delivery, the log line now contains "pipelined" if PIPELINING was used.
5662 4. An SMTP transport process used to panic and die if the bind() call to set
5663 an explicit outgoing interface failed. This has been changed; it is now
5664 treated in the same way as a connect() failure.
5666 5. A reference to $sender_host_name in the part of a conditional expansion
5667 that was being skipped was still causing a DNS lookup. This no longer
5670 6. The def: expansion condition was not recognizing references to header lines
5671 that used bh_ and bheader_.
5673 7. Added the _cache feature to named lists.
5675 8. The code for checking quota_filecount in the appendfile transport was
5676 allowing one more file than it should have been.
5678 9. For compatibility with Sendmail, the command line option
5686 and sets the incoming protocol and host name (for trusted callers). The
5687 host name and its colon can be omitted when only the protocol is to be set.
5688 Note the Exim already has two private options, -pd and -ps, that refer to
5689 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5690 "s", but I don't think that's a major issue.
5692 10. A number of refactoring changes to the code, none of which should affect
5695 (a) The number of logging options was getting close to filling up the
5696 32-bit word that was used as a bit map. I have split them into two classes:
5697 those that are passed in the argument to log_write(), and those that are
5698 only ever tested independently outside of that function. These are now in
5699 separate 32-bit words, so there is plenty of room for expansion again.
5700 There is no change in the user interface or the logging behaviour.
5702 (b) When building, for example, log lines, the code previously used a
5703 macro that called string_cat() twice, in order to add two strings. This is
5704 not really sufficiently general. Furthermore, there was one instance where
5705 it was actually wrong because one of the argument was used twice, and in
5706 one call a function was used. (As it happened, calling the function twice
5707 did not affect the overall behaviour.) The macro has been replaced by a
5708 function that can join an arbitrary number of extra strings onto a growing
5711 (c) The code for expansion conditions now uses a table and a binary chop
5712 instead of a serial search (which was left over from when there were very
5713 few conditions). Also, it now recognizes conditions like "pam" even when
5714 the relevant support is not compiled in: a suitably worded error message is
5715 given if an attempt is made to use such a condition.
5717 11. Added ${time_interval:xxxxx}.
5719 12. A bug was causing one of the ddress fields not to be passed back correctly
5720 from remote delivery subprocesses. The field in question was not being
5721 subsequently used, so this caused to problems in practice.
5723 13. Added new log selectors queue_time and deliver_time.
5725 14. Might have fixed a bug in maildirsizefile handling that threw up
5726 "unexpected character" debug warnings, and recalculated the data
5727 unnecessarily. In any case, I expanded the warning message to give more
5730 15. Added the message "Restricted characters in address" to the statements in
5731 the default ACL that block characters like @ and % in local parts.
5733 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5734 Three changes have been made:
5736 (a) There was a serious bug; a negative response to MAIL caused the whole
5737 recipient domain to be cached as invalid, thereby blocking all messages
5738 to all local parts at the same domain, from all senders. This bug has
5739 been fixed. The domain is no longer cached after a negative response to
5740 MAIL if the sender used is not empty.
5742 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5745 (c) A new callout option, "use_sender" has been added for people who want
5746 the modified behaviour.
5752 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5755 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5756 indeed breaks things for older releases.
5758 3. Added additional logging to the case where there is a problem reading data
5759 from a filter that is running in a subprocess using a pipe, in order to
5760 try to track down a specific problem.
5762 4. Testing facility fudge: when running in the test harness and attempting
5763 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5764 getting "No route to host". Convert this to a timeout.
5766 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5769 6. Some OS don't have socklen_t but use size_t instead. This affects the
5770 fifth argument of getsockopt() amongst other things. This is now
5771 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5772 can be set for individual OS. I have set it for SunOS5, OSF1, and
5773 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5774 some earlier ones do not.
5776 7. Change 4.30/15 was not doing the test caselessly.
5778 8. The standard form for an IPv6 address literal was being rejected by address
5779 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5780 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5781 this, as well as the form without the "IPv6" on the front (but only when
5782 address literals are enabled, of course).
5784 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5786 10. Exim crashed if a message with an empty sender address specified by -f
5787 encountered a router with an errors_to setting. This could be provoked only
5788 by a command such as
5792 where an empty string was supplied; "<>" did not hit this bug.
5794 11. Installed PCRE release 4.5.
5796 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5797 remained set. It is now erased.
5799 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5800 times from message ids (which are base 36 rather than the normal 62).
5802 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5803 were being counted as actual protocol errors, and logged if the log
5804 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5805 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5806 and DATA following a set of rejected RCPTs do not count as protocol errors.
5807 In other words, Exim assumes they were pipelined, though this may not
5808 actually be the case. Of course, in all cases the client gets an
5809 appropriate error code.
5811 15. If a lookup fails in an ACL condition, a message about the failure may
5812 be available; it is used if testing the ACL cannot continue, because most
5813 such messages specify what the cause of the deferral is. However, some
5814 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5815 that caused an old message to be retained and used if a later statement
5816 caused a defer, replacing the real cause of the deferral.
5818 16. If an IP address had so many PTR records that the DNS lookup buffer
5819 was not large enough to hold them, Exim could crash while trying to process
5820 the truncated data. It now detects and logs this case.
5822 17. Further to 4.21/58, another change has been made: if (and only if) the
5823 first line of a message (the first header line) ends with CRLF, a bare LF
5824 in a subsequent header line has a space inserted after it, so as not to
5825 terminate the header.
5827 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5828 unnecessarily, used atoi() instead of strtol(), and didn't check the
5829 termination when getting file sizes from file names by regex.
5831 19. Completely re-implemented the support for maildirsize files, in the light
5832 of a number of problems with the previous contributed implementation
5833 (4.30/29). In particular:
5835 . If the quota is zero, the maildirsize file is maintained, but no quota is
5838 . If the maildir directory does not exist, it is created before any attempt
5839 to write a maildirsize file.
5841 . The quota value in the file is just a cache; if the quota is changed in
5842 the transport, the new value overrides.
5844 . A regular expression is available for excluding directories from the
5847 20. The autoreply transport checks the characters in options that define the
5848 message's headers; it allows continued headers, but it was checking with
5849 isspace() after an embedded newline instead of explicitly looking for a
5852 21. If all the "regular" hosts to which an address was routed had passed their
5853 expiry times, and had not reached their retry times, the address was
5854 bounced, even if fallback hosts were defined. Now Exim should go on to try
5857 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5858 equivalent code in the SMTP transport. Some hosts send humungous responses
5859 to HELO/EHLO, more than 1024 it seems.
5861 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5862 gives compiler warnings in some environments. I've now done it "properly",
5865 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5866 (because of problems with the built-in one) was declared to return uschar *
5867 instead of char *, causing compiler failure.
5869 25. Fixed a file descriptor leak when processing alias/forward files.
5871 26. Fixed a minor format string issue in dbfn.c.
5873 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5875 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5876 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5879 29. When a sender address is verified, it is cached, to save repeating the test
5880 when there is more than one recipient in a message. However, when the
5881 verification involves a callout, it is possible for different callout
5882 options to be set for different recipients. It is too complicated to keep
5883 track of this in the cache, so now Exim always runs a verification when a
5884 callout is required, relying on the callout cache for the optimization.
5885 The overhead is duplication of the address routing, but this should not be
5888 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5889 to be verified with callout=postmaster, and the main callout worked but the
5890 postmaster check failed, the verification correctly failed. However, if a
5891 subsequent RCPT command asked for sender verification *without* the
5892 postmaster check, incorrect caching caused this verification also to fail,
5895 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5896 it was not caching the DNS options (qualify_single, search_parents) that
5897 were used when the lookup failed. A subsequent lookup with different
5898 options therefore always gave the same answer, though there were cases
5899 where it should not have. (Example: a "domains = !$mx_any" option on a
5900 dnslookup router: the "domains" option is always processed without any
5901 widening, but the router might have qualify_single set.) Now Exim uses the
5902 cached value only when the same options are set.
5904 32. Added John Jetmore's "exipick" utility to the distribution.
5906 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5907 than a timeout (e.g. a certificate is required, and is not provided), an
5908 Exim server now closes the connection immediately. Previously it waited for
5909 the client to close - but if the client is SSL, it seems that they each
5910 wait for each other, leading to a delay before one of them times out.
5912 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5913 maintained 0.8.x compatibility because I don't think many are using it, and
5914 it is clearly obsolete.
5916 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5919 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5920 client certificate was expired. A simple patch fixes this, though I don't
5921 understand the full logic of why the verify callback is called multiple
5924 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5925 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5926 which causes problems with some clients (such as the Certicom SSL Plus
5927 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5928 disables the coutermeasure allowing Eudora to connect."
5930 38. Exim was not checking that a write() to a log file succeeded. This could
5931 lead to Bad Things if a log got too big, in particular if it hit a file
5932 size limit. Exim now panics and dies if it cannot write to a log file, just
5933 as it does if it cannot open a log file.
5935 39. Modified OS/Makefile-Linux so that it now contains
5937 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5939 The two -D definitions ensure that Exim is compiled with large file
5940 support, which makes it possible to handle log files that are bigger than
5943 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5944 instance) a domain was checked against a named list that involved a lookup,
5945 causing $domain_data to be set, then another domain was checked against the
5946 same list, then the first domain was re-checked, the value of $domain_data
5947 after the final check could be wrong. In particular, if the second check
5948 failed, it could be set empty. This bug probably also applied to
5951 41. The strip_trailing_dot option was not being applied to the address given
5952 with the -f command-line option.
5954 42. The code for reading a message's header from the spool was incrementing
5955 $received_count, but never initializing it. This meant that the value was
5956 incorrect (doubled) while delivering a message in the same process in which
5957 it was received. In the most common configuration of Exim, this never
5958 happens - a fresh exec is done - but it can happen when
5959 deliver_drop_privilege is set.
5961 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5962 it now includes up to 150 characters of the unexpected data in the log
5965 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5966 and building data strings. The size of both of these buffers was 10 000
5967 bytes - far larger than anybody would *ever* want, thought I. Needless to
5968 say, somebody hit the limit. I have increased the maximum line length to
5969 20 000 and the maximum data length of concatenated lines to 100 000. I have
5970 also fixed two bugs, because there was no checking on these buffers. Tsk,
5971 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5972 buffer is too small.
5974 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5975 lsearch lookups. Now it does.
5977 46. When parsing a route_list item in a manualroute router, a fixed-length
5978 buffer was used for the list of hosts. I made this 1024 bytes long,
5979 thinking that nobody would ever have a list of hosts that long. Wrong.
5980 Somebody had a whole pile of complicated expansion conditions, and the
5981 string was silently truncated, leading to an expansion error. It turns out
5982 that it is easier to change to an unlimited length (owing to other changes
5983 that have happened since this code was originally written) than to build
5984 structure for giving a limitation error. The length of the item that
5985 expands into the list of hosts is now unlimited.
5987 47. The lsearch lookup could not handle data where the length of text line was
5988 more than 4095 characters. Such lines were truncated, leading to shortened
5989 data being returned. It should now handle lines of any length.
5991 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5992 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5995 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5996 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5997 no longer happens in comment lines. A list of replacements is now placed
5998 at the head of all of the source files, except those whose only change is
5999 to replace PERL_COMMAND in the very first #! line.
6001 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6002 messages on the queue, with a bottom-up merge sort, using code contributed
6003 by Michael Haardt. This should make operations like -bp somewhat faster on
6004 large queues. It won't affect queue runners, except when queue_run_in_order
6007 51. Installed eximstats 1.31 in the distribution.
6009 52. Added support for SRV lookups to the dnslookup router.
6011 53. If an ACL referred to $message_body or $message_body_end, the value was not
6012 reset for any messages that followed in the same SMTP session.
6014 54. The store-handling optimization for building very long strings was not
6015 differentiating between the different store pools. I don't think this
6016 actually made any difference in practice, but I've tidied it.
6018 55. While running the routers to verify a sender address, $sender_address
6019 was still set to the sender address. This is wrong, because when routing to
6020 send a bounce to the sender, it would be empty. Therefore, I have changed
6021 it so that, while verifying a sender address, $sender_address is set to <>.
6022 (There is no change to what happens when verifying a recipient address.)
6024 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6025 target A or AAAA records (if not already returned) without resetting the
6026 qualify_single or search_parents options of the DNS resolver. These are
6027 inappropriate in this case because the targets of MX and SRV records must
6028 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6029 target that, when qualified, matched something in the local domain. These
6030 two options are now turned off when doing these lookups.
6032 57. It seems that at least some releases of Reiserfs (which does not have the
6033 concept of a fixed number of inodes) returns zero and not -1 for the
6034 number of available inodes. This interacted badly with check_spool_inodes,
6035 which assumed that -1 was the "no such thing" setting. What I have done is
6036 to check that the total number of inodes is greater than zero before doing
6037 the test of how many are available.
6039 58. When a "warn" ACL statement has a log_message modifier, the message is
6040 remembered, and not repeated. This is to avoid a lot of repetition when a
6041 message has many recipients that cause the same warning to be written.
6042 However, Exim was preserving the list of already written lines for an
6043 entire SMTP session, which doesn't seem right. The memory is now reset if a
6044 new message is started.
6046 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6047 debugging output unless log_rewrite was also set.
6049 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6050 of (int)(handle) when we know that handle contains (void *)(-1).
6052 61. The Exim daemon panic-logs an error return when it closes the incoming
6053 connection. However "connection reset by peer" seems to be common, and
6054 isn't really an error worthy of noting specially, so that particular error
6057 62. When Exim is trying to find all the local interfaces, it used to panic and
6058 die if the ioctl to get the interface flags failed. However, it seems that
6059 on at least one OS (Solaris 9) it is possible to have an interface that is
6060 included in the list of interfaces, but for which you get a failure error
6061 for this call. This happens when the interface is not "plumbed" into a
6062 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6063 failure of the "get flags" call assumes that the interface is down.
6065 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6066 makes life easier for people who are doing arithmetic on fields extracted
6067 from dates, where you often get leading zeros that should not be
6068 interpreted as octal.
6070 64. Added qualify_domain to the redirect router, to override the global
6073 65. If a pathologically long header line contained very many addresses (the
6074 report of this problem mentioned 10 000) and each of them was rewritten,
6075 Exim could use up a very large amount of memory. (It kept on making new
6076 copies of the header line as it rewrote, and never released the old ones.)
6077 At the expense of a bit more processing, the header rewriting function has
6078 been changed so that it no longer eats memory in this way.
6080 66. The generation of the Received: header has been moved from the time that a
6081 message starts to be received, to the time that it finishes. The timestamp
6082 in the Received: header should now be very close to that of the <= log
6083 line. There are two side-effects of this change:
6085 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6086 logged header lines no longer include the local Received: line, because
6087 it has not yet been created. The same applies to a copy of the message
6088 that is returned to a non-SMTP sender when a message is rejected.
6090 (b) When a filter file is tested using -bf, no additional Received: header
6091 is added to the test message. After some thought, I decided that this
6094 This change does not affect the value of $received_for. It is still set
6095 after address rewriting, but before local_scan() is called.
6097 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6099 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6100 gave an unhelpful panic error message, and a defer error. I have managed to
6101 change this behaviour so that it now rejects any supplied certificate,
6102 which seems right, as the list of acceptable certificates is empty.
6104 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6105 gave an unhelpful defer error. I have not managed to make this reject any
6106 supplied certificates, but the error message it gives is "no certificate
6107 supplied", which is not helpful.
6109 70. exigrep's output now also includes lines that are not associated with any
6110 message, but which match the given pattern. Implemented by a patch from
6111 Martin Sluka, which also tidied up the Perl a bit.
6113 71. Recipient callout verification, like sender verification, was using <> in
6114 the MAIL FROM command. This isn't really the right thing, since the actual
6115 sender may affect whether the remote host accepts the recipient or not. I
6116 have changed it to use the actual sender in the callout; this means that
6117 the cache record is now keyed on a recipient/sender pair, not just the
6118 recipient address. There doesn't seem to be a real danger of callout loops,
6119 since a callout by the remote host to check the sender would use <>.
6120 [SEE ABOVE: changed after hitting problems.]
6122 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6123 temporary errors. However, in the case of such a code being given after
6124 the end of a data transmission (i.e. after ".") Exim was failing to write
6125 a retry record for the message. (Yes, there was some broken host that was
6126 actually sending 8xx at this point.)
6128 73. An unknown lookup type in a host list could cause Exim to panic-die when
6129 the list was checked. (An example that provoked this was putting <; in the
6130 middle of a list instead of at the start.) If this happened during a DATA
6131 ACL check, a -D file could be left lying around. This kind of configuration
6132 error no longer causes Exim to die; instead it causes a defer error. The
6133 incident is still logged to the main and panic logs.
6135 74. Buglet left over from Exim 3 conversion. The message "too many messages
6136 in one connection" was written to the rejectlog but not the mainlog, except
6137 when address rewriting (yes!) was being logged.
6139 75. Added write_rejectlog option.
6141 76. When a system filter was run not as root (that is, when system_filter_user
6142 was set), the values of the $n variables were not being returned to the
6143 main process; thus, they were not subsequently available in the $sn
6146 77. Added +return_path_on_delivery log selector.
6148 78. A connection timeout was being treated differently from recipients deferred
6149 when testing hosts_max_try with a message that was older than the host's
6150 retry timeout. (The host should not be counted, thus allowing all hosts to
6151 be tried at least once before bouncing.) This may have been the cause of an
6152 occasionally reported bug whereby a message would remain on the queue
6153 longer than the retry timeout, but would be bounced if a delivery was
6154 forced. I say "may" because I never totally pinned down the problem;
6155 setting up timeout/retry tests is difficult. See also the next item.
6157 79. The ultimate address timeout was not being applied to errors that involved
6158 a combination of host plus message (for example, a timeout on a MAIL
6159 command). When an address resolved to a number of possible hosts, and they
6160 were not all tried for each delivery (e.g. because of hosts_max_try), a
6161 message could remain on the queue longer than the retry timeout.
6163 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6166 81. Fixed an obscure SMTP outgoing bug which required at least the following
6167 conditions: (a) there was another message waiting for the same server;
6168 (b) the server returned 5xx to all RCPT commands in the first message so
6169 that the message was not completed; (c) the server dropped the connection
6170 or gave a negative response to the RSET that Exim sends to abort the
6171 transaction. The observed case was a dropped connection after DATA that had
6172 been sent in pipelining mode. That is, the server had advertised PIPELINING
6173 but was not implementing it correctly. The effect of the bug was incorrect
6174 behaviour, such as trying another host, and this could lead to a crash.
6180 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6181 and daemon.c were passed as pointers to ints; they should have been
6182 pointers to socklen_t variables (which are typically unsigned ints).
6184 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6187 3. Fixed a really odd bug that affected only the testing scheme; patching a
6188 certain fixed string in the binary changed the value of another string that
6189 happened to be identical to the end of the original first string.
6191 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6192 name", it returns that address as the IP address. On some operating
6193 systems (e.g. Solaris), it also passes back the IP address string as the
6194 "host name". However, on others (e.g. Linux), it passes back an empty
6195 string. Exim wasn't checking for this, and was changing the host name to an
6196 empty string, assuming it had been canonicalized.
6198 5. Although rare, it is permitted to have more than one PTR record for a given
6199 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6200 all the names associated with an address, because they do in Solaris.
6201 However, it seems that they do not in Linux for data that comes from the
6202 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6203 I found this out when I moved to a new Linux workstation and tried to run
6204 the Exim test suite.
6206 To get round this problem I have changed the code so that it now does its
6207 own call to the DNS to look up PTR records when searching for a host name.
6208 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6209 addresses that are only in /etc/hosts are still found.
6211 This behaviour is, however, controlled by an option called host_lookup_
6212 order, which defaults to "bydns:byaddr". If people want to use the other
6213 order, or indeed, just use one or the other means of lookup, they can
6214 specify it in this variable.
6216 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6217 some operating systems, this comes back from gethostbyaddr() as an empty
6218 string, and this is what Exim used to test for. However, it seems that in
6219 other systems, "." is yielded. Exim now tests for this case too.
6221 7. The values of check_spool_space and check_log_space are now held internally
6222 as a number of kilobytes instead of an absolute number of bytes. If a
6223 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6224 kilobyte. This means that much larger values can be stored.
6226 8. Exim monitor: an attempt to get the action menu when not actually pointing
6227 at a message produces an empty menu entitled "No message selected". This
6228 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6229 no entries in it ("Shell widget menu has zero width and/or height"). So I
6230 have added a single, blank menu entry in this case.
6232 9. Added ${quote_local_part.
6234 10. MIME decoding is now applied to the contents of Subject: header lines when
6237 11. Now that a reference to $sender_host_address automatically causes a reverse
6238 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6239 host lookup before query-style lookups in lists that might use this
6240 variable. This has therefore been abolished, and the "net-" prefix is no
6241 longer necessary for query-style lookups.
6243 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6244 have been a typo for LFLAGS, so it has been changed.
6246 13. The install script calls Exim with "-C /dev/null" in order to find the
6247 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6248 to be output. However, since Exim outputs its version number before the
6249 error, it didn't break the script. It just looked ugly. I fixed this by
6250 always allowing "-C /dev/null" if the caller is root.
6252 14. Ignore overlarge ACL variable number when reading spool file - insurance
6253 against a later release with more variables having written the file.
6255 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6256 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6257 this, as well as the form without the "IPv6" on the front.
6259 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6260 OS/Makefile-Darwin file.
6262 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6263 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6265 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6266 variable. However, some OS use ints for inodes. I've added cast to long int
6267 to get rid of the compiler warning.
6269 19. I had forgotten to lock out "/../" in configuration file names when
6270 ALT_CONFIG_PREFIX was set.
6272 20. Routers used for verification do not need to specify transports. However,
6273 if such a router generated a host list, and callout was configured, Exim
6274 crashed, because it could not find a port number from the (non-existent)
6275 transport. It now assumes port 25 in this circumstance.
6277 21. Added the -t option to exigrep.
6279 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6280 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6281 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6282 them is set, LOOKUP_LSEARCH is forced.
6284 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6286 24. Added sender and host information to the "rejected by local_scan()" log
6287 line; previously there was no indication of these.
6289 25. Added .include_if_exists.
6291 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6292 Linux. It turns out that not all file systems support this. Apparently some
6293 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6294 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6295 not supported on the file descriptor, is now ignored when Exim is trying to
6296 sync a directory. This applies only to Linux.
6298 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6300 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6301 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6304 29. Added support for maildirsize files from supplied patch (modified a bit).
6306 30. The use of :fail: followed by an empty string could lead Exim to respond to
6307 sender verification failures with (e.g.):
6309 550 Verification failed for <xxx>
6310 550 Sender verify failed
6312 where the first response line was missing the '-' that indicates it is not
6313 the final line of the response.
6315 31. The loop for finding the name of the user that called Exim had a hardwired
6316 limit of 10; it now uses the value of finduser_retries, which is used for
6317 all other user lookups.
6319 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6322 33. Exim was neglecting to zero errno before one call of strtol() when
6323 expanding a string and expecting an integer value. On some systems this
6324 resulted in spurious "integer overflow" errors. Also, it was casting the
6325 result into an int without checking.
6327 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6328 did not work. The code looks as if it has *never* worked, though it appears
6329 to have been documented since at least release 1.62. I have made it work.
6331 35. The "timeout_DNS" error in retry rules, also documented since at least
6332 1.62, also never worked. As it isn't clear exactly what this means, and
6333 clearly it isn't a major issue, I have abolished the feature by treating it
6334 as "timeout", and writing a warning to the main and panic logs.
6336 36. The display of retry rules for -brt wasn't always showing the error code
6339 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6340 timeout_connect_A, timeout_connect_MX.
6342 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6343 to the empty sender.
6345 39. The daemon was not analysing the content of -oX till after it had closed
6346 stderr and disconnected from the controlling terminal. This meant that any
6347 syntax errors were only noted on the panic log, and the return code from
6348 the command was 0. By re-arranging the code a little, I've made the
6349 decoding happen first, so such errors now appear on stderr, and the return
6350 code is 1. However, the actual setting up of the sockets still happens in
6351 the disconnected process, so errors there are still only recorded on the
6354 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6355 connections (as happens on some IP stacks) was logged at start up time as
6356 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6357 it from "IPv6 and IPv4", which means that two separate sockets are being
6360 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6361 says whether AF_INET or AF_INET6 was passed as an argument.
6363 42. Exiwhat output was messed up when time zones were included in log
6366 43. Exiwhat now gives more information about the daemon's listening ports,
6367 and whether -tls-on-connect was used.
6369 44. The "port" option of the smtp transport is now expanded.
6371 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6372 silently ignored. Now an error message is written to the main and panic
6375 46. There's a new ACL modifier called "logwrite" which writes to a log file
6376 as soon as it is encountered.
6378 47. Added $local_user_uid and $local_user_gid at routing time.
6380 48. Exim crashed when trying to verify a sender address that was being
6383 49. Exim was recognizing only a space character after ".include". It now also
6384 recognizes a tab character.
6386 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6387 extracting the relevant information from the specification. The man page no
6388 longer contains scrambled data for the -d option, and I've added a section
6389 at the front about calling Exim under different names.
6391 51. Added "extra_headers" argument to the "mail" command in filter files.
6393 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6396 53. Installed eximstats 1.29.
6398 54. Added transport_filter_timeout as a generic transport option.
6400 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6401 Cc: header lines. This was required by RFC 822, but it not required by RFC
6404 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6405 incoming messages that did not have them. Now it does so only if the
6406 message originates locally, that is, if there is no associated remote host
6407 address. When Resent- header lines are present, this applies to the Resent-
6408 lines rather than the non-Resent- lines.
6410 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6411 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6413 58. Messages for configuration errors now include the name of the main
6414 configuration file - useful now that there may be more than one file in a
6415 list (.included file names were always shown).
6417 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6418 for those rare installations that do not start the daemon as root or run it
6419 setuid root. I've cut out the call to initgroups() if the daemon is not
6422 60. The Exim user and group can now be bound into the binary as text strings
6423 that are looked up at the start of Exim's processing.
6425 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6427 62. Added $mailstore_basename variable.
6429 63. Installed patch to sieve.c from Michael Haardt.
6431 64. When Exim failed to open the panic log after failing to open the main log,
6432 the original message it was trying to log was written to stderr and debug
6433 output, but if they were not available (the usual case in production), it
6434 was lost. Now it is written to syslog before the two lines that record the
6435 failures to open the logs.
6437 65. Users' Exim filters run in subprocesses under the user's uid. It is
6438 possible for a "deliver" command or an alias in a "personal" command to
6439 provoke an address rewrite. If logging of address rewriting is configured,
6440 this fails because the process is not running as root or exim. There may be
6441 a better way of dealing with this, but for the moment (because 4.30 needs
6442 to be released), I have disabled address rewrite logging when running a
6443 filter in a non-root, non-exim process.
6449 1. The buildconfig auxiliary program wasn't quoting the value set for
6450 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6451 not defined. This bug was masked in 4.22 by the effect that was fixed in
6454 2. Some messages that were rejected after a message id was allocated were
6455 shown as "incomplete" by exigrep. It no longer does this for messages that
6456 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6458 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6459 have allow_domain_literals set, the ID did not get logged in the <= line.
6460 Domain literals are now always recognized in Message-ID: header lines.
6462 4. The first argument for a ${extract expansion item is the key name or field
6463 number. Leading and trailing spaces in this item were not being ignored,
6464 causing some misleading effects.
6466 5. When deliver_drop_privilege was set, single queue runner processes started
6467 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6468 same command in the process it spins off) were not dropping privilege.
6470 6. When the daemon running as "exim" started a queue runner, it always
6471 re-executed Exim in the spun-off process. This is a waste of effort when
6472 deliver_drop_privilege is set. The new process now just calls the
6473 queue-runner function directly.
6479 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6482 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6483 ignored. Though the use of -oP was forcing the writing of a pid file, it
6484 was always written to the default place.
6486 3. If the message "no IP address found for host xxxx" is generated during
6487 incoming verification, it is now followed by identification of the incoming
6488 connection (so you can more easily find what provoked it).
6490 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6492 5. Added some features to "harden" Exim a bit more against certain attacks:
6494 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6495 be put in Local/Makefile. This is like the never_users runtime option,
6496 but it cannot be overridden. The default setting is "root".
6498 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6499 prefix string with which any file named in a -C command line option
6502 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6503 is retained for -C and -D only if the caller of Exim is root. Without
6504 it, the exim user may also use -C and -D and retain privilege.
6506 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6507 command line option is disabled.
6509 6. Macro names set by the -D option must start with an upper case letter, just
6510 like macro names defined in the configuration file.
6512 7. Added "dereference=" facility to LDAP.
6514 8. Two instances of the typo "uknown" in the source files are fixed.
6516 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6517 the Configure-Makefile script screwed up while processing it.
6519 10. Incorporated PCRE 4.4.
6521 11. The SMTP synchronization check was not operating right at the start of an
6522 SMTP session. For example, it could not catch a HELO sent before the client
6523 waited for the greeting. There is now a check for outstanding input at the
6524 point when the greeting is written. Because of the duplex, asynchronous
6525 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6526 way, but not yet received, when the check is performed.
6528 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6529 on TCP/IP sockets, because this apparently causes some broken clients to
6532 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6533 unchanged) from the Cygwin maintainer.
6535 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6536 was supported instead of testing for mbx. Effectively a typo.
6538 15. The spa authenticator server code was not checking that the input it
6539 received was valid base64.
6541 16. The debug output line for the "set" modifier in ACLs was not showing the
6542 name of the variable that was being set.
6544 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6546 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6547 lookup on its own. Something else was needed to trigger the lookup. For
6548 example, a match in host_lookup or the need for a host name in a host list.
6549 Now, if $sender_host_name is referenced and the host name has not yet been
6550 looked up, a lookup is performed. If the lookup fails, the variable remains
6551 empty, and $host_lookup_failed is set to "1".
6553 19. Added "eqi" as a case-independent comparison operator.
6555 20. The saslauthd authentication condition could segfault if neither service
6556 nor realm was specified.
6558 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6559 error message that was logged was misleading, and incoming SMTP
6560 connections were dropped. The message is now more accurate, and temporary
6561 errors are given to SMTP connections.
6563 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6564 (except RSET) with a 421 error, until QUIT is received. However, it was
6565 failing to send a response to QUIT.
6567 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6568 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6569 ACL is now run after the helo_try_verify_hosts code.
6571 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6572 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6573 case-independent, so other case variants are also recognized). Apparently
6574 some systems use these upper case variants.
6576 25. If more than two messages were waiting for the same host, and a transport
6577 filter was specified for the transport, Exim sent two messages over the
6578 same TCP/IP connection, and then failed with "socket operation on non-
6579 socket" when it tried to send the third.
6581 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6583 27. The extern definition of crypt16() in expand.c was not being excluded when
6584 the OS had its own crypt16() function.
6586 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6587 as a preferred synonym for return_size_limit, both as an option and as an
6590 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6592 30. Changed the default configuration ACL to relax the local part checking rule
6593 for addresses that are not in any local domains. For these addresses,
6594 slashes and pipe symbols are allowed within local parts, but the sequence
6595 /../ is explicitly forbidden.
6597 31. SPA server authentication was not clearing the challenge buffer before
6600 32. log_message in a "warn" ACL statement was writing to the reject log as
6601 well as to the main log, which contradicts the documentation and doesn't
6602 seem right (because no rejection is happening). So I have stopped it.
6604 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6605 However, I am unable to do any testing of this.
6607 34. Fixed an infelicity in the appendfile transport. When checking directories
6608 for a mailbox, to see if any needed to be created, it was accidentally
6609 using path names with one or more superfluous leading slashes; tracing
6610 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6612 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6613 contents are added to the log line that is written for every discarded
6614 recipient. (Previously a log_message setting was ignored.)
6616 36. The ${quote: operator now quotes the string if it is empty.
6618 37. The install script runs exim in order to find its version number. If for
6619 some reason other than non-existence or emptiness, which it checks, it
6620 could not run './exim', it was installing it with an empty version number,
6621 i.e. as "exim-". This error state is now caught, and the installation is
6624 38. An argument was missing from the function that creates an error message
6625 when Exim fails to connect to the socket for saslauthd authentication.
6626 This could cause Exim to crash, or give a corrupted message.
6628 39. Added isip, isip4, and isip6 to ${if conditions.
6630 40. The ACL variables $acl_xx are now saved with the message, and can be
6631 accessed later in routers, transports, and filters.
6633 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6634 strings in the file are not string-expanded.
6636 42. If a MAIL command specified a SIZE value that was too large to fit into an
6637 int variable, the check against message_size_limit failed. Such values are
6638 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6639 day this will have to be increased, but I don't think I want to be around
6640 when emails are that large.
6647 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6648 iconv() is not standard in FreeBSD.
6650 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6651 IPv6 enabled. The observed symptom was a segmentation fault on return from
6652 the function os_common_find_running_interfaces() in src/os.c.
6654 3. In the check_special_case() function in daemon.c I had used "errno" as an
6655 argument name, which causes warnings on some systems. This was basically a
6656 typo, since it was named "eno" in the comments!
6658 4. The code that waits for the clock to tick (at a resolution of some fraction
6659 of a second) so as to ensure message-id uniqueness was always waiting for
6660 at least one whole tick, when it could have waited for less. [This is
6661 almost certainly not relevant at current processor speeds, where it is
6662 unlikely to ever wait at all. But we try to future-proof.]
6664 5. The function that sleeps for a time interval that includes fractions of a
6665 second contained a race. It did not block SIGALRM between setting the
6666 timer, and suspending (a couple of lines later). If the interval was short
6667 and the sigsuspend() was delayed until after it had expired, the suspension
6668 never ended. On busy systems this could lead to processes getting stuck for
6671 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6672 process, before it forks any delivery processes. The open lookup caching
6673 mechanism meant that the open file or database connection was passed into
6674 the delivery process. The problem was that delivery processes always tidy
6675 up cached lookup data. This could cause a problem for the next delivery
6676 process started by the queue runner, because the external queue runner
6677 process does not know about the closure. So the next delivery process
6678 still has data in the lookup cache. In the case of a file lookup, there was
6679 no problem because closing a file descriptor in a subprocess doesn't affect
6680 the parent. However, if the lookup was caching a connection to a database,
6681 the connection was closed, and the second delivery process was likely to
6682 see errors such as "PGSQL: query failed: server closed the connection
6683 unexpectedly". The problem has been fixed by closing all cached lookups
6684 in a queue runner before running a delivery process.
6686 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6687 seem to have the "const" qualifier which it has on other OS. I've
6690 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6691 *of the same type* (client or server) with the same public name that an
6692 error should be diagnosed.
6694 9. When Exim looked up a host name for an IP address, but failed to find the
6695 original IP address when looking up the host name (a safety check), it
6696 output the message "<ip address> does not match any IP for NULL", which was
6697 confusing, to say the least. The bug was that the host name should have
6698 appeared instead of "NULL".
6700 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6701 user that is built into the binary, and the -C or -D options is used, root
6702 privilege is dropped before the configuration file is read. In addition,
6703 logging is switched to stderr instead of the normal log files. If the
6704 configuration then re-defines the Exim user, the unprivileged environment
6705 is probably not what is expected, so Exim logs a panic warning message (but
6708 However, if deliver_drop_privilege is set, the unprivileged state may well
6709 be exactly what is intended, so the warning has been cut out in that case,
6710 and Exim is allowed to try to write to its normal log files.
6716 1. smtp_return_error_details was not giving details for temporary sender
6717 or receiver verification errors.
6719 2. Diagnose a configuration error if two authenticators have the same public
6722 3. Exim used not to create the message log file for a message until the first
6723 delivery attempt. This could be confusing when incoming messages were held
6724 for policy or load reasons. The message log file is now created at the time
6725 the message is received, and an initial "Received" line is written to it.
6727 4. The automatically generated man page for command line options had a minor
6728 bug that caused no ill effects; however, a more serious problem was that
6729 the procedure for building the man page automatically didn't always
6730 operate. Consequently, release 4.20 contains an out-of-date version. This
6731 shouldn't happen again.
6733 5. When building Exim with embedded Perl support, the script that builds the
6734 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6735 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6737 6. The freeze_tell option was not being used for messages that were frozen on
6738 arrival, either by an ACL or by local_scan().
6740 7. Added the smtp_incomplete_transaction log selector.
6742 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6743 was accepting AUTH without a new EHLO.
6745 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6746 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6749 10. Logging of TCP/IP connections (when configured) now happens in the main
6750 daemon process instead of the child process, so that the TCP/IP connection
6751 count is more accurate (but it can never be perfect).
6753 11. The use of "drop" in a nested ACL was not being handled correctly in the
6754 outer ACL. Now, if condition failure induced by the nested "drop" causes
6755 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6756 or "require"), the connection is dropped.
6758 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6759 that yield "discard" can now be used with an "accept" or a "discard" verb,
6760 but an error is generated for any others (because I can't see a useful way
6761 to define what should happen).
6763 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6764 are now processed in the same way as lines in the Exim configuration file.
6765 In particular, continuation lines are supported.
6767 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6769 15. Added -ti meaning -t -i.
6771 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6772 domains, and warn by logging if others are found.
6774 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6775 returned from GIFCONF when Exim is trying to find the list of interfaces on
6776 a host. The code in os.c has been modified to copy each ifreq to an aligned
6777 structure in all cases.
6779 Also, in some cases, the returned ifreq's were being copied to a 'struct
6780 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6781 means the last couple of bytes of an IPv6 address could be chopped if the
6782 ifreq contained only a normal sockaddr (14 bytes storage).
6784 18. Named domain lists were not supported in the hosts_treat_as_local option.
6785 An entry such as +xxxx was not recognized, and was treated as a literal
6788 19. Ensure that header lines added by a DATA ACL are included in the reject log
6789 if the ACL subsequently rejects the message.
6791 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6792 MD5 (which is deprecated).
6794 21. When testing a filter file using -bf, Exim was writing a message when it
6795 took the sender from a "From " line in the message, but it was not doing so
6796 when it took $return_path from a Return-Path: header line. It now does.
6798 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6799 with a valid header line field name (a series of printing characters
6800 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6802 23. Changed "disc" in the source to "disk" to conform to the documentation and
6803 the book and for uniformity.
6805 24. Ignore Sendmail's -Ooption=value command line item.
6807 25. When execve() failed while trying to run a command in a pipe transport,
6808 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6809 could be confused with a return value of 69 from the command itself. This
6810 has been changed to 127, the value the shell returns if it is asked to run
6811 a non-existent command. The wording for the related log line suggests a
6812 non-existent command as the problem.
6814 26. If received_header_text expands to an empty string, do not add a Received:
6815 header line to the message. (Well, it adds a token one on the spool, but
6816 marks it "old" so that it doesn't get used or transmitted.)
6818 27. Installed eximstats 1.28 (addition of -nt option).
6820 28. There was no check for failure on the call to getsockname() in the daemon
6821 code. This can fail if there is a shortage of resources on the system, with
6822 ENOMEM, for example. A temporary error is now given on failure.
6824 29. Contrary to the C standard, it seems that in some environments, the
6825 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6826 program. Exim now does this explicitly; it affects the formatting of
6827 timestamps using strftime().
6829 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6830 complaints. I've now initialized all the variables, to avoid this.
6832 32. Header lines added by a system filter were not being "seen" during
6833 transport-time rewrites.
6835 33. The info_callback() function passed to OpenSSL is set up with type void
6836 (*)(SSL *, int, int), as described somewhere. However, when calling the
6837 function (actually a macro) that sets it up, the type void(*)() is
6838 expected. I've put in a cast to prevent warnings from picky compilers.
6840 34. If a DNS black list lookup found a CNAME record, but there were no A
6841 records associated with the domain it pointed at, Exim crashed.
6843 35. If a DNS black list lookup returned more than one A record, Exim ignored
6844 all but the first. It now scans all returned addresses if a particular IP
6845 value is being sought. In this situation, the contents of the
6846 $dnslist_value variable are a list of all the addresses, separated by a
6849 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6850 to accept a host name and all its aliases if the forward lookup for any of
6851 them yielded the IP address of the incoming connection. Now it accepts only
6852 those names whose forward lookup yields the correct IP address. Any other
6853 names are discarded. This closes a loophole whereby a rogue DNS
6854 administrator could create reverse DNS records to break through a
6855 wildcarded host restriction in an ACL.
6857 37. If a user filter or a system filter that ran in a subprocess used any of
6858 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6859 the wrong values were passed to the pipe command ($thisaddress had the
6860 value of $0, $0 had the value of $1, etc). This bug was introduced by
6861 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6863 38. Improved the line breaking for long SMTP error messages from ACLs.
6864 Previously, if there was no break point between 40 and 75 characters, Exim
6865 left the rest of the message alone. Two changes have been made: (a) I've
6866 reduced the minimum length to 35 characters; (b) if it can't find a break
6867 point between 35 and 75 characters, it looks ahead and uses the first one
6868 that it finds. This may give the occasional overlong line, but at least the
6869 remaining text gets split now.
6871 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6872 file descriptors might be low, and that setting 1000 would always raise it.
6873 It turns out that in some environments, the limit is already over 1000 and
6874 that lowering it causes trouble. So now Exim takes care not to decrease it.
6876 40. When delivering a message, the value of $return_path is set to $sender_
6877 address at the start of routing (routers may change the value). By an
6878 oversight, this default was not being set up when an address was tested by
6879 -bt or -bv, which affected the outcome if any router or filter referred to
6882 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6883 log only when "message" or "log_message" are set. However, if one of the
6884 conditions was an address verification, or a call to a nested ACL, the
6885 messages generated by the underlying test were being passed through. This
6886 no longer happens. The underlying message is available in $acl_verify_
6887 message for both "message" and "log_message" expansions, so it can be
6888 passed through if needed.
6890 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6891 new expansion $bh_ to give the encoded byte string without charset
6892 translation. Translation happens only if iconv() is available; HAVE_ICONV
6893 indicates this at build time. HEADERS_CHARSET gives the charset to
6894 translate to; headers_charset can change it in the configuration, and
6895 "headers charset" can change it in an individual filter file.
6897 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6898 that creates RFC 2047 encoded "words" labels them as that charset instead
6899 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6900 expansion operator; (ii) when Exim creates a From: line for a local
6901 message; (iii) when a header line is rewritten to include a "phrase" part.
6903 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6904 buggy, causing it to skip the first lines of messages whose message ID
6905 ended in 'D'. This would not have bitten before Exim release 4.14, because
6906 message IDs were unlikely to end in 'D' before then. The effect was to have
6907 incorrect size information for certain domains.
6909 45. #include "config.h" was missing at the start of the crypt16.c module. This
6910 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6913 46. If there was a timeout during a "random" callout check, Exim treated it as
6914 a failure of the random address, and carried on sending RSET and the real
6915 address. If the delay was just some slowness somewhere, the response to the
6916 original RCPT would be taken as a response to RSET and so on, causing
6917 mayhem of various kinds.
6919 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6920 when I implemented it. It didn't allow for the fact that some option values
6921 may legitimately be negative (e.g. size_addition), and it didn't even do
6922 the right test for positive values.
6924 48. Domain names in DNS records are case-independent. Exim always looks them up
6925 in lower case. Some resolvers return domain names in exactly the case they
6926 appear in the zone file, that is, they may contain uppercase letters. Not
6927 all resolvers do this - some return always lower case. Exim was treating a
6928 change of case by a resolver as a change of domain, similar to a widening
6929 of a domain abbreviation. This triggered its re-routing code and so it was
6930 trying to route what was effectively the same domain again. This normally
6931 caused routing to fail (because the router wouldn't handle the domain
6932 twice). Now Exim checks for this case specially, and just changes the
6933 casing of the domain that it ultimately uses when it transmits the message
6936 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6939 50. If a filter generated a file delivery with a non-absolute name (possible if
6940 no home directory exists for the router), the forbid_file option was not
6943 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6944 the existing equality matching.
6946 52. Exim was using ints instead of ino_t variables in some places where it was
6947 dealing with inode numbers.
6949 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6950 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6951 called TMPDIR, and if it finds it is different, it changes its value.
6953 54. The smtp_printf() function is now made available to local_scan() so
6954 additional output lines can be written before returning. There is also an
6955 smtp_fflush() function to enable the detection of a dropped connection.
6956 The variables smtp_input and smtp_batched_input are exported to
6959 55. Changed the default runtime configuration: the message "Unknown user"
6960 has been removed from the ACL, and instead placed on the localuser router,
6961 using the cannot_route_message feature. This means that any verification
6962 failures that generate their own messages won't get overridden. Similarly,
6963 the "Unrouteable address" message that was in the ACL for unverifiable
6964 relay addresses has also been removed.
6966 56. Added hosts_avoid_esmtp to the smtp transport.
6968 57. The exicyclog script was not checking for the esoteric option
6969 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6970 will work only if exicyclog is run under the appropriate euid.
6972 58. Following a discussion on the list, the rules by which Exim recognises line
6973 endings on incoming messages have been changed. The -dropcr and drop_cr
6974 options are now no-ops, retained only for backwards compatibility. The
6975 following line terminators are recognized: LF CRLF CR. However, special
6976 processing applies to CR:
6978 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6979 nor a local message in the state where . is a terminator.
6981 (ii) If a bare CR is encountered in a header line, an extra space is added
6982 after the line terminator so as not to end the header. The reasoning
6983 behind this is that bare CRs in header lines are most likely either
6984 to be mistakes, or people trying to play silly games.
6986 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6987 was being incorrectly given as 18 bytes larger than it should have been.
6988 This is a VOB (very old bug).
6990 60. This may never have affected anything current, but just in case it has:
6991 When the local host is found other than at the start of a list of hosts,
6992 the local host, those with the same MX, and any that follow, are discarded.
6993 When the list in question was part of a longer list of hosts, the following
6994 hosts (not currently being processed) were also being discarded. This no
6995 longer happens. I'm not sure if this situation could ever has previously
6998 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7001 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7002 header. This is something that is recommended in a new Internet Draft, and
7003 is something that is documented as being done by Sendmail. There are two
7004 possible values. For messages generated by the autoreply transport, Exim
7007 Auto-Submitted: auto-replied
7009 whereas for all other generated messages (e.g. bounces) it adds
7011 Auto-Submitted: auto-generated
7013 63. The "personal" condition in filters now includes a test for the
7014 Auto-Submitted: header. If it contains the string "auto-" the message it
7015 not considered personal.
7017 64. Added rcpt_include_affixes as a generic transport option.
7019 65. Added queue_only_override (default true).
7021 66. Added the syslog_duplication option.
7023 67. If what should have been the first header line of a message consisted of
7024 a space followed by a colon, Exim was mis-interpreting it as a header line.
7025 It isn't of course - it is syntactically invalid and should therefore be
7026 treated as the start of the message body. The misbehaviour could have
7027 caused a number of strange effects, including loss of data in subsequent
7028 header lines, and spool format errors.
7030 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7031 client host had authenticated. This control can now be exercised by an ACL
7032 for more flexibility.
7034 69. By default, callouts do not happen when testing with -bh. There is now a
7035 variant, -bhc, which does actually run the callout code, including
7036 consulting and updating the callout cache.
7038 70. Added support for saslauthd authentication, courtesy of Alexander
7041 71. If statvfs() failed on the spool or log directories while checking their
7042 size for availability, Exim confusingly gave the error "space shortage".
7043 Furthermore, in debugging mode it crashed with a floating point exception.
7044 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7045 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7046 serious problem, Exim now writes to the main and panic logs when this
7047 happens, with details of the failure. It then refuses to accept the
7048 incoming message, giving the message "spool directory problem" or "log
7049 directory problem" with a 421 code for SMTP messages.
7051 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7052 0, 1, and 2 exist, because some OS complain for execs without them (see
7053 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7054 descriptors. However, the code omitted to check that the open succeeded,
7055 causing mysterious errors if for some reason the permissions on /dev/null
7056 got screwed. Now Exim writes a message to the main and panic logs, and
7057 bombs out if it can't open /dev/null.
7059 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7060 interact so that it is all more flexible. It is supposed to remain
7061 backwards compatible. Also added extra_local_interfaces.
7063 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7064 to bomb out with an assertion failure - to the client this appears as a
7065 connection drop. This problem occurs in the part of the code that was taken
7066 from the Samba project. Fortunately, the assertion is in a very simple
7067 function, so I have fixed this by reproducing the function inline in the
7068 one place where it is called, and arranging for authentication to fail
7069 instead of killing the process with assert().
7071 75. The SPA client code was not working when the server requested OEM rather
7072 than Unicode encoding.
7074 76. Added code to make require_files with a specific uid setting more usable in
7075 the case where statting the file as root fails - usually a non-root-mounted
7076 NFS file system. When this happens and the failure is EACCES, Exim now
7077 forks a subprocess and does the per-uid checking as the relevant uid.
7079 77. Added process_log_path.
7081 78. If log_file_path was not explicitly set, a setting of check_log_space or
7082 check_log_inodes was ignored.
7084 79. If a space check for the spool or log partitions fails, the incident is now
7085 logged. Of course, in the latter case the data may get lost...
7087 80. Added the %p formatting code to string_format() so that it can be used to
7088 print addresses in debug_print(). Adjusted all the address printing in the
7089 debugging in store.c to use %p rather than %d.
7091 81. There was a concern that a line of code in smtp_in.c could overflow a
7092 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7093 initially expressed, the concern was not well-founded, because trailing
7094 spaces are removed early. However, if the trailing spaces were followed by
7095 a NULL, they did not get removed, so the overflow was possible. Two fixes
7098 (a) I re-wrote the offending code in a cleaner fashion.
7099 (b) If an incoming SMTP command contains a NULL character, it is rejected
7102 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7103 runs initgroups(), so that if the Exim user is in any additional groups,
7104 they will be used during message reception.
7110 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob