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
137 JH/01 Rework error string handling in TLS interface so that the caller in
138 more cases is responsible for logging. This permits library-sourced
139 string to be attached to addresses during delivery, and collapses
140 pairs of long lines into single ones.
142 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
143 during configuration. Wildcards are allowed and expanded.
145 JH/02 Rework error string handling in DKIM to pass more info back to callers.
146 This permits better logging.
148 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
149 do not close it down and have the child transport start it up again on
150 the passed-on TCP connection. Instead, proxy the child (and any
151 subsequent ones) for TLS via a unix-domain socket channel. Logging is
152 affected: the continued delivery log lines do not have any DNSSEC, TLS
153 Certificate or OCSP information. TLS cipher information is still logged.
155 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
156 identical IP addresses on different listening ports. Will also affect
159 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
160 add noisy ifdef guards to special-case this sillyness.
161 Patch from Bernd Kuhls.
163 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
164 than 255 are no longer allowed.
166 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
167 Disable the session-cache too, which might reduce our load. Since we
168 currrectly use a new context for every connection, both as server and
169 client, there is no benefit for these.
170 GnuTLS appears to not support tickets server-side by default (we don't
171 call gnutls_session_ticket_enable_server()) but client side is enabled
172 by default on recent versions (3.1.3 +) unless the PFS priority string
175 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
176 <https://reproducible-builds.org/specs/source-date-epoch/>.
178 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
179 the check for any unsuccessful recipients did not notice the limit, and
180 erroneously found still-pending ones.
182 JH/08 Pipeline CHUNKING command and data together, on kernels that support
183 MSG_MORE. Only in-clear (not on TLS connections).
185 JH/09 Avoid using a temporary file during transport using dkim. Unless a
186 transport-filter is involved we can buffer the headers in memory for
187 creating the signature, and read the spool data file once for the
188 signature and again for transmission.
190 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
191 4.77 as the kernel support then wasn't solid, having issues in 64bit
192 mode. Now, it's been long enough. Add support for FreeBSD also.
194 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
195 case where the routing stage had gathered several addresses to send to
196 a host before calling the transport for the first, we previously failed
197 to close down TLS in the old transport process before passing the TCP
198 connection to the new process. The new one sent a STARTTLS command
199 which naturally failed, giving a failed delivery and bloating the retry
200 database. Investigation and fix prototype from Wolfgang Breyha.
202 JH/12 Fix check on SMTP command input synchronisation. Previously there were
203 false-negatives in the check that the sender had not preempted a response
204 or prompt from Exim (running as a server), due to that code's lack of
205 awareness of the SMTP input buffering.
207 PP/04 Add commandline_checks_require_admin option.
208 Exim drops privileges sanely, various checks such as -be aren't a
209 security problem, as long as you trust local users with access to their
210 own account. When invoked by services which pass untrusted data to
211 Exim, this might be an issue. Set this option in main configuration
212 AND make fixes to the calling application, such as using `--` to stop
215 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
216 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
217 responses to those, into a single TLS record each way (this usually means
218 a single packet). As a side issue, smtp_enforce_sync now works on TLS
221 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
222 affects you only if you're dancing at the edge of the param size limits.
223 If you are, and this message makes sense to you, then: raise the
224 configured limit or use OpenSSL 1.1. Nothing we can do for older
227 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
228 cmdline element to get the documented default one. Previously it was
231 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
232 are unsuable for non-message-specific reasons - eg. connection timeout,
233 banner-time rejection.
235 JH/16 Fix logging of delivery remote port, when specified by router, under
238 PP/06 Repair manualroute's ability to take options in any order, even if one
239 is the name of a transport.
242 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
244 JH/17 Change the list-building routines interface to use the expanding-string
245 triplet model, for better allocation and copying behaviour.
247 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
248 Previously it was constructed the first time a possibly-matching string
249 was met in the configuration file input during startup; now it is done
252 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
253 compatible one, to avoid the (poorly documented) possibility of a config
254 file in the working directory redirecting the DB files, possibly correpting
255 some existing file. CVE-2017-10140 assigned for BDB.
257 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
258 cache-hot. Previously, although the result was properly cached, the
259 initial verify call returned a defer.
261 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
262 the main verify for receipient in uncached-mode.
264 JH/22 Retire historical build files to an "unsupported" subdir. These are
265 defined as "ones for which we have no current evidence of testing".
267 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
268 if present. Previously it was ignored.
270 JH/24 Start using specified-initialisers in C structure init coding. This is
271 a C99 feature (it's 2017, so now considered safe).
273 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
274 if was a fixed-sized field and bitmask ops via macros; it is now more
277 PP/07 GitHub PR 56: Apply MariaDB build fix.
278 Patch provided by Jaroslav Škarvada.
280 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
281 during Coverity cleanups [4.87 JH/47]
282 Diagnosis and fix provided by Michael Fischer v. Mollard.
284 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
285 the right size to place the terminating semicolon on its own folded
286 line, the header hash was calculated to an incorrect value thanks to
287 the (relaxed) space the fold became.
289 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
290 and confused the parent.
292 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
293 which could crash as a result. This could lead to undeliverable messages.
295 JH/28 Logging: "next input sent too soon" now shows where input was truncated
298 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
299 matters on fast-turnover and PID-randomising systems, which were getting
300 out-of-order delivery.
302 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
303 a possibly-overlapping copy. The symptom was that "Remote host closed
304 connection in response to HELO" was logged instead of the actual 4xx
307 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
308 Previously only that bufferd was discarded, resulting in SYMTP command
311 JH/32 DKIM: when a message has multiple signatures matching an identity given
312 in dkim_verify_signers, run the dkim acl once for each. Previously only
313 one run was done. Bug 2189.
315 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
316 "panic the current process" to "deliberately defer". The panic log is
317 still written with the problem list name; the mail and reject logs now
318 get a temp-reject line for the message that was being handled, saying
319 something like "domains check lookup or other defer". The SMTP 451
320 message is still "Temporary local problem".
322 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
323 A crafted sequence of BDAT commands could result in in-use memory beeing
324 freed. CVE-2017-16943.
326 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
327 from SMTP input. Previously it was always done; now only done for DATA
328 and not BDAT commands. CVE-2017-16944.
330 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
331 to the message (such as an overlong header line). Previously this was
332 not done and we did not exit BDAT mode. Followon from the previous item
333 though a different problem.
339 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
340 than -2003 did; needs libidn2 in addition to libidn.
342 JH/02 The path option on a pipe transport is now expanded before use.
344 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
345 Patch provided by "Björn", documentation fix added too.
347 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
348 missing a wire-to-host endian conversion.
350 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
351 close after a BDAT command line could be taken as a following command,
352 giving a synch failure. Fix by only checking for synch immediately
353 before acknowledging the chunk.
355 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
356 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
357 macro. Patches provided by Josh Soref.
359 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
360 Previously we did not; the RFC seems ambiguous and VRFY is not listed
361 by IANA as a service extension. However, John Klensin suggests that we
364 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
365 the dkim code may be unix-mode line endings rather than smtp wire-format
366 CRLF, so prepend a CR to any bare LF.
368 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
369 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
371 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
372 the first were themselves being wrongly included in the feed into dkim
373 processing; with most chunk sizes in use this resulted in an incorrect
374 body hash calculated value.
376 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
377 DKIM signature block, for verification. Although advised against by
378 standards it is specifically not ruled illegal.
380 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
382 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
383 missing a body hash (the bh= tag).
385 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
386 It seems that HAProxy sends the Proxy Protocol information in clear and
387 only then does a TLS startup, so do the same.
389 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
390 TCP connections (such as for Spamd) unless the daemon successfully set
391 Fast Open mode on its listening sockets. This fixes breakage seen on
392 too-old kernels or those not configured for Fast Open, at the cost of
393 requiring both directions being enabled for TFO, and TFO never being used
394 by non-daemon-related Exim processes.
396 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
397 endings, at least on the first header line. Try to canonify any that get
398 past that check, despite the cost.
400 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
401 now limited to an arbitrary five deep, while parsing addresses with the
402 strip_excess_angle_brackets option enabled.
404 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
405 instead leave the unprompted TLS handshake in socket buffer for the
406 TLS library to consume.
408 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
410 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
412 JH/16 Drop variables when they go out of scope. Memory management drops a whole
413 region in one operation, for speed, and this leaves assigned pointers
414 dangling. Add checks run only under the testsuite which checks all
415 variables at a store-reset and panics on a dangling pointer; add code
416 explicitly nulling out all the variables discovered. Fixes one known
417 bug: a transport crash, where a dangling pointer for $sending_ip_address
418 originally assigned in a verify callout, is re-used.
420 PP/06 Drop '.' from @INC in various Perl scripts.
422 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
424 PP/08 Reduce a number of compilation warnings under clang; building with
425 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
426 should be warning-free.
428 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
430 HS/01 Fix portability problems introduced by PP/08 for platforms where
431 realloc(NULL) is not equivalent to malloc() [SunOS et al].
433 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
434 chunk. This allows us to accept broken chunked messages. We need a more
435 general solution here.
437 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
438 already-broken messages in the queue.
440 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
442 JH/19 Fix reference counting bug in routing-generated-address tracking.
448 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
449 supports it and a size is available (ie. the sending peer gave us one).
451 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
452 years of being deprecated). The replacements are the ACLs
453 acl_smtp_mime and acl_not_smtp_mime.
455 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
456 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
457 or even an in-clear connection were permitted. Now, if the host lookup
458 was dnssec and dane was requested then the host is only used if the
459 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
460 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
461 if one fails this test.
462 This means that a poorly-configured remote DNS will make it incommunicado;
463 but it protects against a DNS-interception attack on it.
465 JH/04 Bug 1810: make continued-use of an open smtp transport connection
466 non-noisy when a race steals the message being considered.
468 JH/05 If main configuration option tls_certificate is unset, generate a
469 self-signed certificate for inbound TLS connections.
471 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
472 in rewrites and routers.
474 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
475 and logged a warning sing 4.83; now they are a configuration file error.
477 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
478 (lacking @domain). Apply the same qualification processing as RCPT.
480 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
482 JH/10 Support ${sha256:} applied to a string (as well as the previous
485 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
486 a cutthrough deliver is pending, as we always want to make a connection.
487 This also avoids re-routing the message when later placing the cutthrough
488 connection after a verify cache hit.
489 Do not update it with the verify result either.
491 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
492 when routing results in more than one destination address.
494 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
495 signing (which inhibits the cutthrough capability). Previously only
496 the presence of an option was tested; now an expansion evaluating as
497 empty is permissible (obviously it should depend only on data available
498 when the cutthrough connection is made).
500 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
501 the relevant preceding SMTP command did not note the pipelining mode.
503 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
504 Previously they were not counted.
506 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
507 as one having no matching records. Previously we deferred the message
508 that needed the lookup.
510 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
511 distinguished as "(=".
513 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
514 for missing MX records. Previously it only worked for missing A records.
516 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
518 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
519 after the data-go-ahead and data-ack. Patch from Jason Betts.
521 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
522 even for a "none" policy. Patch from Tony Meyer.
524 JH/22 Fix continued use of a connection for further deliveries. If a port was
525 specified by a router, it must also match for the delivery to be
528 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
529 When one of the recipients of a message was unsuitable for the connection
530 (has no matching addresses), we lost track of needing to mark it
531 deferred. As a result mail would be lost.
533 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
535 JH/25 Decoding ACL controls is now done using a binary search; the source code
536 takes up less space and should be simpler to maintain. Merge the ACL
537 condition decode tables also, with similar effect.
539 JH/26 Fix problem with one_time used on a redirect router which returned the
540 parent address unchanged. A retry would see the parent address marked as
541 delivered, so not attempt the (identical) child. As a result mail would
544 JH/27 Fix a possible security hole, wherein a process operating with the Exim
545 UID can gain a root shell. Credit to http://www.halfdog.net/ for
546 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
549 JH/28 Enable {spool,log} filesystem space and inode checks as default.
550 Main config options check_{log,spool}_{inodes,space} are now
551 100 inodes, 10MB unless set otherwise in the configuration.
553 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
554 Previously it only applied to the main-section connection policy
557 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
559 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
560 by me. Added RFC7919 DH primes as an alternative.
562 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
563 are not in the system include path.
565 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
566 GnuTLS, when a session startup failed (eg because the client disconnected)
567 Exim did stdio operations after fclose. This was exposed by a recent
568 change which nulled out the file handle after the fclose.
570 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
571 signed directly by the cert-signing cert, rather than an intermediate
572 OCSP-signing cert. This is the model used by LetsEncrypt.
574 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
576 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
577 an incoming connection.
579 HS/02 Bug 1802: Do not half-close the connection after sending a request
582 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
583 fallback to "prime256v1".
585 JH/34 SECURITY: Use proper copy of DATA command in error message.
586 Could leak key material. Remotely exploitable. CVE-2016-9963.
592 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
593 and 3.4.4 - once the server is enabled to respond to an OCSP request
594 it does even when not requested, resulting in a stapling non-aware
595 client dropping the TLS connection.
597 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
598 support variable-length bit vectors. No functional change.
600 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
601 The I= interface field on outgoing lines is now after the H= remote
602 host field, same as incoming lines. There is a separate
603 outgoing_interface log selector which allows you to disable the
606 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
607 If not running log_selector +smtp_connection the mainlog would be held
608 open indefinitely after a "too many connections" event, including to a
609 deleted file after a log rotate. Leave the per net connection logging
610 leaving it open for efficiency as that will be quickly detected by the
611 check on the next write.
613 HS/01 Bug 1671: Fix post transport crash.
614 Processing the wait-<transport> messages could crash the delivery
615 process if the message IDs didn't exist for some reason. When
616 using 'split_spool_directory=yes' the construction of the spool
617 file name failed already, exposing the same netto behaviour.
619 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
620 mime_regex ACL conditions.
622 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
623 to DSN fail messages (bounces): remote IP, remote greeting, remote response
624 to HELO, local diagnostic string.
626 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
627 log line to debug. Even when configured with a tls authenticator many
628 client connections are expected to not authenticate in this way, so
629 an authenticate fail is not an error.
631 HS/02 Add the Exim version string to the process info. This way exiwhat
632 gives some more detail about the running daemon.
634 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
635 matter for fast-change records such as DNSBLs.
637 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
638 retry record, even if constant. There may be multiple transports with
639 different interface settings and the retry behaviour needs to be kept
642 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
644 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
646 JH/10 Bug 840: fix log_defer_output option of pipe transport
648 JH/11 Bug 830: use same host for all RCPTS of a message, even under
649 hosts_randomize. This matters a lot when combined with mua_wrapper.
651 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
652 ${quote_pgsql:<string>} operator.
654 JH/13 Bug 1708: avoid misaligned access in cached lookup.
656 JH/14 Change header file name for freeradius-client. Relevant if compiling
657 with Radius support; from the Gentoo tree and checked under Fedora.
659 JH/15 Bug 1712: Introduce $prdr_requested flag variable
661 JH/16 Bug 1714: Permit an empty string as expansion result for transport
662 option transport_filter, meaning no filtering.
664 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
666 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
667 defaults to "*" (all hosts). The variable is now available when not built
668 with TLS, default unset, mainly to enable keeping the testsuite sane.
669 If a server certificate is not supplied (via tls_certificate) an error is
670 logged, and clients will find TLS connections fail on startup. Presumably
671 they will retry in-clear.
672 Packagers of Exim are strongly encouraged to create a server certificate
673 at installation time.
675 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
676 with the $config_file variable.
678 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
679 in transport context, after the attempt, and per-recipient. The latter type
680 is per host attempted. The event data is the error message, and the errno
681 information encodes the lookup type (A vs. MX) used for the (first) host,
682 and the trailing two digits of the smtp 4xx response.
684 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
685 to write to mainlog (or rejectlog, paniclog) in the window between file
686 creation and permissions/ownership being changed. Particularly affects
687 installations where exicyclog is run as root, rather than exim user;
688 result is that the running daemon panics and dies.
690 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
692 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
693 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
694 "pri" and "weight". Note that the previous implicit priority given by the
695 list order is no longer honoured.
697 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
700 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
701 by defining SUPPORT_SOCKS.
703 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
704 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
705 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
706 variables are renamed to proxy_{local,external}_{address,port}.
708 JH/27 Move Internationalisation support from Experimental to mainline, enabled
709 for a build by defining SUPPORT_I18N
711 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
712 of the query string, and make ${quote_redis:} do that quoting.
714 JH/29 Move Events support from Experimental to mainline, enabled by default
715 and removable for a build by defining DISABLE_EVENT.
717 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
719 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
720 cached by the daemon.
722 JH/32 Move Redis support from Experimental to mainline, enabled for a build
723 by defining LOOKUP_REDIS. The libhiredis library is required.
725 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
726 keys are given for lookup.
728 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
729 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
730 only supported when built with TLS support. The PolarSSL SHA routines
731 are still used when the TLS library is too old for convenient support.
733 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
734 openssl_options), for security. OpenSSL forces this from version 1.1.0
735 server-side so match that on older versions.
737 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
738 allocation for $value could be released as the expansion processing
739 concluded, but leaving the global pointer active for it.
741 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
742 and to use the domains and local_parts ACL conditions.
744 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
745 incorrectly not doubled on cutthrough transmission, hence seen as a
746 body-termination at the receiving system - resulting in truncated mails.
747 Commonly the sender saw a TCP-level error, and retransmitted the message
748 via the normal store-and-forward channel. This could result in duplicates
749 received - but deduplicating mailstores were liable to retain only the
750 initial truncated version.
752 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
754 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
756 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
757 we're in there, support oversigning also; bug 1309.
759 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
761 HS/04 Add support for keep_environment and add_environment options.
763 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
764 either intentional arithmetic overflow during PRNG, or testing config-
767 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
768 delivery resulted in actual delivery. Cancel cutthrough before DATA
771 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
772 rejects a recipient: pass the reject to the originator.
774 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
775 Many were false-positives and ignorable, but it's worth fixing the
778 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
779 for the new environment-manipulation done at startup. Move the routines
780 from being local to tls.c to being global via the os.c file.
782 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
783 an extract embedded as result-arg for a map, the first arg for extract
784 is unavailable so we cannot tell if this is a numbered or keyed
785 extraction. Accept either.
791 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
794 JH/02 The smtp transport option "multi_domain" is now expanded.
796 JH/03 The smtp transport now requests PRDR by default, if the server offers
799 JH/04 Certificate name checking on server certificates, when exim is a client,
800 is now done by default. The transport option tls_verify_cert_hostnames
801 can be used to disable this per-host. The build option
802 EXPERIMENTAL_CERTNAMES is withdrawn.
804 JH/05 The value of the tls_verify_certificates smtp transport and main options
805 default to the word "system" to access the system default CA bundle.
806 For GnuTLS, only version 3.0.20 or later.
808 JH/06 Verification of the server certificate for a TLS connection is now tried
809 (but not required) by default. The verification status is now logged by
810 default, for both outbound TLS and client-certificate supplying inbound
813 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
816 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
817 Status Notification (bounce) messages are now MIME format per RFC 3464.
818 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
819 under the control of the dsn_advertise_hosts option, and routers may
820 have a dsn_lasthop option.
822 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
823 default, modifiable by a malware= option. The list separator for
824 the options can now be changed in the usual way. Bug 68.
826 JH/10 The smtp_receive_timeout main option is now expanded before use.
828 JH/11 The incoming_interface log option now also enables logging of the
829 local interface on delivery outgoing connections.
831 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
832 if the interface and destination host and port all match.
834 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
837 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
838 Patch from Andrew Lewis.
840 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
841 now supports optional time-restrictions, weighting, and priority
842 modifiers per server. Patch originally by <rommer@active.by>.
844 JH/16 The spamd_address main option now supports a mixed list of local
845 and remote servers. Remote servers can be IPv6 addresses, and
846 specify a port-range.
848 JH/17 Bug 68: The spamd_address main option now supports an optional
849 timeout value per server.
851 JH/18 Bug 1581: Router and transport options headers_add/remove can
852 now have the list separator specified.
854 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
857 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
860 JH/21 Support for the A6 type of dns record is withdrawn.
862 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
863 rather than the verbs used.
865 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
866 from 255 to 1024 chars.
868 JH/24 Verification callouts now attempt to use TLS by default.
870 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
871 are generic router options now. The defaults didn't change.
873 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
874 Original patch from Alexander Shikoff, worked over by JH.
876 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
877 files if ZCAT_COMMAND is not executable.
879 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
881 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
883 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
884 Normally benign, it bites when the pair was led to by a CNAME;
885 modern usage is to not canonicalize the domain to a CNAME target
886 (and we were inconsistent anyway for A-only vs AAAA+A).
888 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
890 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
891 when evaluating $sender_host_dnssec.
893 JH/31 Check the HELO verification lookup for DNSSEC, adding new
894 $sender_helo_dnssec variable.
896 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
898 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
900 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
902 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
903 documented as working, but never had. Support all but $spam_report.
905 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
906 added for tls authenticator.
908 HS/03 Add perl_taintmode main config option
914 TL/01 When running the test suite, the README says that variables such as
915 no_msglog_check are global and can be placed anywhere in a specific
916 test's script, however it was observed that placement needed to be near
917 the beginning for it to behave that way. Changed the runtest perl
918 script to read through the entire script once to detect and set these
919 variables, reset to the beginning of the script, and then run through
920 the script parsing/test process like normal.
922 TL/02 The BSD's have an arc4random API. One of the functions to induce
923 adding randomness was arc4random_stir(), but it has been removed in
924 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
925 function when detected.
927 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
928 cause callback expansion.
930 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
931 syntax errors in an expansion can be treated as a string instead of
932 logging or causing an error, due to the internal use of bool_lax
933 instead of bool when processing it.
935 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
936 server certificates when making smtp deliveries.
938 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
940 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
942 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
944 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
945 Merged patch from Sebastian Wiedenroth.
947 JH/05 Fix results-pipe from transport process. Several recipients, combined
948 with certificate use, exposed issues where response data items split
949 over buffer boundaries were not parsed properly. This eventually
950 resulted in duplicates being sent. This issue only became common enough
951 to notice due to the introduction of connection certificate information,
952 the item size being so much larger. Found and fixed by Wolfgang Breyha.
954 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
955 size buffer was used, resulting in syntax errors when an expansion
958 JH/07 Add support for directories of certificates when compiled with a GnuTLS
959 version 3.3.6 or later.
961 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
962 is EXPERIMENTAL_EVENT, the main-configuration and transport options
963 both become "event_action", the variables become $event_name, $event_data
964 and $event_defer_errno. There is a new variable $verify_mode, usable in
965 routers, transports and related events. The tls:cert event is now also
966 raised for inbound connections, if the main configuration event_action
969 TL/06 In test suite, disable OCSP for old versions of openssl which contained
970 early OCSP support, but no stapling (appears to be less than 1.0.0).
972 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
973 server certificate names available under the smtp transport option
974 "tls_verify_cert_hostname" now do not permit multi-component wildcard
977 JH/10 Time-related extraction expansions from certificates now use the main
978 option "timezone" setting for output formatting, and are consistent
979 between OpenSSL and GnuTLS compilations. Bug 1541.
981 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
982 encoded parameter in the incoming message. Bug 1558.
984 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
985 include certificate info, eximon was claiming there were spoolfile
988 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
990 JH/14 Log delivery-related information more consistently, using the sequence
991 "H=<name> [<ip>]" wherever possible.
993 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
994 are problematic for Debian distribution, omit them from the release
997 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
999 JH/16 Fix string representation of time values on 64bit time_t architectures.
1002 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1003 output list separator was used.
1008 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1009 checkers that were complaining about end of non-void function with no
1012 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1013 This was a regression introduced in 4.83 by another bugfix.
1015 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1017 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1018 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1024 TF/01 Correctly close the server side of TLS when forking for delivery.
1026 When a message was received over SMTP with TLS, Exim failed to clear up
1027 the incoming connection properly after forking off the child process to
1028 deliver the message. In some situations the subsequent outgoing
1029 delivery connection happened to have the same fd number as the incoming
1030 connection previously had. Exim would try to use TLS and fail, logging
1031 a "Bad file descriptor" error.
1033 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1034 utilities have not been installed.
1036 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1037 temporary space as the ACL may create new global variables.
1039 TL/01 LDAP support uses per connection or global context settings, depending
1040 upon the detected version of the libraries at build time.
1042 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1043 to extract and use the src ip:port in logging and expansions as if it
1044 were a direct connection from the outside internet. PPv2 support was
1045 updated based on HAProxy spec change in May 2014.
1047 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1049 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1050 Properly escape header and check for NULL return.
1052 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1055 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1057 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1058 characters in header names, implemented as a verify condition.
1059 Contributed by Michael Fischer v. Mollard.
1061 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1062 results permerror and temperror. Previous values are deprecated but
1063 still accepted. In a future release, err_perm and err_temp will be
1064 completely removed, which will be a backward incompatibility if the
1065 ACL tests for either of these two old results. Patch contributed by
1066 user bes-internal on the mailing list.
1068 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1070 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1071 selectors, in both main and reject logs.
1073 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1076 JH/07 Add malware type "sock" for talking to simple daemon.
1078 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1080 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1081 routers/transports under cutthrough routing.
1083 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1084 numbers. Touch up "bool" conditional to keep the same definition.
1086 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1088 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1090 JH/12 Expand items in router/transport headers_add or headers_remove lists
1091 individually rather than the list as a whole. Bug 1452.
1093 Required for reasonable handling of multiple headers_ options when
1094 they may be empty; requires that headers_remove items with embedded
1095 colons must have them doubled (or the list-separator changed).
1097 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1098 view the policy declared in the DMARC record. Currently, $dmarc_status
1099 is a combined value of both the record presence and the result of the
1102 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1104 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1105 dnslookup router and the smtp transport (applying to the forward
1108 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1109 of ldap servers used for a specific lookup. Patch provided by Heiko
1112 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1113 New variable $lookup_dnssec_authenticated for observability.
1115 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1116 Patch submitted by Lars Timman.
1118 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1120 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1121 Requires trusted mode and valid format message id, aborts otherwise.
1122 Patch contributed by Heiko Schlichting.
1124 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1125 certextract with support for various fields. Bug 1358.
1127 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1128 is requested by default, modifiable by smtp transport option
1131 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1132 operate on certificate variables to give certificate fingerprints
1133 Also new ${sha256:cert_variable}.
1135 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1137 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1138 Christian Aistleitner.
1140 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1142 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1143 file. Patch from Wolfgang Breyha.
1145 JH/25 Expand the coverage of the delivery $host and $host_address to
1146 client authenticators run in verify callout. Bug 1476.
1148 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1149 align with daemon_smtp_ports. Bug 72.
1151 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1152 support and error reporting did not work properly.
1154 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1155 and is readable. Patch from Andrew Colin Kissa.
1157 TL/14 Enhance documentation of ${run expansion and how it parses the
1158 commandline after expansion, particularly in the case when an
1159 unquoted variable expansion results in an empty value.
1161 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1163 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1164 by renaming away from it. Addresses GH issue 15, reported by
1167 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1169 TL/15 SECURITY: prevent double expansion in math comparison functions
1170 (can expand unsanitized data). Not remotely exploitable.
1177 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1179 PP/02 Make -n do something, by making it not do something.
1180 When combined with -bP, the name of an option is not output.
1182 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1185 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1186 $sender_host_name and config options to manage this, and basic check
1189 PP/05 DSCP support for outbound connections and control modifier for inbound.
1191 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1192 (Only plugin which currently uses this is kerberos4, which nobody should
1193 be using, but we should make it available and other future plugins might
1194 conceivably use it, even though it would break NAT; stuff *should* be
1195 using channel bindings instead).
1197 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1198 name; added for Sendmail compatibility; requires admin caller.
1199 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1200 just ignore it); requires trusted caller.
1201 Also parse but ignore: -Ac -Am -X<logfile>
1204 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1206 TL/02 Add +smtp_confirmation as a default logging option.
1208 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1209 Patch by Magnus Holmgren from 2007-02-20.
1211 TL/04 Bugzilla 1281 - Spec typo.
1212 Bugzilla 1283 - Spec typo.
1213 Bugzilla 1290 - Spec grammar fixes.
1215 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1217 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1219 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1220 mailing list by by Dmitry Isaikin.
1222 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1224 JH/02 Support "G" suffix to numbers in ${if comparisons.
1226 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1228 NM/01 Bugzilla 1197 - Spec typo
1229 Bugzilla 1196 - Spec examples corrections
1231 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1233 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1234 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1237 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1238 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1240 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1241 "acl {{name}{arg}...}", and optional args on acl condition
1244 JH/05 Permit multiple router/transport headers_add/remove lines.
1246 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1248 JH/07 Avoid using a waiting database for a single-message-only transport.
1249 Performance patch from Paul Fisher. Bugzilla 1262.
1251 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1254 JH/09 Add $headers_added variable, with content from use of ACL modifier
1255 add_header (but not yet added to the message). Bugzilla 199.
1257 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1258 Pulled from Bugzilla 817 by Wolfgang Breyha.
1260 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1262 (nb: this is the same fix as in Exim 4.80.1)
1264 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1267 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1270 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1272 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1273 advertise SMTP AUTH mechanism to us, instead of a generic
1274 protocol violation error. Also, make Exim more robust to bad
1275 data from the Dovecot auth socket.
1277 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1279 When a queue runner is handling a message, Exim first routes the
1280 recipient addresses, during which it prunes them based on the retry
1281 hints database. After that it attempts to deliver the message to
1282 any remaining recipients. It then updates the hints database using
1285 So if a recipient address works intermittently, it can get repeatedly
1286 deferred at routing time. The retry hints record remains fresh so the
1287 address never reaches the final cutoff time.
1289 This is a fairly common occurrence when a user is bumping up against
1290 their storage quota. Exim had some logic in its local delivery code
1291 to deal with this. However it did not apply to per-recipient defers
1292 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1294 This change adds a proper retry rule check during routing so that the
1295 final cutoff time is checked against the message's age. We only do
1296 this check if there is an address retry record and there is not a
1297 domain retry record; this implies that previous attempts to handle
1298 the address had the retry_use_local_parts option turned on. We use
1299 this as an approximation for the destination being like a local
1300 delivery, as in LMTP.
1302 I suspect this new check makes the old local delivery cutoff check
1303 redundant, but I have not verified this so I left the code in place.
1305 TF/02 Correct gecos expansion when From: is a prefix of the username.
1307 Test 0254 submits a message to Exim with the header
1311 When I ran the test suite under the user fanf2, Exim expanded
1312 the header to contain my full name, whereas it should have added
1313 a Resent-Sender: header. It erroneously treats any prefix of the
1314 username as equal to the username.
1316 This change corrects that bug.
1318 GF/01 DCC debug and logging tidyup
1319 Error conditions log to paniclog rather than rejectlog.
1320 Debug lines prefixed by "DCC: " to remove any ambiguity.
1322 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1324 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1325 Bug spotted by Jeremy Harris; was flawed since initial commit.
1326 Would have resulted in OCSP responses post-SNI triggering an Exim
1327 NULL dereference and crash.
1329 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1331 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1332 Bug detection, analysis and fix by Samuel Thibault.
1333 Bugzilla 1331, Debian bug #698092.
1335 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1337 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1338 Server implementation by Todd Lyons, client by JH.
1339 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1340 config variable "prdr_enable" controls whether the server
1341 advertises the facility. If the client requests PRDR a new
1342 acl_data_smtp_prdr ACL is called once for each recipient, after
1343 the body content is received and before the acl_smtp_data ACL.
1344 The client is controlled by both of: a hosts_try_prdr option
1345 on the smtp transport, and the server advertisement.
1346 Default client logging of deliveries and rejections involving
1347 PRDR are flagged with the string "PRDR".
1349 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1350 fclose(). Diagnosis by Todd Lyons.
1352 PP/17 Update configure.default to handle IPv6 localhost better.
1353 Patch by Alain Williams (plus minor tweaks).
1356 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1357 This is now consistent with GnuTLS, and is now documented: the
1358 previous undocumented portable approach to treating the option as
1359 unset was to force an expansion failure. That still works, and
1360 an empty string is now equivalent.
1362 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1363 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1364 not performing validation itself.
1366 PP/20 Added force_command boolean option to pipe transport.
1367 Patch from Nick Koston, of cPanel Inc.
1369 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1372 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1374 PP/21 Fix eximon continuous updating with timestamped log-files.
1375 Broken in a format-string cleanup in 4.80, missed when I repaired the
1376 other false fix of the same issue.
1377 Report and fix from Heiko Schlichting.
1380 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1381 Report from Prashanth Katuri.
1383 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1384 It's SecureTransport, so affects any MacOS clients which use the
1385 system-integrated TLS libraries, including email clients.
1387 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1388 using a MIME ACL for non-SMTP local injection.
1389 Report and assistance in diagnosis by Warren Baker.
1391 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1393 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1395 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1396 last id that failed so it may be referenced in subsequent ACL's.
1398 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1401 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1402 ldap library initialization, allowing self-signed CA's to be
1403 used. Also properly sets require_cert option later in code by
1404 using NULL (global ldap config) instead of ldap handle (per
1405 session). Bug diagnosis and testing by alxgomz.
1407 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1408 the src/util/ subdirectory.
1410 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1411 renamed to Transport Post Delivery Action by Jeremy Harris, as
1414 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1415 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1416 redis_servers = needs to be configured which will be used by the redis
1417 lookup. Patch from Warren Baker, of The Packet Hub.
1419 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1421 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1422 hostname or reverse DNS when processing a host list. Used suggestions
1423 from multiple comments on this bug.
1425 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1427 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1428 Missed a few lines, added it to make the runtest require no keyboard
1431 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1432 contains upper case chars. Make router use caseful_local_part.
1434 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1435 support when GnuTLS has been built with p11-kit.
1441 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1443 This, or similar/improved, will also be change PP/11 of 4.82.
1449 PP/01 Handle short writes when writing local log-files.
1450 In practice, only affects FreeBSD (8 onwards).
1451 Bugzilla 1053, with thanks to Dmitry Isaikin.
1453 NM/01 Bugzilla 949 - Documentation tweak
1455 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1458 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1460 PP/02 Implemented gsasl authenticator.
1462 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1464 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1465 `pkg-config foo` for cflags/libs.
1467 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1468 with rest of GSASL and with heimdal_gssapi.
1470 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1471 `pkg-config foo` for cflags/libs for the TLS implementation.
1473 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1474 properties get this fed in as external SSF. A number of robustness
1475 and debugging improvements to the cyrus_sasl authenticator.
1477 PP/08 cyrus_sasl server now expands the server_realm option.
1479 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1480 Patch by Jeremy Harris.
1482 PP/10 Added dbmjz lookup type.
1484 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1486 PP/12 MAIL args handles TAB as well as SP, for better interop with
1487 non-compliant senders.
1488 Analysis and variant patch by Todd Lyons.
1490 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1491 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1492 Patch from Dirk Mueller <dmueller@suse.com>
1494 PP/13 tls_peerdn now print-escaped for spool files.
1495 Observed some $tls_peerdn in wild which contained \n, which resulted
1496 in spool file corruption.
1498 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1499 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1500 or write after TLS renegotiation, which otherwise led to messages
1503 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1504 as a tracking header (ie: a signed header comes before the signature).
1505 Patch from Wolfgang Breyha.
1507 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1508 comma-sep list; embedded commas doubled.
1510 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1512 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1514 Report and patch from Dmitry Banschikov.
1516 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1517 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1518 failures. We appear to now support TLS1.1+ with Exim.
1520 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1521 lets Exim select keys and certificates based upon TLS SNI from client.
1522 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1523 before an outbound SMTP session. New log_selector, +tls_sni.
1525 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1526 NULL dereference. Report and patch from Alun Jones.
1528 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1529 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1530 Not seeing resolver debug output on NetBSD, but suspect this is a
1531 resolver implementation change.
1533 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1534 Left warnings. Added "eximon gdb" invocation mode.
1536 PP/21 Defaulting "accept_8bitmime" to true, not false.
1538 PP/22 Added -bw for inetd wait mode support.
1540 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1541 locate the relevant includes and libraries. Made this the default.
1543 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1544 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1546 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1547 This may cause build issues on older platforms.
1549 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1550 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1551 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1552 Added SNI support via GnuTLS too.
1553 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1555 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1557 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1558 Applied second patch from Janne, implementing suggestion to default
1559 multiple-strings-in-record handling to match SPF spec.
1561 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1563 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1564 read-only, out of scope).
1565 Patch from Wolfgang Breyha, report from Stuart Northfield.
1567 PP/29 Fix three issues highlighted by clang analyser static analysis.
1568 Only crash-plausible issue would require the Cambridge-specific
1569 iplookup router and a misconfiguration.
1570 Report from Marcin Mirosław.
1572 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1574 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1575 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1576 As part of this, removing so much warning spew let me fix some minor
1577 real issues in debug logging.
1579 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1580 assignment on my part. Fixed.
1582 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1583 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1584 Janne Snabb (who went above and beyond: thank you).
1586 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1587 string otherwise requires a connection and a bunch more work and it's
1588 relatively easy to get wrong. Should also expose TLS library linkage
1591 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1592 64-bit ${eval} (JH/03).
1594 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1595 GNU libc to support some of the 64-bit stuff, should not lead to
1596 conflicts. Defined before os.h is pulled in, so if a given platform
1597 needs to override this, it can.
1599 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1600 protection layer was required, which is not implemented.
1601 Bugzilla 1254, patch from Wolfgang Breyha.
1603 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1604 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1605 tls_dhparam take prime identifiers. Also unbreak combination of
1606 OpenSSL+DH_params+TLSSNI.
1608 PP/39 Disable SSLv2 by default in OpenSSL support.
1614 PP/01 Solaris build fix for Oracle's LDAP libraries.
1615 Bugzilla 1109, patch from Stephen Usher.
1617 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1619 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1622 TF/02 Fix a couple more cases where we did not log the error message
1623 when unlink() failed. See also change 4.74-TF/03.
1625 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1626 lock up or crash if it happened to be inside a call to libc when it
1627 got a SIGUSR1 from exiwhat.
1629 The SIGUSR1 handler appends the current process status to the process
1630 log which is later printed by exiwhat. It used to use the general
1631 purpose logging code to do this, but several functions it calls are
1632 not safe for signals.
1634 The new output code in the SIGUSR1 handler is specific to the process
1635 log, and simple enough that it's easy to inspect for signal safety.
1636 Removing some special cases also simplifies the general logging code.
1637 Removing the spurious timestamps from the process log simplifies
1640 TF/04 Improved ratelimit ACL condition.
1642 The /noupdate option has been deprecated in favour of /readonly which
1643 has clearer semantics. The /leaky, /strict, and /readonly update modes
1644 are mutually exclusive. The update mode is no longer included in the
1645 database key; it just determines when the database is updated. (This
1646 means that when you upgrade Exim will forget old rate measurements.)
1648 Exim now checks that the per_* options are used with an update mode that
1649 makes sense for the current ACL. For example, when Exim is processing a
1650 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1651 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1652 must specify per_mail/readonly. If you omit the update mode it defaults to
1653 /leaky where that makes sense (as before) or /readonly where required.
1655 The /noupdate option is now undocumented but still supported for
1656 backwards compatibility. It is equivalent to /readonly except that in
1657 ACLs where /readonly is required you may specify /leaky/noupdate or
1658 /strict/noupdate which are treated the same as /readonly.
1660 A useful new feature is the /count= option. This is a generalization
1661 of the per_byte option, so that you can measure the throughput of other
1662 aggregate values. For example, the per_byte option is now equivalent
1663 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1665 The per_rcpt option has been generalized using the /count= mechanism
1666 (though it's more complicated than the per_byte equivalence). When it is
1667 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1668 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1669 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1670 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1671 Note that using per_rcpt with a non-readonly update mode in more than
1672 one ACL will cause the recipients to be double-counted. (The per_mail
1673 and per_byte options don't have this problem.)
1675 The handling of very low rates has changed slightly. If the computed rate
1676 is less than the event's count (usually one) then this event is the first
1677 after a long gap. In this case the rate is set to the same as this event's
1678 count, so that the first message of a spam run is counted properly.
1680 The major new feature is a mechanism for counting the rate of unique
1681 events. The new per_addr option counts the number of different
1682 recipients that someone has sent messages to in the last time period. It
1683 behaves like per_rcpt if all the recipient addresses are different, but
1684 duplicate recipient addresses do not increase the measured rate. Like
1685 the /count= option this is a general mechanism, so the per_addr option
1686 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1687 example, measure the rate that a client uses different sender addresses
1688 with the options per_mail/unique=$sender_address. There are further
1689 details in the main documentation.
1691 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1693 TF/06 Removed a few PCRE remnants.
1695 TF/07 Automatically extract Exim's version number from tags in the git
1696 repository when doing development or release builds.
1698 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1699 Bugzilla 879. Patch from Paul Fisher.
1701 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1702 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1705 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1707 PP/05 Variable $av_failed, true if the AV scanner deferred.
1708 Bugzilla 1078. Patch from John Horne.
1710 PP/06 Stop make process more reliably on build failure.
1711 Bugzilla 1087. Patch from Heiko Schlittermann.
1713 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1714 Bugzilla 1089. Patch from Heiko Schlittermann.
1716 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1717 Bugzilla 1131. Patch from Holger Weiß.
1719 PP/09 Handle IPv6 addresses with SPF.
1720 Bugzilla 860. Patch from Wolfgang Breyha.
1722 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1724 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1727 PP/11 match_* no longer expand right-hand-side by default.
1728 New compile-time build option, EXPAND_LISTMATCH_RHS.
1729 New expansion conditions, "inlist", "inlisti".
1731 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1733 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1735 PP/14 fix log_write() format string regression from TF/03.
1736 Bugzilla 1152. Patch from Dmitry Isaikin.
1742 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1744 PP/02 Harmonised TLS library version reporting; only show if debugging.
1745 Layout now matches that introduced for other libraries in 4.74 PP/03.
1747 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1749 PP/04 New "dns_use_edns0" global option.
1751 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1754 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1755 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1757 TK/01 Updated PolarSSL code to 0.14.2.
1758 Bugzilla 1097. Patch from Andreas Metzler.
1760 PP/07 Catch divide-by-zero in ${eval:...}.
1761 Fixes bugzilla 1102.
1763 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1766 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1767 format-string attack -- SECURITY: remote arbitrary code execution.
1769 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1770 time unintentionally subject to list matching rules, letting the header
1771 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1772 arbitrary string expansion). This allowed for information disclosure.
1774 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1775 INT_MIN/-1 -- value coerced to INT_MAX.
1781 NM/01 Workaround for PCRE version dependency in version reporting
1784 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1785 This fixes portability to compilers other than gcc, notably
1786 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1788 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1789 makefiles for portability to HP-UX and POSIX correctness.
1791 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1792 Also via indented variable definition in the Makefile.
1793 (Debugging by Oliver Heesakkers).
1795 PP/02 Restore caching of spamd results with expanded spamd_address.
1796 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1798 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1799 Improves build reliability. Fix from: Frank Elsner
1801 NM/02 Fix wide character breakage in the rfc2047 coding
1802 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1804 NM/03 Allow underscore in dnslist lookups
1805 Fixes bug 1026. Patch from Graeme Fowler
1807 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1808 Code patches from Adam Ciarcinski of NetBSD.
1810 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1813 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1814 is logged, to avoid truncation. Patch from John Horne.
1816 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1817 Patch from Jakob Hirsch.
1819 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1820 SQL string expansion failure details.
1821 Patch from Andrey Oktyabrski.
1823 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1824 Patch from Simon Arlott.
1826 PP/09 New lookups functionality failed to compile on old gcc which rejects
1827 extern declarations in function scope.
1828 Patch from Oliver Fleischmann
1830 PP/10 Use sig_atomic_t for flags set from signal handlers.
1831 Check getgroups() return and improve debugging.
1832 Fixed developed for diagnosis in bug 927 (which turned out to be
1835 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1836 Patch from Mark Zealey.
1838 PP/12 Bugzilla 1056: Improved spamd server selection.
1839 Patch from Mark Zealey.
1841 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1842 Based on patch from Heiko Schlittermann.
1844 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1845 Patch from Uwe Doering, sign-off by Michael Haardt.
1847 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1848 variable declaration deep within a block. Bug and patch from
1851 PP/15 lookups-Makefile IRIX compatibility coercion.
1853 PP/16 Make DISABLE_DKIM build knob functional.
1855 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1856 Patch by Simon Arlott
1858 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1859 variable argument macros. Our copy now differs from upstream.
1865 TF/01 Failure to get a lock on a hints database can have serious
1866 consequences so log it to the panic log.
1868 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1869 controlled using the smtp_confirmation log selector.
1871 TF/03 Include the error message when we fail to unlink a spool file.
1873 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1874 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1875 for maintaining out-of-tree patches for some time.
1877 PP/01 Bugzilla 139: Documentation and portability issues.
1878 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1879 Handle per-OS dynamic-module compilation flags.
1881 PP/02 Let /dev/null have normal permissions.
1882 The 4.73 fixes were a little too stringent and complained about the
1883 permissions on /dev/null. Exempt it from some checks.
1884 Reported by Andreas M. Kirchwitz.
1886 PP/03 Report version information for many libraries, including
1887 Exim version information for dynamically loaded libraries. Created
1888 version.h, now support a version extension string for distributors
1889 who patch heavily. Dynamic module ABI change.
1891 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1892 privilege escalation vulnerability whereby the Exim run-time user
1893 can cause root to append content of the attacker's choosing to
1896 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1899 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1900 If dropping privileges for untrusted macros, we disabled normal logging
1901 on the basis that it would fail; for the Exim run-time user, this is not
1902 the case, and it resulted in successful deliveries going unlogged.
1903 Fixed. Reported by Andreas Metzler.
1909 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1910 only prepend for the Resent-* case. Fixes regression introduced in
1911 Exim 4.70 by NM/22 for Bugzilla 607.
1913 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1914 increasing numbers of administrators be bitten by this.
1916 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1918 PP/03 Bugzilla 994: added openssl_options main configuration option.
1920 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1922 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1924 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1926 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1927 without a peer certificate, leading to a segfault because of an
1928 assumption that peers always have certificates. Be a little more
1929 paranoid. Problem reported by Martin Tscholak.
1931 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1932 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1933 NB: ClamAV planning to remove STREAM in "middle of 2010".
1934 CL also introduces -bmalware, various -d+acl logging additions and
1935 more caution in buffer sizes.
1937 PP/09 Implemented reverse_ip expansion operator.
1939 PP/10 Bugzilla 937: provide a "debug" ACL control.
1941 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1943 PP/12 Bugzilla 973: Implement --version.
1945 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1947 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1949 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1951 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1952 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1953 ignore trailing whitespace.
1955 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1957 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1960 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1961 Notification from Dr Andrew Aitchison.
1963 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1964 ExtendedDetectionInfo response format.
1965 Notification from John Horne.
1967 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1970 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1971 XSL and documented dependency on system catalogs, with examples of how
1974 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1977 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1978 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1979 configuration file which is writeable by the Exim user or group.
1981 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1982 of configuration files to cover files specified with the -C option if
1983 they are going to be used with root privileges, not just the default
1986 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1987 option (effectively making it always true).
1989 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1990 files to be used while preserving root privileges.
1992 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1993 that rogue child processes cannot use them.
1995 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1996 run-time user, instead of root.
1998 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1999 Exim run-time user without dropping privileges.
2001 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2002 result string, instead of calling string_vformat() twice with the same
2005 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2006 for other users. Others should always drop root privileges if they use
2007 -C on the command line, even for a whitelisted configure file.
2009 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2011 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2017 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2018 $data_path, and $header_path variables; fixed documentation bugs and
2021 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2022 exipick to access non-standard spools, including the "frozen" queue
2025 NM/01 Bugzilla 965: Support mysql stored procedures.
2026 Patch from Alain Williams
2028 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2030 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2031 Patch from Andreas Metzler
2033 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2034 Patch from Kirill Miazine
2036 NM/05 Bugzilla 671: Added umask to procmail example.
2038 JJ/03 installed exipick 20100323.0, fixing doc bug
2040 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2041 directory. Notification and patch from Dan Rosenberg.
2043 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2045 TK/02 Improve log output when DKIM signing operation fails.
2047 MH/01 Treat the transport option dkim_domain as a colon separated
2048 list, not as a single string, and sign the message with each element,
2049 omitting multiple occurences of the same signer.
2051 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2052 Bugzilla 985, 986. Patch by Simon Arlott
2054 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2055 Patch by Simon Arlott
2057 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2058 MBX locking. Notification from Dan Rosenberg.
2064 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2066 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2068 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2070 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2072 NM/04 Bugzilla 915: Flag broken perl installation during build.
2078 TK/01 Added patch by Johannes Berg that expands the main option
2079 "spamd_address" if it starts with a dollar sign.
2081 TK/02 Write list of recipients to X-Envelope-Sender header when building
2082 the mbox-format spool file for content scanning (suggested by Jakob
2085 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2086 (http://www.dcc-servers.net/) support via dccifd. Activated by
2087 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2089 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2090 by Mark Daniel Reidel <mr@df.eu>.
2092 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2093 When building exim an external PCRE library is now needed -
2094 PCRE is a system library on the majority of modern systems.
2095 See entry on PCRE_LIBS in EDITME file.
2097 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2098 conversation. Added nologin parameter to request.
2099 Patch contributed by Kirill Miazine.
2101 TF/01 Do not log submission mode rewrites if they do not change the address.
2103 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2105 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2106 log files in place. Contributed by Roberto Lima.
2108 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2110 TF/03 Bugzilla 615: When checking the local_parts router precondition
2111 after a local_part_suffix or local_part_prefix option, Exim now
2112 does not use the address's named list lookup cache, since this
2113 contains cached lookups for the whole local part.
2115 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2116 Robert Millan. Documentation is in experimental-spec.txt.
2118 TF/04 Bugzilla 668: Fix parallel build (make -j).
2120 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2122 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2123 Patch provided by Jan Srzednicki.
2125 TF/05 Leading white space used to be stripped from $spam_report which
2126 wrecked the formatting. Now it is preserved.
2128 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2129 that they are available at delivery time.
2131 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2133 TF/08 TLS error reporting now respects the incoming_interface and
2134 incoming_port log selectors.
2136 TF/09 Produce a more useful error message if an SMTP transport's hosts
2137 setting expands to an empty string.
2139 NM/06 Bugzilla 744: EXPN did not work under TLS.
2140 Patch provided by Phil Pennock.
2142 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2143 Patch provided by Richard Godbee.
2145 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2146 acl_smtp_notquit, added index entry.
2148 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2149 Patch provided by Eugene Bujak.
2151 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2152 accept(). Patch provided by Maxim Dounin.
2154 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2155 Patch provided by Phil Pennock.
2157 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2159 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2160 Patch provided by Brad "anomie" Jorsch.
2162 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2163 Patch provided by Dean Brooks.
2165 TK/05 Add native DKIM support (does not depend on external libraries).
2167 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2168 Patch provided by Graeme Fowler.
2170 NM/16 Bugzilla 851: Documentation example syntax fix.
2172 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2174 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2177 NM/19 Bugzilla 745: TLS version reporting.
2178 Patch provided by Phil Pennock.
2180 NM/20 Bugzilla 167: bool: condition support.
2181 Patch provided by Phil Pennock.
2183 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2184 clients. Patch provided by Phil Pennock.
2186 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2187 Patch provided by Brad "anomie" Jorsch.
2189 NM/23 Bugzilla 687: Fix misparses in eximstats.
2190 Patch provided by Heiko Schlittermann.
2192 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2193 Patch provided by Heiko Schlittermann.
2195 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2196 plus update to original patch.
2198 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2200 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2201 Patch provided by David Brownlee.
2203 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2205 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2207 NM/30 Bugzilla 888: TLS documentation bugfixes.
2209 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2211 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2212 Unlike the original bugzilla I have changed all shell scripts in src tree.
2214 NM/33 Bugzilla 898: Transport filter timeout fix.
2215 Patch by Todd Rinaldo.
2217 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2218 Patch by Serge Demonchaux.
2220 NM/35 Bugzilla 39: Base64 decode bug fixes.
2221 Patch by Jakob Hirsch.
2223 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2225 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2227 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2229 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2235 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2236 ALT-N's libdkim that I have put here:
2237 http://duncanthrax.net/exim-experimental/
2239 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2240 were called 'true' and it seems that C99 defines that as a reserved
2241 keyword to be used with 'bool' variable types. That means you could
2242 not include C99-style headers which use bools without triggering
2243 build errors in sieve.c.
2245 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2246 as mailq or other aliases. Changed the --help handling significantly
2247 to do whats expected. exim_usage() emits usage/help information.
2249 SC/01 Added the -bylocaldomain option to eximstats.
2251 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2253 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2255 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2261 PH/01 Another patch from the Sieve maintainer.
2263 PH/02 When an IPv6 address is converted to a string for single-key lookup
2264 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2265 dots are used instead of colons so that keys in lsearch files need not
2266 contain colons. This was done some time before quoting was made available
2267 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2268 (notated using the quote facility) so as to distinguish them from IPv4
2269 keys. This meant that lookups for IP addresses in host lists did not work
2270 for iplsearch lookups.
2272 This has been fixed by arranging for IPv6 addresses to be expressed with
2273 colons if the lookup type is iplsearch. This is not incompatible, because
2274 previously such lookups could never work.
2276 The situation is now rather anomalous, since one *can* have colons in
2277 ordinary lsearch keys. However, making the change in all cases is
2278 incompatible and would probably break a number of configurations.
2280 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2283 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2284 conversion specification without a maximum field width, thereby enabling
2285 a rogue spamd server to cause a buffer overflow. While nobody in their
2286 right mind would setup Exim to query an untrusted spamd server, an
2287 attacker that gains access to a server running spamd could potentially
2288 exploit this vulnerability to run arbitrary code as the Exim user.
2290 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2291 $primary_hostname instead of what libspf2 thinks the hosts name is.
2293 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2294 a directory entry by the name of the lookup key. Previously, if a
2295 symlink pointed to a non-existing file or a file in a directory that
2296 Exim lacked permissions to read, a lookup for a key matching that
2297 symlink would fail. Now it is enough that a matching directory entry
2298 exists, symlink or not. (Bugzilla 503.)
2300 PH/03 The body_linecount and body_zerocount variables are now exported in the
2303 PH/04 Added the $dnslist_matched variable.
2305 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2306 This means they are set thereafter only if the connection becomes
2309 PH/06 Added the client_condition to authenticators so that some can be skipped
2310 by clients under certain conditions.
2312 PH/07 The error message for a badly-placed control=no_multiline_responses left
2313 "_responses" off the end of the name.
2315 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2317 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2318 (without spaces) instead of just copying the configuration text.
2320 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2322 PH/11 Added $max_received_linelength.
2324 PH/12 Added +ignore_defer and +include_defer to host lists.
2326 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2327 way in which PCRE > 7.0 is built.
2329 PH/14 Implemented queue_only_load_latch.
2331 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2332 MAIL command. The effect was to mangle the value on 64-bit systems.
2334 PH/16 Another patch from the Sieve maintainer.
2336 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2338 PH/18 If a system quota error occurred while trying to create the file for
2339 a maildir delivery, the message "Mailbox is full" was not appended to the
2340 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2341 only to a quota excession during the actual writing of the file.
2343 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2344 characters?) which causes problems in log lines. The DN values are now
2345 passed through string_printing() before being added to log lines.
2347 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2348 and InterBase are left for another time.)
2350 PH/21 Added message_body_newlines option.
2352 PH/22 Guard against possible overflow in moan_check_errorcopy().
2354 PH/23 POSIX allows open() to be a macro; guard against that.
2356 PH/24 If the recipient of an error message contained an @ in the local part
2357 (suitably quoted, of course), incorrect values were put in $domain and
2358 $local_part during the evaluation of errors_copy.
2364 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2365 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2368 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2369 issue a MAIL command.
2371 PH/02 In an ACL statement such as
2373 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2375 if a client was not listed at all, or was listed with a value other than
2376 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2377 the condition was not true (as it should be), so access was not denied.
2378 The bug was that the ! inversion was incorrectly passed on to the second
2379 item. This has been fixed.
2381 PH/03 Added additional dnslists conditions == and =& which are different from
2382 = and & when the dns lookup returns more than one IP address.
2384 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2385 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2387 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2388 FSYNC, which compiles an option called disable_fsync that allows for
2389 bypassing fsync(). The documentation is heavily laced with warnings.
2391 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2393 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2394 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2395 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2396 including adding "make clean"; (3) Added -fPIC when compiling the test
2397 dynamically loaded module, to get rid of a warning.
2399 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2400 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2401 = 0s. The bug is otherwise harmless.
2403 PH/07 There was a bug in the dovecot authenticator such that the value of
2404 $auth1 could be overwritten, and so not correctly preserved, after a
2405 successful authentication. This usually meant that the value preserved by
2406 the server_setid option was incorrect.
2408 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2410 PH/09 Installed PCRE release 7.0.
2412 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2413 run for batched SMTP input. It is now run at the start of every message
2414 in the batch. While fixing this I discovered that the process information
2415 (output by running exiwhat) was not always getting set for -bs and -bS
2416 input. This is fixed, and it now also says "batched" for BSMTP.
2418 PH/11 Added control=no_pipelining.
2420 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2421 patch, slightly modified), and move the expansion of helo_data till after
2422 the connection is made in the smtp transport (so it can use these
2425 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2427 PH/14 Added log_selector = +pid.
2429 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2431 PH/16 Add ${if forany and ${if forall.
2433 PH/17 Added dsn_from option to vary the From: line in DSNs.
2435 PH/18 Flush SMTP output before performing a callout, unless control =
2436 no_callout_flush is set.
2438 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2439 was true (the default) a successful delivery failed to delete the retry
2440 item, thus causing premature timeout of the address. The bug is now
2443 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2445 PH/21 Long custom messages for fakedefer and fakereject are now split up
2446 into multiline responses in the same way that messages for "deny" and
2447 other ACL rejections are.
2449 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2450 with slight modification.
2452 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2453 draft, changing the syntax and factoring some duplicate code".
2455 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2456 for deliveries of the second and subsequent messages over the same SMTP
2459 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2460 ${reduce, with only minor "tidies".
2462 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2464 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2465 expansion side effects.
2467 PH/27 When a message times out after an over-quota error from an Exim-imposed
2468 quota, the bounce message says "mailbox is full". This message was not
2469 being given when it was a system quota that was exceeded. It now should
2472 MH/03 Made $recipients available in local_scan(). local_scan() already has
2473 better access to the recipient list through recipients_list[], but
2474 $recipients can be useful in postmaster-provided expansion strings.
2476 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2477 in the case of a MAIL command with additional options following the
2478 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2479 were accidentally chopped off.
2481 PH/29 SMTP synchronization checks are implemented when a command is read -
2482 there is a check that no more input is waiting when there shouldn't be
2483 any. However, for some commands, a delay in an ACL can mean that it is
2484 some time before the response is written. In this time, more input might
2485 arrive, invalidly. So now there are extra checks after an ACL has run for
2486 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2487 pipelining has not been advertised.
2489 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2491 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2492 returned in the SMTP response when smtp_return_error_details was false.
2493 This has been fixed.
2495 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2496 instead of the C I/O that was originally supplied, because problems were
2497 reported on Solaris.
2499 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2500 Exim which did not show up earlier: it was assuming that a call to
2501 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2502 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2503 was a macro that became an assignment, so it seemed to work. This has
2504 changed to a proper function call with a void return, hence the compile
2505 error. Exim's code has been fixed.
2507 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2510 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2512 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2514 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2515 because it turns out to be too big or there is a timeout) there is no
2516 "Completed" line in the log. When some messages of this type were
2517 selected by exigrep, they were listed as "not completed". Others were
2518 picked up by some special patterns. I have improved the selection
2519 criteria to be more general.
2521 PH/38 The host_find_failed option in the manualroute router can now be set
2522 to "ignore", to completely ignore a host whose IP address cannot be
2523 found. If all hosts are ignored, the behaviour is controlled by the new
2524 host_all_ignored option.
2526 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2527 homing or because of multiple MX records with /mx) generated more than
2528 one IP address, and the following item turned out to be the local host,
2529 all the secondary addresses of the first item were incorrectly removed
2530 from the list, along with the local host and any following hosts (which
2531 is what is supposed to happen).
2533 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2534 whoever called Exim into the -H file. In the case of the daemon it was
2535 behaving confusingly. When first started, it used values for whoever
2536 started the daemon, but after a SIGHUP it used the Exim user (because it
2537 calls itself on a restart). I have changed the code so that it now always
2540 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2541 message are rejected with the same error (e.g. no authentication or bad
2542 sender address), and a DATA command is nevertheless sent (as can happen
2543 with PIPELINING or a stupid MUA), the error message that was given to the
2544 RCPT commands is included in the rejection of the DATA command. This is
2545 intended to be helpful for MUAs that show only the final error to their
2548 PH/42 Another patch from the Sieve maintainer.
2550 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2551 Eximstats - Fixed some broken HTML links and added missing column headers
2553 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2554 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2556 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2558 PH/43 Yet another patch from the Sieve maintainer.
2560 PH/44 I found a way to check for a TCP/IP connection going away before sending
2561 the response to the final '.' that terminates a message, but only in the
2562 case where the client has not sent further data following the '.'
2563 (unfortunately, this is allowed). However, in many cases there won't be
2564 any further data because there won't be any more messages to send. A call
2565 to select() can be used: if it shows that the input is "ready", there is
2566 either input waiting, or the socket has been closed. An attempt to read
2567 the next input character can distinguish the two cases. Previously, Exim
2568 would have sent an OK response which the client would never have see.
2569 This could lead to message repetition. This fix should cure that, at
2570 least in a lot of common cases.
2572 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2573 advertised in response to EHLO.
2579 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2580 fixed by 4.65/MH/01 (is this a record?) are fixed:
2582 (i) An empty string was always treated as zero by the numeric comparison
2583 operators. This behaviour has been restored.
2585 (ii) It is documented that the numeric comparison operators always treat
2586 their arguments as decimal numbers. This was broken in that numbers
2587 starting with 0 were being interpreted as octal.
2589 While fixing these problems I realized that there was another issue that
2590 hadn't been noticed. Values of message_size_limit (both the global option
2591 and the transport option) were treated as octal if they started with 0.
2592 The documentation was vague. These values are now always treated as
2593 decimal, and I will make that clear in the documentation.
2599 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2600 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2603 MH/01 Don't check that the operands of numeric comparison operators are
2604 integers when their expansion is in "skipping" mode (fixes bug
2605 introduced by 4.64-PH/07).
2607 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2608 child addresses, Exim now panics and dies. Previously, because the count
2609 is held in a short int, deliveries were likely to be lost. As such a
2610 large number of recipients for a single message is ridiculous
2611 (performance will be very, very poor), I have chosen to impose a limit
2612 rather than extend the field.
2618 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2619 leftover -K file (the existence of which was triggered by #402).
2620 While we were at it, introduced process PID as part of the -K
2621 filename. This should rule out race conditions when creating
2624 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2625 processing considerably. Previous code took too long for large mails,
2626 triggering a timeout which in turn triggers #401.
2628 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2629 in the DK code in transports.c. sendfile() is not really portable,
2630 hence the _LINUX specificness.
2632 TF/01 In the add_headers option to the mail command in an Exim filter,
2633 there was a bug that Exim would claim a syntax error in any
2634 header after the first one which had an odd number of characters
2637 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2638 callout verification, Exim cached a "reject" for the entire domain. This
2639 is correct for most verifications, but it is not correct for a recipient
2640 verification with use_sender or use_postmaster set, because in that case
2641 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2642 case of MAIL FROM:<> rejection from other early rejections (e.g.
2643 rejection of HELO). When verifying a recipient using a non-null MAIL
2644 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2645 Whatever the result of the callout, the value of the domain cache is
2646 left unchanged (for any other kind of callout, getting as far as trying
2647 RCPT means that the domain itself is ok).
2649 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2652 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2653 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2654 session, the daemon ignores ECONNECT errors and logs others; it now
2655 ignores EPIPE as well.
2657 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2658 (quoted-printable decoding).
2660 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2661 later the small subsequent patch to fix an introduced bug.
2663 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2665 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2667 PH/08 An error is now given if message_size_limit is specified negative.
2669 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2670 to be given (somewhat) arbitrary names.
2672 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2675 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2676 miscellaneous code fixes
2678 PH/10 Added the log_reject_target ACL modifier to specify where to log
2681 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2682 hostname. This is wrong, because it relates to the incoming message (and
2683 probably the interface on which it is arriving) and not to the outgoing
2684 callout (which could be using a different interface). This has been
2685 changed to use the value of the helo_data option from the smtp transport
2686 instead - this is what is used when a message is actually being sent. If
2687 there is no remote transport (possible with a router that sets up host
2688 addresses), $smtp_active_hostname is used.
2690 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2691 tweaks were necessary in order to get it to work (see also 21 below):
2692 (a) The code assumed that strncpy() returns a negative number on buffer
2693 overflow, which isn't the case. Replaced with Exim's string_format()
2695 (b) There were several signed/unsigned issues. I just did the minimum
2696 hacking in of casts. There is scope for a larger refactoring.
2697 (c) The code used strcasecmp() which is not a standard C function.
2698 Replaced with Exim's strcmpic() function.
2699 (d) The code set only $1; it now sets $auth1 as well.
2700 (e) A simple test gave the error "authentication client didn't specify
2701 service in request". It would seem that Dovecot has changed its
2702 interface. Fortunately there's a specification; I followed it and
2703 changed what the client sends and it appears to be working now.
2705 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2708 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2709 address A is aliased to B and C, where B exists and C does not. Without
2710 -v the output is "A verified" because verification stops after a
2711 successful redirection if more than one address is generated. However,
2712 with -v the child addresses are also verified. Exim was outputting "A
2713 failed to verify" and then showing the successful verification for C,
2714 with its parentage. It now outputs "B failed to verify", showing B's
2715 parentage before showing the successful verification of C.
2717 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2718 look up a TXT record in a specific list after matching in a combined
2721 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2722 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2723 they consult the DNS. I had assumed they would set it the way they
2724 wanted; and indeed my experiments on Linux seem to show that in some
2725 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2726 To be on the safe side, however, I have now made the interface to
2727 host_find_byname() similar to host_find_bydns(), with an argument
2728 containing the DNS resolver options. The host_find_byname() function now
2729 sets these options at its start, just as host_find_bydns() does. The smtp
2730 transport options dns_qualify_single and dns_search_parents are passed to
2731 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2732 of host_find_byname() use the default settings of RES_DEFNAMES
2733 (qualify_single) but not RES_DNSRCH (search_parents).
2735 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2736 spool_read_header() do less string testing, by means of a preliminary
2737 switch on the second character of optional "-foo" lines. (This is
2738 overdue, caused by the large number of possibilities that now exist.
2739 Originally there were few.) While I was there, I also converted the
2740 str(n)cmp tests so they don't re-test the leading "-" and the first
2741 character, in the hope this might squeeze out yet more improvement.
2743 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2744 flag allowing group syntax was set by the header_syntax check but not
2745 turned off, possible causing trouble later; (2) The flag was not being
2746 set at all for the header_verify test, causing "group"-style headers to
2747 be rejected. I have now set it in this case, and also caused header_
2748 verify to ignore an empty address taken from a group. While doing this, I
2749 came across some other cases where the code for allowing group syntax
2750 while scanning a header line wasn't quite right (mostly, not resetting
2751 the flag correctly in the right place). These bugs could have caused
2752 trouble for malformed header lines. I hope it is now all correct.
2754 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2755 with the "reply" argument non-NULL. The code, however (which originally
2756 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2757 but it didn't always do it. This confused somebody who was copying the
2758 code for some other use. I have removed all the tests.
2760 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2761 feature that was used to support insecure browsers during the U.S. crypto
2762 embargo. It requires special client support, and Exim is probably the
2763 only MTA that supported it -- and would never use it because real RSA is
2764 always available. This code has been removed, because it had the bad
2765 effect of slowing Exim down by computing (never used) parameters for the
2766 RSA_EXPORT functionality.
2768 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2769 authenticator to fail if there's a tab character in the incoming data
2770 (there should never be unless someone is messing about, as it's supposed
2771 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2772 if the connection is using TLS or if the remote IP is the same as the
2773 local IP, and the "valid-client-cert option" if a client certificate has
2776 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2777 authenticators. This can be used for authorization after authentication
2778 succeeds. (In the case of plaintext, it servers for both authentication
2781 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2782 if any retry times were supplied.
2784 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2785 connection, where there is no client IP address to check. In this
2786 situation, the verify now always succeeds.
2788 PH/25 Applied John Jetmore's -Mset patch.
2790 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2792 PH/27 In a string expansion for a processed (not raw) header when multiple
2793 headers of the same name were present, leading whitespace was being
2794 removed from all of them, but trailing whitespace was being removed only
2795 from the last one. Now trailing whitespace is removed from each header
2796 before concatenation. Completely empty headers in a concatenation (as
2797 before) are ignored.
2799 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2800 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2802 PH/29 [Removed. This was a change that I later backed out, and forgot to
2803 correct the ChangeLog entry (that I had efficiently created) before
2804 committing the later change.]
2806 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2807 address errors (4xx response to RCPT) over the same connection as other
2808 messages routed to the same hosts. Such deliveries are always "forced",
2809 so retry times are not inspected. This resulted in far too many retries
2810 for the affected addresses. The effect occurred only when there were more
2811 hosts than the hosts_max_try setting in the smtp transport when it had
2812 the 4xx errors. Those hosts that it had tried were not added to the list
2813 of hosts for which the message was waiting, so if all were tried, there
2814 was no problem. Two fixes have been applied:
2816 (i) If there are any address or message errors in an SMTP delivery, none
2817 of the hosts (tried or untried) are now added to the list of hosts
2818 for which the message is waiting, so the message should not be a
2819 candidate for sending over the same connection that was used for a
2820 successful delivery of some other message. This seems entirely
2821 reasonable: after all the message is NOT "waiting for some host".
2822 This is so "obvious" that I'm not sure why it wasn't done
2823 previously. Hope I haven't missed anything, but it can't do any
2824 harm, as the worst effect is to miss an optimization.
2826 (ii) If, despite (i), such a delivery is accidentally attempted, the
2827 routing retry time is respected, so at least it doesn't keep
2828 hammering the server.
2830 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2831 in ${readsocket because some servers need this prod.
2833 PH/32 Added some extra debug output when updating a wait-xxx database.
2835 PH/33 The hint "could be header name not terminated by colon", which has been
2836 given for certain expansion errors for a long time, was not being given
2837 for the ${if def:h_colon_omitted{... case.
2839 PH/34 The spec says: "With one important exception, whenever a domain list is
2840 being scanned, $domain contains the subject domain." There was at least
2841 one case where this was not true.
2843 PH/35 The error "getsockname() failed: connection reset by peer" was being
2844 written to the panic log as well as the main log, but it isn't really
2845 panic-worthy as it just means the connection died rather early on. I have
2846 removed the panic log writing for the ECONNRESET error when getsockname()
2849 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2850 runs only) independently of the message's sender address. This meant
2851 that, if the 4xx error was in fact related to the sender, a different
2852 message to the same recipient with a different sender could confuse
2853 things. In particular, this can happen when sending to a greylisting
2854 server, but other circumstances could also provoke similar problems.
2855 I have changed the default so that the retry time for these errors is now
2856 based a combination of the sender and recipient addresses. This change
2857 can be overridden by setting address_retry_include_sender=false in the
2860 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2861 remote server are returned as part of bounce messages. This was not
2862 happening for LMTP over a pipe (the lmtp transport), but now it is the
2863 same for both kinds of LMTP.
2865 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2866 in header lines that were in fact CNAMEs. This is no longer the case.
2868 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2869 and queue runs started by the daemon processed all messages. This has
2870 been fixed so that -R and -S can now usefully be given with -q<time>.
2872 PH/40 Import PCRE release 6.7 (fixes some bugs).
2874 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2876 PH/42 Give an error if -q is specified more than once.
2878 PH/43 Renamed the variables $interface_address and $interface_port as
2879 $received_ip_address and $received_port, to make it clear that these
2880 values apply to message reception, and not to the outgoing interface when
2881 a message is delivered. (The old names remain recognized, of course.)
2883 PH/44 There was no timeout on the connect() call when using a Unix domain
2884 socket in the ${readsocket expansion. There now is.
2886 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2887 be meaningful with "accept".
2889 SC/01 Eximstats V1.43
2890 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2892 SC/02 Eximstats V1.44
2893 Use a glob alias rather than an array ref in the generated
2894 parser. This improves both readability and performance.
2896 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2897 Collect SpamAssassin and rejection statistics.
2898 Don't display local sender or destination tables unless
2899 there is data to show.
2900 Added average volumes into the top table text output.
2902 SC/04 Eximstats V1.46
2903 Collect data on the number of addresses (recipients)
2904 as well as the number of messages.
2906 SC/05 Eximstats V1.47
2907 Added 'Message too big' to the list of mail rejection
2908 reasons (thanks to Marco Gaiarin).
2910 SC/06 Eximstats V1.48
2911 Mainlog lines which have GMT offsets and are too short to
2912 have a flag are now skipped.
2914 SC/07 Eximstats V1.49 (Alain Williams)
2915 Added the -emptyok flag.
2917 SC/08 Eximstats V1.50
2918 Fixes for obtaining the IP address from reject messages.
2920 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2921 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2922 whitespace changes from 4.64-PH/27
2924 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2927 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2928 are found, allow negative numbers in numeric criteria)
2930 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2932 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2933 to match changes made in 4.64-PH/43
2935 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2937 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2938 "domain" is actually an IP address, and force a failure. This locks out
2939 those revolvers/nameservers that support "A-for-A" lookups, in
2940 contravention of the specifications.
2942 PH/48 When a host name was looked up from an IP address, and the subsequent
2943 forward lookup of the name timed out, the host name was left in
2944 $sender_host_name, contrary to the specification.
2946 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2947 restricted to single-key lookups, Exim was not diagnosing an error if
2948 * or *@ was used with a query-style lookup.
2950 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2952 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2953 long ago, but noone interested enough thought of it. Let's just say that
2954 the "1.1" means that there are some new functions that weren't there at
2955 some point in the past.
2957 PH/51 Error processing for expansion failure of helo_data from an smtp
2958 transport during callout processing was broken.
2960 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2961 tested/used via the -bh/-bhc/-bs options.
2963 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2964 bug, fixed in subsequent PCRE releases).
2966 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2967 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2969 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2975 SC/01 Use a glob alias rather than an array ref in eximstats generated
2976 parser. This improves both readability and performance.
2978 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2979 Don't display local sender or destination tables in eximstats unless
2980 there is data to show.
2981 Added average volumes into the eximstats top table text output.
2983 SC/03 Collect data on the number of addresses (recipients) as well
2984 as the number of messages in eximstats.
2986 TF/01 Correct an error in the documentation for the redirect router. Exim
2987 does (usually) call initgroups() when daemonizing.
2989 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2990 with consistent privilege compared to when running as a daemon.
2992 TF/03 Note in the spec that $authenticated_id is not set for local
2993 submissions from trusted users.
2995 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2996 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2998 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2999 by adding some example configuration directives to the default
3000 configuration file. A little bit of work is required to uncomment the
3001 directives and define how usernames and passwords are checked, but
3002 there is now a framework to start from.
3004 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3005 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3006 without this. I don't know how relevant this is to other LDAP libraries.
3008 PH/02 Add the verb name to the "unknown ACL verb" error.
3010 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3012 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3014 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3015 directory not expanded when it should be if an expanded home directory
3016 was set for the address (which is overridden by the transport).
3018 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3021 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3022 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3023 because it is too late at that time, and has no effect.
3025 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3026 security issue with \' (bugzilla #107). I could not use the
3027 PQescapeStringConn() function, because it needs a PGconn value as one of
3030 PH/08 When testing addresses using -bt, indicate those final addresses that
3031 are duplicates that would not cause an additional delivery. At least one
3032 person was confused, thinking that -bt output corresponded to deliveries.
3033 (Suppressing duplicates isn't a good idea as you lose the information
3034 about possibly different redirections that led to the duplicates.)
3036 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3037 systems where poll() doesn't work, in particular OS X.
3039 PH/10 Added more information to debugging output for retry time not reached.
3041 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3042 operations in malware.c.
3044 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3047 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3048 syslog_duplication set false, log lines that would normally be written
3049 both the the main log and to the reject log were not written to syslog at
3052 PH/14 In the default configuration, change the use of "message" in ACL warn
3053 statements to "add_header".
3055 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3056 not followed by a command (e.g. "seen endif").
3058 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3059 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3062 PH/17 Added extra conditions to the default value of delay_warning_condition
3066 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3067 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3068 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3071 The Auto-Submitted: and various List- headers are standardised, whereas I
3072 don't think Precedence: ever was.
3074 PH/18 Refactored debugging code in route_finduser() to show more information,
3075 in particular, the error code if getpwnam() issues one.
3077 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3078 This is apparently needed in addition to the PH/07 change above to avoid
3079 any possible encoding problems.
3081 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3082 but not after initializing Perl.
3084 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3085 output them only if debugging. By default they are written stderr,
3086 apparently, which is not desirable.
3088 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3091 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3094 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3096 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3097 authenticated or an ident call has been made. Suppress the default
3098 values for $authenticated_id and $authenticated_sender (but permit -oMai
3099 and -oMas) when testing with -bh.
3101 PH/24 Re-jigged the order of the tests in the default configuration so that the
3102 tests for valid domains and recipients precede the DNS black list and CSA
3103 tests, on the grounds that those ones are more expensive.
3105 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3106 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3107 This bug exists in every version of Exim that I still have, right back to
3110 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3111 However, an attempt to turn on case-sensitivity in a regex key by
3112 including (?-i) didn't work because the subject string was already
3113 lowercased, and the effects were non-intuitive. It turns out that a
3114 one-line patch can be used to allow (?-i) to work as expected.
3120 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3121 other effects) broke the use of negated acl sub-conditions.
3123 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3126 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3127 "Deny" causes Exim to reject the incoming connection with a 554 error.
3128 Unfortunately, if there is a major crisis, such as a disk failure,
3129 tcp-wrappers gives "deny", whereas what one would like would be some
3130 kind of temporary error. A kludge has been added to help with this.
3131 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3132 554 error is used if errno is still zero or contains ENOENT (which occurs
3133 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3136 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3138 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3139 errors. Otherwise a message that provokes a temporary error (when other
3140 messages do not) can cause a whole host to time out.
3142 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3143 addresses were routed directly to files or pipes from a redirect router.
3144 File deliveries just didn't batch; pipe deliveries might have suffered
3147 PH/06 A failure to get a lock for a hints database would erroneously always say
3148 "Failed to get write lock", even when it was really a read lock.
3150 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3151 of 0600. This has been changed to use the value of the lockfile_mode
3152 option (which defaults to 0600).
3154 PH/08 Applied small patch from the Sieve maintainer.
3156 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3157 folder from quota calculations, a direct delivery into this folder messed
3158 up the contents of the maildirsize file. This was because the regex was
3159 used only to exclude .Trash (or whatever) when the size of the mailbox
3160 was calculated. There was no check that a delivery was happening into an
3161 excluded directory. This bug has been fixed by ignoring all quota
3162 processing for deliveries into excluded directories.
3164 PH/10 Added the maildirfolder_create_regex option to appendfile.
3170 PH/01 The code for finding all the local interface addresses on a FreeBSD
3171 system running IPv6 was broken. This may well have applied to all BSD
3172 systems, as well as to others that have similar system calls. The broken
3173 code found IPv4 interfaces correctly, but gave incorrect values for the
3174 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3175 that it would not match correctly against @[] and not recognize the IPv6
3178 PH/02 The ipliteral router was not recognizing addresses of the form user@
3179 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3181 PH/03 Added disable_ipv6.
3183 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3184 decoded form, because it is most often used to construct To: headers
3185 lines in autoreplies, and the decoded form may well be syntactically
3186 invalid. However, $reply_address has leading white space removed, and all
3187 newlines turned into spaces so that the autoreply transport does not
3190 PH/05 If group was specified without a user on a router, and no group or user
3191 was specified on a transport, the group from the router was ignored.
3193 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3194 for visible compile-time settings that can be used to change these
3195 numbers, for those that want even more. Backwards compatibility with old
3196 spool files has been maintained. However, going back to a previous Exim
3197 release will lost any variables that are in spool files.
3199 PH/07 Two small changes when running in the test harness: increase delay when
3200 passing a TCP/IP connection to a new process, in case the original
3201 process has to generate a bounce, and remove special handling of
3202 127.0.0.2 (sic), which is no longer necessary.
3204 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3205 be the same on different OS.
3207 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3210 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3211 whether --show-vars was specified or not
3213 JJ/02 exipick: Added support for new ACL variable spool format introduced
3216 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3217 syntactically invalid From: or Reply-to: line, and a filter used this to
3218 generate an autoreply, and therefore failed to obtain an address for the
3219 autoreply, Exim could try to deliver to a non-existent relative file
3220 name, causing unrelated and misleading errors. What now happens is that
3221 it logs this as a hard delivery error, but does not attempt to create a
3224 PH/11 The exinext utility has a -C option for testing purposes, but although
3225 the given file was scanned by exinext itself; it wasn't being passed on
3226 when Exim was called.
3228 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3229 an end-of-file indication when reading a command response.
3231 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3232 compiled. In many other places in Exim, IPv6 addresses are always
3233 recognized, so I have changed this. It also means that IPv4 domain
3234 literals of the form [IPV4:n.n.n.n] are now always recognized.
3236 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3237 used if the router is not running as root, for example, when verifying at
3238 ACL time, or when using -bh. The debugging output from this situation was
3239 non-existent - all you got was a failure to exec. I have made two
3242 (a) Failures to set uid/gid, the current directory, or a process leader
3243 in a subprocess such as that created by queryprogram now generate
3244 suitable debugging output when -d is set.
3246 (b) The queryprogram router detects when it is not running as root,
3247 outputs suitable debugging information if -d is set, and then runs
3248 the subprocess without attempting to change uid/gid.
3250 PH/15 Minor change to Makefile for building test_host (undocumented testing
3253 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3254 additional section of a DNS packet that returns MX or SRV records.
3255 Instead, it always explicitly searches for A/AAAA records. This avoids
3256 major problems that occur when a DNS server includes only records of one
3257 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3258 fixed another bug: if SRV records were looked up and the corresponding
3259 address records were *not* found in the additional section, the port
3260 values from the SRV records were lost.
3262 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3263 using the correct key (the original address) when searching the retry
3264 rules in order to find which one to use for generating the retry hint.
3266 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3267 adding the default one. Similarly, if it contains a Reply-To: header, the
3268 errors_reply_to option, if set, is not used.
3270 PH/19 When calculating a retry time, Exim used to measure the "time since
3271 failure" by looking at the "first failed" field in the retry record. Now
3272 it does not use this if it is later than than the arrival time of the
3273 message. Instead it uses the arrival time. This makes for better
3274 behaviour in cases where some deliveries succeed, thus re-setting the
3275 "first failed" field. An example is a quota failure for a huge message
3276 when small messages continue to be delivered. Without this change, the
3277 "time since failure" will always be short, possible causing more frequent
3278 delivery attempts for the huge message than are intended.
3279 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3281 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3282 $1, $2, $3) because the numerical variables can be reset during some
3283 expansion items (e.g. "match"), thereby losing the authentication data.
3285 PH/21 Make -bV show the size of off_t variables so that the test suite can
3286 decide whether to run tests for quotas > 2G.
3288 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3289 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3290 filecount value is greater than 2G or if a quota value is greater than 2G
3291 on a system where the size of off_t is not greater than 4, a panic error
3294 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3295 never match. The debug and -bh output now contains an explicit error
3296 message indicating a malformed IPv4 address or mask.
3298 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3299 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3300 PH/23 above applies.
3302 PH/25 Do not write to syslog when running in the test harness. The only
3303 occasion when this arises is a failure to open the main or panic logs
3304 (for which there is an explicit test).
3306 PH/26 Added the /no_tell option to "control=freeze".
3308 PH/27 If a host name lookup failed very early in a connection, for example, if
3309 the IP address matched host_lookup and the reverse lookup yielded a name
3310 that did not have a forward lookup, an error message of the form "no IP
3311 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3312 could be logged. Now it outputs the IP address instead of "NULL".
3314 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3315 allows the sender and the authenticated sender to be set when
3316 submitting a message from within Exim. Since child_open_exim() is
3317 documented for local_scan(), the new function should be too.
3319 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3320 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3321 results in an empty string is now treated as unset.
3323 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3325 PH/31 Added sender_verify_fail logging option.
3327 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3328 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3329 tidied the source and removed it altogether.
3331 PH/33 When a queue run was abandoned because the load average was too high, a
3332 log line was always written; now it is written only if the queue_run log
3333 selector is set. In addition, the log line for abandonment now contains
3334 information about the queue run such as the pid. This is always present
3335 in "start" and "stop" lines but was omitted from the "abandon" line.
3337 PH/34 Omit spaces between a header name and the colon in the error message that
3338 is given when verify = headers_syntax fails (if there are lots of them,
3339 the message gets confusing).
3341 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3342 names, as there are now some PTR records that contain slashes. This check
3343 is only to protect against broken name servers that fall over on strange
3344 characters, so the fact that it applies to all lookups doesn't matter.
3346 PH/36 Now that the new test suite is complete, we can remove some of the
3347 special code in Exim that was needed for the old test suite. For example,
3348 sorting DNS records because real resolvers return them in an arbitrary
3349 order. The new test suite's fake resolver always returns records in the
3352 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3353 bounces) except when queue_only is set, to avoid logging races between
3354 the different processes.
3356 PH/38 Panic-die if .include specifies a non-absolute path.
3358 PH/39 A tweak to the "H" retry rule from its user.
3360 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3361 a label. They prevented compilation on older perls.
3363 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3364 a warning to be raised on newish perls.
3366 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3367 on queue. Changes to match documented behaviour of showing count of
3368 messages matching specified criteria.
3370 PH/40 Changed the default ident timeout from 30s to 5s.
3372 PH/41 Added support for the use of login_cap features, on those BSD systems
3373 that have them, for controlling the resources used by pipe deliveries.
3375 PH/42 The content-scanning code uses fopen() to create files in which to put
3376 message data. Previously it was not paying any attention to the mode of
3377 the files. Exim runs with umask(0) because the rest of the code creates
3378 files with open(), and sets the required mode explicitly. Thus, these
3379 files were ending up world-writeable. This was not a big issue, because,
3380 being within the spool directory, they were not world-accessible. I have
3381 created a function called modefopen, which takes an additional mode
3382 argument. It sets umask(777), creates the file, chmods it to the required
3383 mode, then resets the umask. All the relevant calls to fopen() in the
3384 content scanning code have been changed to use this function.
3386 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3387 to 24 hours. This avoids potential overflow problems when processing G
3388 and H retry rules. I suspect nobody ever tinkers with this value.
3390 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3392 PH/45 When the plaintext authenticator is running as a client, the server's
3393 challenges are checked to ensure they are valid base64 strings. By
3394 default, the authentication attempt is cancelled if an invalid string is
3395 received. Setting client_ignore_invalid_base64 true ignores these errors.
3396 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3397 they are received. Thus, the responses can be made to depend on the
3398 challenges. If an invalid string is ignored, an empty string is placed in
3401 PH/46 Messages that are created by the autoreply transport now contains a
3402 References: header, in accordance with RFCs 2822 and 3834.
3404 PH/47 Added authenticated_sender_force to the smtp transport.
3406 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3408 PH/49 Installed latest patch from the Sieve maintainer.
3410 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3411 was also set, the appendfile transport was unnecessarily scanning a
3412 directory of message files (e.g. for maildir delivery) to find the count
3413 of files (along with the size), even though it did not need this
3414 information. It now does the scan only if it needs to find either the
3415 size of the count of files.
3417 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3419 PH/52 Two bugs concerned with error handling when the smtp transport is
3422 (i) Exim was not creating retry information for temporary errors given
3423 for individual recipients after the DATA command when the smtp transport
3424 was used in LMTP mode. This meant that they could be retried too
3425 frequently, and not timed out correctly.
3427 (ii) Exim was setting the flag that allows error details to be returned
3428 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3429 recipients that were returned after the DATA command.
3431 PH/53 This is related to PH/52, but is more general: for any failing address,
3432 when detailed error information was permitted to be returned to the
3433 sender, but the error was temporary, then after the final timeout, only
3434 "retry timeout exceeded" was returned. Now it returns the full error as
3435 well as "retry timeout exceeded".
3437 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3438 do this, and (what is worse) MTAs that accept it.
3440 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3441 will now be deprecated.
3443 PH/56 New os.c-cygwin from the Cygwin maintainer.
3445 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3446 formats (previously only available in exim formats via -bpr, -bpru,
3447 and -bpra. Now also available in native and exiqgrep formats)
3449 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3450 with very large, slow to parse queues
3452 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3454 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3456 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3457 responses to authentication challenges, though it was showing the
3458 challenges; (ii) I've removed the CR characters from the debug output for
3461 PH/58 Allow for the insertion of a newline as well as a space when a string
3462 is turned into more than one encoded-word during RFC 2047 encoding. The
3463 Sieve code now uses this.
3465 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3466 data_4xx, lost_connection, tls_required.
3468 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3469 message was being sent as an SMTP response.
3471 PH/61 Add -l and -k options to exicyclog.
3473 PH/62 When verifying, if an address was redirected to one new address, so that
3474 verification continued, and the new address failed or deferred after
3475 having set something in $address_data, the value of $address_data was not
3476 passed back to the ACL. This was different to the case when no
3477 redirection occurred. The value is now passed back in both cases.
3479 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3480 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3481 use login_cap.h, so on its own it isn't the distinguishing feature. The
3482 new name refers directly to the setclassresources() function.
3484 PH/65 Added configuration files for NetBSD3.
3486 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3488 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3489 is preferred over IPv4.
3491 PH/68 The bounce_return_message and bounce_return_body options were not being
3492 honoured for bounces generated during the reception of non-SMTP messages.
3493 In particular, this applied to messages rejected by the ACL. This bug has
3494 been fixed. However, if bounce_return_message is true and bounce_return_
3495 body is false, the headers that are returned for a non-SMTP message
3496 include only those that have been read before the error was detected.
3497 (In the case of an ACL rejection, they have all been read.)
3499 PH/69 The HTML version of the specification is now built in a directory called
3500 spec_html instead of spec.html, because the latter looks like a path with
3501 a MIME-type, and this confuses some software.
3503 PH/70 Catch two compiler warnings in sieve.c.
3505 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3506 function verify_get_ident() calls ip_connect() to connect a socket, but
3507 if the "connect()" function timed out, ip_connect() used to close the
3508 socket. However, verify_get_ident() also closes the socket later, and in
3509 between Exim writes to the log, which may get opened at this point. When
3510 the socket was closed in ip_connect(), the log could get the same file
3511 descriptor number as the socket. This naturally causes chaos. The fix is
3512 not to close the socket in ip_connect(); the socket should be closed by
3513 the function that creates it. There was only one place in the code where
3514 this was missing, in the iplookup router, which I don't think anybody now
3515 uses, but I've fixed it anyway.
3517 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3518 well as to direct DNS lookups. Otherwise the handling of names in host
3519 lists is inconsistent and therefore confusing.
3525 PH/01 Two changes to the default runtime configuration:
3527 (1) Move the checks for relay_from_hosts and authenticated clients from
3528 after to before the (commented out) DNS black list checks.
3530 (2) Add control=submission to the relay_from_hosts and authenticated
3531 clients checks, on the grounds that messages accepted by these
3532 statements are most likely to be submissions.
3534 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3536 (1) Generate an error if the third argument for the ${prvs expansion is
3539 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3542 (3) Reset the variables that are obtained from the first argument of
3543 ${prvscheck and used in the second argument before leaving the code,
3544 because their memory is reclaimed, so using them afterwards may do
3547 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3548 one (it's much easier than Tom thought :-).
3550 (5) Because of (4), we can now allow for the use of $prvscheck_result
3551 inside the third argument.
3553 PH/03 For some reason, the default setting of PATH when running a command from
3554 a pipe transport was just "/usr/bin". I have changed it to
3557 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3558 anything to be listed in the output from -bV.
3560 PH/05 When a filter generated an autoreply, the entire To: header line was
3561 quoted in the delivery log line, like this:
3563 => >A.N.Other <ano@some.domain> <original@ddress> ...
3565 This has been changed so that it extracts the operative address. There
3566 may be more than one such address. If so, they are comma-separated, like
3569 => >ano@some.domain,ona@other.domain <original@ddress> ...
3571 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3572 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3573 not being looked up in the rDNS to get a host name, Exim was showing the
3574 IP address twice in Received: lines, even though the IP addresses were
3575 identical. For example:
3577 Received: from [1.2.3.4] (helo=[1.2.3.4])
3579 However, if the real host name was known, it was omitting the HELO data
3580 if it matched the actual IP address. This has been tidied up so that it
3581 doesn't show the same IP address twice.
3583 PH/07 When both +timestamp and +memory debugging was on, the value given by
3584 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3585 called by the debug printing, thereby overwriting the timestamp buffer.
3586 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3588 PH/08 When the original message was included in an autoreply transport, it
3589 always said "this is a copy of the message, including all the headers",
3590 even if body_only or headers_only was set. It now gives an appropriate
3593 PH/09 Applied a patch from the Sieve maintainer which:
3595 o fixes some comments
3596 o adds the (disabled) notify extension core
3597 o adds some debug output for the result of if/elsif tests
3598 o points to the current vacation draft in the documentation
3599 and documents the missing references header update
3603 o fixes a bug in processing the envelope test (when testing
3604 multiple envelope elements, the last element determined the
3607 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3608 Electronic Mail") by including:
3610 Auto-submitted: auto-generated
3612 in the messages that it generates (bounce messages and others, such as
3613 warnings). In the case of bounce messages for non-SMTP messages, there was
3614 also a typo: it was using "Auto_submitted" (underscore instead of
3615 hyphen). Since every message generated by Exim is necessarily in response
3616 to another message, thes have all been changed to:
3618 Auto-Submitted: auto-replied
3620 in accordance with these statements in the RFC:
3622 The auto-replied keyword:
3624 - SHOULD be used on messages sent in direct response to another
3625 message by an automatic process,
3627 - MUST NOT be used on manually-generated messages,
3629 - MAY be used on Delivery Status Notifications (DSNs) and Message
3630 Disposition Notifications (MDNs),
3632 - MUST NOT be used on messages generated by automatic or periodic
3633 processes, except for messages which are automatic responses to
3636 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3637 to the default Received: header definition.
3639 PH/12 Added log selector acl_warn_skipped (default on).
3641 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3642 variables because (a) they are in the wrong storage pool and (b) even if
3643 they were copied, it wouldn't work properly because of the caching.
3645 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3646 checking when decoding. Apparently there are clients that generate
3647 overlong encoded strings. Why am I not surprised?
3649 PH/15 If the first argument of "${if match_address" was not empty, but did not
3650 contain an "@" character, Exim crashed. Now it writes a panic log message
3651 and treats the condition as false.
3653 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3655 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3656 extension "envelope-auth". The code is finished and in agreement with
3657 other implementations, but there is no documentation so far and in fact,
3658 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3659 not changing the active code.
3661 Print executed "if" and "elsif" statements when debugging is used. This
3662 helps a great deal to understand what a filter does.
3664 Document more things not specified clearly in RFC3028. I had all this
3665 sorted out, when out of a sudden new issues came to my mind. Oops."
3667 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3670 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3671 canonical form (as documented). However, after a host name lookup from
3672 the IP address, check_host() was doing a simple string comparison with
3673 addresses acquired from the DNS when checking that the found name did
3674 have the original IP as one of its addresses. Since any found IPv6
3675 addresses are likely to be in abbreviated form, the comparison could
3676 fail. Luckily, there already exists a function for doing the comparison
3677 by converting both addresses to binary, so now that is used instead of
3678 the text comparison.
3680 PH/20 There was another similar case to PH/19, when a complete host name was
3681 given in a host list; looking up its IP address could give an abbreviated
3682 form, whereas the current host's name might or might not be abbreviated.
3683 The same fix has been applied.
3689 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3690 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3693 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3694 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3696 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3698 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3699 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3700 Also, for "Auto-Submitted", treat anything other than "no" as
3701 non-personal, in accordance with RFC 3834. (Previously it treated
3702 anything starting "auto-" as non-personal.)
3704 TF/01 The control=submission/name=... option had a problem with syntax
3705 errors if the name included a slash character. The /name= option
3706 now slurps the rest of the string, so it can include any characters
3707 but it must come last in the list of options (after /sender_retain
3710 PH/05 Some modifications to the interface to the fake nameserver for the new
3718 TK/01 Added the "success_on_redirect" address verification option. See
3719 NewStuff for rationale and an example.
3721 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3723 PH/02 Patch to exigrep to allow it to work on syslog lines.
3725 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3726 fread() to skip over the body file's header line, because in Cygwin the
3727 header line is locked and is inaccessible.
3729 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3730 co-exist for some time) to make it clear that it is the Exim ID that is
3731 referenced, not the Message-ID: header line.
3733 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3734 string_format() function, because snprintf() does not exist on all
3737 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3738 ${stat: expansion item.
3740 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3741 protocol synchronization error", to keep the pedants happy.
3743 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3744 well as for IRIX systems, when gcc is being used. See the host.c source
3747 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3749 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3752 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3753 transport and to the smtp transport in LMTP mode.
3755 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3757 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3758 run a filter in a subprocess. This could lead to confusion in subsequent
3759 lookups in the parent process. There should also be a search_tidyup() at
3760 the end of the subprocess.
3762 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3763 only if the host matched helo_try_verify_hosts, which caused the
3764 verification to occur when the EHLO/HELO command was issued. The ACL just
3765 tested the remembered result. Now, if a previous verification attempt has
3766 not happened, "verify = helo" does it there and then.
3768 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3770 TK/03 Fix log output including CR from clamd.
3772 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3773 exist provoked a memory error which could cause a segfault.
3775 PH/15 Installed PCRE 6.2
3777 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3779 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3780 of the problem. Specifically, suggested +O2 rather than +O1 for the
3783 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3785 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3786 record, the port was not correctly logged when the outgoing_port log
3787 selector was set (it logged the transort's default port).
3789 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3790 fallback_hosts, and "hosts" in the smtp transport.
3792 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3793 host errors such as "Connection refused".
3795 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3796 authentication with radiusclient 0.4.9:
3798 - Error returned from rc_read_config was caught wrongly
3799 - Username/password not passed on to radius server due to wrong length.
3801 The presumption is that some radiusclient API changes for 4.51/PH/17
3802 were not taken care of correctly. The code is still untested by me (my
3803 Linux distribution still has 0.3.2 of radiusclient), but it was
3804 contributed by a Radius user.
3806 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3807 expanding the "port" option of the smtp transport.
3809 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3810 while reading a MIME header. Thanks to Tom Hughes for a patch.
3812 PH/24 Include config.h inside local_scan.h so that configuration settings are
3815 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3816 that in an ACL for RCPT (for example), you can examine exactly what was
3819 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3820 commands, but it was not correctly comparing the address with the actual
3821 client host address. Thus, it would show the EHLO address in Received:
3822 header lines when this was not necessary.
3824 PH/27 Added the % operator to ${eval:}.
3826 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3827 it should be ignoring failures (because with -C, for example, it has lost
3828 privilege). It wasn't ignoring creation failures other than "already
3831 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3834 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3835 because an input line was too long, either on its own, or by virtue of
3836 too many continuations, the temporary file was not being removed, and the
3837 return code was incorrect.
3839 PH/31 Missing "BOOL" in function definition in filtertest.c.
3841 PH/32 Applied Sieve patches from the maintainer.
3843 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3845 PH/33 Added "verify = not_blind".
3847 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3848 Local/Makefile (with some defaults set). These are used in built scripts
3849 such as exicyclog, but they have never been used in the exim_install
3850 script (though there are many overriding facilities there). I have
3851 arranged that the exim_install script now takes note of these two
3854 PH/35 Installed configuration files for Dragonfly.
3856 PH/36 When a locally submitted message by a trusted user did not contain a
3857 From: header, and the sender address was obtained from -f or from an SMTP
3858 MAIL command, and the trusted user did not use -F to supply a sender
3859 name, $originator_name was incorrectly used when constructing a From:
3860 header. Furthermore, $originator_name was used for submission mode
3861 messages from external hosts without From: headers in a similar way,
3862 which is clearly wrong.
3864 PH/37 Added control=suppress_local_fixups.
3866 PH/38 When log_selector = +received_sender was set, and the addition of the
3867 sender made the log line's construction buffer exactly full, or one byte
3868 less than full, an overflow happened when the terminating "\n" was
3871 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3872 when the result of a list match is failure because a DNS lookup failed.
3874 PH/40 RM_COMMAND is now used in the building process.
3876 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3877 the "build-* directories that it finds.
3879 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3880 address in a domain literal was a prefix of an interface address.
3882 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3883 when verifying a sender address, unless rewrite_headers is false.
3885 PH/44 Wrote a long comment about why errors_to addresses are verified as
3886 recipients, not senders.
3888 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3889 the ratelimit ACL was added.
3891 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3893 PH/46 Added extra information about PostgreSQL errors to the error string.
3895 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3896 avoiding the need to install special zones in a real server. This is
3897 backwards compatible; if it can't find the fake resolver, it drops back.
3898 Thus, both old and new test suites can be run.
3900 TF/02 Added util/ratelimit.pl
3902 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3903 clock is set back in time.
3905 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3906 Candler <B.Candler@pobox.com>.
3908 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3909 OK for addresses that are the result of redirections.
3911 PH/48 A number of further additions for the benefit of the new test suite,
3912 including a fake gethostbyname() that interfaces to the fake DNS resolver
3915 TF/06 The fix for widen_domains has also been applied to qualify_single and
3916 search_parents which are the other dnslookup options that can cause
3919 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3922 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3924 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3925 service" if it encountered an MX record with an empty target hostname.
3926 The message is now "an MX or SRV record indicated no SMTP service".
3928 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3929 if the DNS of the sending site is misconfigured. This is quite a
3930 common situation. This change restores the behaviour of treating a
3931 helo verification defer as a failure.
3933 PH/51 If self=fail was set on a router, the bounce message did not include the
3934 actual error message.
3940 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3942 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3943 command itself ended in error, the underlying message about the transport
3944 filter timeout was being overwritten with the pipe command error. Now the
3945 underlying error message should be appended to the second error message.
3947 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3949 PH/02 Reduce the amount of output that "make" produces by default. Full output
3950 can still be requested.
3952 PH/03 The warning log line about a condition test deferring for a "warn" verb
3953 was being output only once per connection, rather than after each
3954 occurrence (because it was using the same function as for successful
3955 "warn" verbs). This seems wrong, so I have changed it.
3957 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3958 it should not have, which might have caused a crash in the right
3959 circumstances, but probably never did.
3961 PH/04 Installed a modified version of Tony Finch's patch to make submission
3962 mode fix the return path as well as the Sender: header line, and to
3963 add a /name= option so that you can make the user's friendly name appear
3966 TF/03 Added the control = fakedefer ACL modifier.
3968 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3969 Mark Lowes for thorough testing.
3971 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3973 TK/03 Merged latest SRS patch from Miles Wilton.
3975 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3976 with the definition in sysexits.h (which is #included earlier).
3977 Fortunately, Exim does not actually use EX_OK. The code used to try to
3978 preserve the sysexits.h value, by assuming that macro definitions were
3979 scanned for macro replacements. I have been disabused of this notion,
3980 so now the code just undefines EX_OK before #including unistd.h.
3982 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3983 in the smtp transport. When a block could not be written in a single
3984 write() function, the timeout was being re-applied to each part-write.
3985 This seems wrong - if the receiver was accepting one byte at a time it
3986 would take for ever. The timeout is now adjusted when this happens. It
3987 doesn't have to be particularly precise.
3989 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3990 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3992 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3993 without a domain if the check to <postmaster@domain> fails.
3995 SC/01 Eximstats: added -xls and the ability to specify output files
3996 (patch written by Frank Heydlauf).
3998 SC/02 Eximstats: use FileHandles for outputting results.
4000 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4002 SC/04 Eximstats: fixed display of large numbers with -nvr option
4004 SC/05 Eximstats: fixed merging of reports with empty tables.
4006 SC/06 Eximstats: added the -include_original_destination flag
4008 SC/07 Eximstats: removed tabs and trailing whitespace.
4010 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4012 TK/06 MBOX spool code: Add real "From " MBOX separator line
4013 so the .eml file is really in mbox format (even though
4014 most programs do not really care). Patch from Alex Miller.
4016 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4017 The latter is generated from $received_to and is only set if the
4018 message has one envelope recipient. SA can use these headers,
4019 obviously out-of-the-box. Patch from Alex Miller.
4021 PH/08 The ${def test on a variable was returning false if the variable's
4022 value was "0", contrary to what the specification has always said!
4023 The result should be true unless the variable is empty.
4025 PH/09 The syntax error of a character other than { following "${if
4026 def:variable_name" (after optional whitespace) was not being diagnosed.
4027 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4028 accidental colon was present, for example, could give incorrect results.
4030 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4031 result is used (not including appendfile, where other changes are about
4034 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4035 This involved changing a lot of size variables from int to off_t. It
4036 should work with maildirs and everything.
4038 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4039 spamd dying while we are connected to it.
4041 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4044 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4045 function for BDB 4.3.
4047 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4049 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4050 the sources for PCRE have been re-arranged and more files are now
4053 PH/15 The code I had for printing potentially long long variables in PH/11
4054 above was not the best (it lost precision). The length of off_t variables
4055 is now inspected at build time, and an appropriate printing format (%ld
4056 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4057 to be "long long int" or "long int". This is needed for the internal
4058 formatting function string_vformat().
4060 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4061 the configuration file to be ":syslog", then the script "guesses" where
4062 the logs files are, rather than using the compiled in default. In our
4063 case the guess is not the same as the compiled default, so the script
4064 suddenly stopped working when I started to use syslog. The patch checks
4065 to see if log_file_path is "". If so, it attempts to read it from exim
4066 with no configuration file to get the compiled in version, before it
4067 falls back to the previous guessing code."
4069 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4070 implementing BATV in an Exim configuration. See NewStuff for the gory
4073 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4074 Makefile that are specific to HP-UX.
4076 PH/18 If the "use_postmaster" option was set for a recipient callout together
4077 with the "random" option, the postmaster address was used as the MAIL
4078 FROM address for the random test, but not for the subsequent recipient
4079 test. It is now used for both.
4081 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4082 patch removes a few documentation additions to RFC 3028, because the
4083 latest draft now contains them. It adds the new en;ascii-case comparator
4084 and a new error check for 8bit text in MIME parts. Comparator and
4085 require names are now matched exactly. I enabled the subaddress
4086 extension, but it is not well tested yet (read: it works for me)."
4088 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4089 rework some of the code of TK/09 above to avoid the hardwired use of
4090 "%lld" and "long long". Replaced the call to snprintf() with a call to
4093 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4094 records point to non-existent hosts", "retry timeout exceeded", and
4095 "retry time not reached for any host after a long failure period".
4097 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4098 experimental DomainKeys support:
4100 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4101 (2) On an error such as an illegally used "control", the wrong name for
4102 the control was given.
4104 These problems did NOT occur unless DomainKeys support was compiled.
4106 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4108 PH/24 Added ${if match_ip condition.
4110 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4111 databases so that it will be absolutely obvious if a crash occurs in the
4112 DB library. This is a regular occurrence (often caused by mis-matched
4115 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4116 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4117 picked up on a user's system that detects such things. There doesn't seem
4118 to be a gcc warning option for this - only an attribute that has to be
4119 put on the function's prototype. It seems that in Fedora Core 4 they have
4120 set this on a number of new functions. No doubt there will be more in due
4123 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4124 specify a transport. However, if an address that was verified by such a
4125 router was the subject of a callout, Exim crashed because it tried to
4126 read the rcpt_include_affixes from the non-existent transport. Now it
4127 just assumes that the setting of that option is false. This bug was
4128 introduced by 4.51/PH/31.
4130 PH/28 Changed -d+all to exclude +memory, because that information is very
4131 rarely of interest, but it makes the output a lot bigger. People tend to
4132 do -d+all out of habit.
4134 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4135 code in os-type was giving problems when libc.so lives in lib64, like on
4138 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4139 aren't the modern standard, and it seems that some systems' include files
4140 don't always have them. Exim was already checking for some of the newer
4141 ones like T_AAAA, and defining it itself. I've added checks for all the
4142 record types that Exim uses.
4144 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4145 not automatically generating a new one, as it is supposed to. This
4146 prevented TLS from working. If the file did exist, but contained invalid
4147 data, a new version was generated, as expected. It was only the case of a
4148 non-existent file that was broken.
4150 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4151 with a change in libdomainkeys > 0.64.
4153 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4154 from DNS. If the selector record carries the flag, it now has
4155 precedence over the domain-wide flag.
4157 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4159 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4160 the use of a port name that isn't defined in /etc/services) occurred, the
4161 message was deferred as in a normal delivery, and thus remained on the
4162 spool, instead of being failed because of the mua_wrapper setting. This
4163 is now fixed, and I tidied up some of the mua_wrapper messages at the
4166 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4167 the messages in a hash of arrays rather than using individual hashes.
4168 This is a bit cleaner and results in dramatic memory savings, albeit
4169 at a slight CPU cost.
4171 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4172 as requested by Marc Sherman.
4174 SC/10 Eximstats: added histograms for user specified patterns as requested
4177 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4179 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4180 fopen() in the content-scanning modules that did not already have it.
4186 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4187 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4189 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4191 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4193 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4194 to test Sieve filters that use "vacation".
4196 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4197 that changes the way the GnuTLS parameters are stored in the cache file.
4198 The new format can be generated externally. For backward compatibility,
4199 if the data in the cache doesn't make sense, Exim assumes it has read an
4200 old-format file, and it generates new data and writes a new file. This
4201 means that you can't go back to an older release without removing the
4204 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4205 work if there are any delivery delays because "one_time" forces the
4206 parent to be marked "delivered", so its unseen clone is never tried
4207 again. For this reason, Exim now forbids the simultaneous setting of
4210 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4211 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4212 entry if you want to know the details. The fix, however, neglected to
4213 consider the case when local delivery batching is involved. The test for
4214 "previously delivered" was not happening when checking to see if an
4215 address could be batched with a previous (undelivered) one; under
4216 certain circumstances this could lead to multiple deliveries to the same
4219 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4220 in its include files, and this causes problems building Exim.
4222 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4223 header_syntax) but Exim was just ignoring anything given after a slash.
4224 In particular, this caused confusion with an attempt to use "verify =
4225 reverse_host_lookup/defer_ok". An error is now given when options are
4226 supplied for verify items that do not have them. (Maybe reverse_host_
4227 lookup should have a defer_ok option, but that's a different point.)
4229 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4230 defined by RFC 821) to 2048, because there were problems with some AUTH
4231 commands, and RFC 1869 says the size should be increased for extended
4232 SMTP commands that take arguments.
4234 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4237 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4238 "unknown" error; now it says that the functionality isn't in the binary.
4240 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4241 an address' error message when a string expansion fails (syntax or
4242 whatever). Otherwise the password may appear in the log. Following change
4243 PH/42 below, there is no longer a chance of it appearing in a bounce
4246 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4248 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4249 of its addresses was ever tried. (Bugzilla bug #2.)
4251 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4252 the result incorrectly in the debug output. (It correctly added a newline
4253 to what was transported.)
4255 TF/01 Added $received_time.
4257 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4258 commented out examples of how to interface to a virus scanner and to
4259 SpamAssassin. Also added commented examples of av_scanner and
4260 spamd_address settings.
4262 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4263 and controls are allowed in which ACLs. There were a couple of minor
4264 errors. Some of the entries in the conditions table (which is a table of
4265 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4266 negation of where the condition IS allowed.
4268 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4270 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4271 header file does not have a version number, so I've had to invent a new
4272 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4273 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4274 radiusclient), but it was contributed by a Radius user.
4276 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4277 files or directories, for OpenSSL.
4279 PH/19 When an Exim process that is running as root has to create an Exim log
4280 file, it does so in a subprocess that runs as exim:exim so as to get the
4281 ownership right at creation (otherwise, other Exim processes might see
4282 the file with the wrong ownership). There was no test for failure of this
4283 fork() call, which would lead to the process getting stuck as it waited
4284 for a non-existent subprocess. Forks do occasionally fail when resources
4285 run out. I reviewed all the other calls to fork(); they all seem to check
4288 PH/20 When checking for unexpected SMTP input at connect time (before writing
4289 the banner), Exim was not dealing correctly with a non-positive return
4290 from the read() function. If the client had disconnected by this time,
4291 the result was a log entry for a synchronization error with an empty
4292 string after "input=" when read() returned zero. If read() returned -1
4293 (an event I could not check), uninitialized data bytes were printed.
4294 There were reports of junk text (parts of files, etc) appearing after
4297 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4299 PH/22 Added support for macro redefinition, and (re)definition in between
4300 driver and ACL definitions.
4302 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4303 forgetting to use the resulting value; it was using the unexpanded value.
4305 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4306 hadn't been configured. The fix is from Juergen Kreileder, who
4307 understands it better than I do:
4309 "Here's what I see happening with three configured cyrus_sasl
4310 authenticators configured (plain, login, cram-md5):
4312 On startup auth_cyrus_sasl_init() gets called for each of these.
4313 This means three calls to sasl_listmech() without a specified mech_list.
4314 => SASL tests which mechs of all available mechs actually work
4315 => three warnings about OTP not working
4316 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4318 With the patch, sasl_listmech() also gets called three times. But now
4319 SASL's mech_list option is set to the server_mech specified in the the
4320 authenticator. Or in other words, the answer from sasl_listmech()
4321 gets limited to just the mech you're testing for (which is different
4323 => the return list contains just 'plain' or 'login', 'cram-md5' or
4324 nothing depending on the value of ob->server_mech.
4326 I've just tested the patch: Authentication still works fine,
4327 unavailable mechs specified in the exim configuration are still
4328 caught, and the auth.log warnings about OTP are gone."
4330 PH/25 When debugging is enabled, the contents of the command line are added
4331 to the debugging output, even when log_selector=+arguments is not
4334 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4335 answer is "GNU", and only if the return is "GNU/something" is the answer
4338 PH/27 $acl_verify_message is now set immediately after the failure of a
4339 verification in an ACL, and so is available in subsequent modifiers. In
4340 particular, the message can be preserved by coding like this:
4342 warn !verify = sender
4343 set acl_m0 = $acl_verify_message
4345 Previously, $acl_verify_message was set only while expanding "message"
4346 and "log_message" when a very denied access.
4348 PH/28 Modified OS/os.c-Linux with
4350 -#ifndef OS_LOAD_AVERAGE
4351 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4353 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4354 nomenclature these days.)
4356 PH/29 Installed patch from the Sieve maintainer that adds the options
4357 sieve_useraddress and sieve_subaddress to the redirect router.
4359 PH/30 In these circumstances:
4360 . Two addresses routed to the same list of hosts;
4361 . First host does not offer TLS;
4362 . First host accepts first address;
4363 . First host gives temporary error to second address;
4364 . Second host offers TLS and a TLS session is established;
4365 . Second host accepts second address.
4366 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4367 and peerdn, if requested) that were in fact used only for the second
4370 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4371 attention to any local part prefix or suffix that was matched by the
4372 router that accepted the address. It now behaves in the same way as it
4373 does for delivery: the affixes are removed from the local part unless
4374 rcpt_include_affixes is set on the transport.
4376 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4377 timeout during the DATA phase of an incoming message.
4379 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4380 applied a patch sanctioned by the Sieve maintainer.
4382 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4383 the uid or gid is negative. A case of a negative gid caused this to be
4384 noticed. The fix allows for either to be negative.
4386 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4387 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4390 PH/36 The change PH/12 above was broken. Fixed it.
4392 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4393 the grounds that routing the same address twice would always produce the
4394 same answer. This might have been true once, but it is certainly no
4395 longer true now. Routing a child address may depend on the previous
4396 routing that produced that child. Some complicated redirection strategies
4397 went wrong when messages had multiple recipients, and made Exim's
4398 behaviour dependent on the order in which the addresses were given.
4400 I have moved the duplicate checking until after the routing is complete.
4401 Exim scans the addresses that are assigned to local and remote
4402 transports, and removes any duplicates. This means that more work will be
4403 done, as duplicates will always all be routed, but duplicates are
4404 presumably rare, so I don't expect this is of any significance.
4406 For deliveries to pipes, files, and autoreplies, the duplicate checking
4407 still happens during the routing process, since they are not going to be
4410 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4411 It corrects a timeout issue with spamd. This is Ian's comment: "The
4412 background is that sometimes spamd either never reads data from a
4413 connection it has accepted, or it never writes response data. The exiscan
4414 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4415 blindly assumes that writes won't block so it may never time out."
4417 PH/39 Allow G after quota size as well as K and M.
4419 PH/40 The value set for $authenticated_id in an authenticator may not contain
4420 binary zeroes or newlines because the value is written to log lines and
4421 to spool files. There was no check on this. Now the value is run through
4422 the string_printing() function so that such characters are converted to
4423 printable escape sequences.
4425 PH/41 $message_linecount is a new variable that contains the total number of
4426 lines in the message. Compare $body_linecount, which is the count for the
4429 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4430 bounce and delay warning messages, except in certain special cases, which
4433 (a) An SMTP error message from a remote host;
4434 (b) A message specified in a :fail: redirection;
4435 (c) A message specified in a "fail" command in a system filter;
4436 (d) A message specified in a FAIL return from the queryprogram router;
4437 (e) A message specified by the cannot_route_message router option.
4439 In these cases only, Exim does include the error details in bounce and
4440 warning messages. There are also a few cases where bland messages such
4441 as "unrouteable address" or "local delivery error" are given.
4443 PH/43 $value is now also set for the "else" part of a ${run expansion.
4445 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4446 being worked on, but at least Exim now implements the latest version to
4449 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4450 process to complete was treated as a delivery failure, a timeout while
4451 writing the message to the pipe was logged, but erroneously treated as a
4452 successful delivery. Such timeouts include transport filter timeouts. For
4453 consistency with the overall process timeout, these timeouts are now
4454 treated as errors, giving rise to delivery failures by default. However,
4455 there is now a new Boolean option for the pipe transport called
4456 timeout_defer, which, if set TRUE, converts the failures into defers for
4457 both kinds of timeout. A transport filter timeout is now identified in
4460 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4461 systems where "make" and "gmake" are different, calling "gmake" at top
4462 level broke things. I've arranged for the value of $(MAKE) to be passed
4463 from the Makefile to this script so that it can call the same version of
4467 A note about Exim versions 4.44 and 4.50
4468 ----------------------------------------
4470 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4471 changes of various kinds. As a consequence, a big documentation update was
4472 needed. This delayed the release for rather longer than seemed good, especially
4473 in the light of a couple of (minor) security issues. Therefore, the changes
4474 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4475 release. So 4.44 and 4.50 are in effect two different branches that both start
4478 I have left the 4.50 change log unchanged; it contains all the changes since
4479 4.43. The change log for 4.44 is below; many of its items are identical to
4480 those for 4.50. This seems to be the most sensible way to preserve the
4481 historical information.
4487 1. Minor wording change to the doc/README.SIEVE file.
4489 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4490 computation of the current number of files was incorrect.
4492 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4493 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4496 4. Give more explanation in the error message when the command for a transport
4497 filter fails to execute.
4499 5. There are several places where Exim runs a non-Exim command in a
4500 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4501 was being done only for the command run by the queryprogram router. It is
4502 now done for all such subprocesses. The other cases are: ${run, transport
4503 filters, and the commands run by the lmtp and pipe transports.
4505 6. Added CONFIGURE_GROUP build-time option.
4507 7. Some older OS have a limit of 256 on the maximum number of file
4508 descriptors. Exim was using setrlimit() to set 1000 as a large value
4509 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4510 systems. I've change it so that if it can't get 1000, it tries for 256.
4512 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4513 was an oversight, and furthermore, ever since the addition of extra
4514 controls (e.g. 4.43/32), the checks on when to allow different forms of
4515 "control" were broken. There should now be diagnostics for all cases when a
4516 control that does not make sense is encountered.
4518 9. Added the /retain_sender option to "control=submission".
4520 10. $recipients is now available in the predata ACL (oversight).
4522 11. Tidy the search cache before the fork to do a delivery from a message
4523 received from the command line. Otherwise the child will trigger a lookup
4524 failure and thereby defer the delivery if it tries to use (for example) a
4525 cached ldap connection that the parent has called unbind on.
4527 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4528 of $address_data from the recipient verification was clobbered by the
4529 sender verification.
4531 13. The value of address_data from a sender verification is now available in
4532 $sender_address_data in subsequent conditions in the ACL statement.
4534 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4536 15. Added a new option "connect=<time>" to callout options, to set a different
4539 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4540 was its contents. (It was OK if the option was not defined at all.)
4542 17. A "Completed" log line is now written for messages that are removed from
4543 the spool by the -Mrm option.
4545 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4546 information about exactly what failed.
4548 19. Added -dd to debug only the daemon process.
4550 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4551 handles timeouts, both on the server side and network timeouts. Renamed the
4552 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4554 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4555 It is now set to "smtps".
4557 22. $host_address is now set to the target address during the checking of
4558 ignore_target_hosts.
4560 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4561 being passed; this would have caused $sender_host_name to have been used if
4562 matching the list had actually called for a host name (not very likely,
4563 since this list is usually IP addresses). A host name is now passed as
4566 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4567 code that specifies a non-restarting handler (typically sigaction() in
4568 modern systems) in an attempt to fix a rare and obscure crash bug.
4570 25. Narrowed the window for a race in the daemon that could cause it to ignore
4571 SIGCHLD signals. This is not a major problem, because they are used only to
4572 wake it up if nothing else does.
4574 26. A malformed maildirsize file could cause Exim to calculate negative values
4575 for the mailbox size or file count. Odd effects could occur as a result.
4576 The maildirsize information is now recalculated if the size or filecount
4579 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4580 support for a long time. Removed HAVE_SYS_VFS_H.
4582 28. Installed the latest version of exipick from John Jetmore.
4584 29. In an address list, if the pattern was not a regular expression, an empty
4585 subject address (from a bounce message) matched only if the pattern was an
4586 empty string. Non-empty patterns were not even tested. This was the wrong
4587 because it is perfectly reasonable to use an empty address as part of a
4588 database query. An empty address is now tested by patterns that are
4589 lookups. However, all the other forms of pattern expect the subject to
4590 contain a local part and a domain, and therefore, for them, an empty
4591 address still always fails if the pattern is not itself empty.
4593 30. Exim went into a mad DNS loop when attempting to do a callout where the
4594 host was specified on an smtp transport, and looking it up yielded more
4595 than one IP address.
4597 31. Re-factored the code for checking spool and log partition space into a
4598 function that finds that data and another that does the check. The former
4599 is then used to implement four new variables: $spool_space, $log_space,
4600 $spool_inodes, and $log_inodes.
4602 32. The RFC2047 encoding function was originally intended for short strings
4603 such as real names; it was not keeping to the 75-character limit for
4604 encoded words that the RFC imposes. It now respects the limit, and
4605 generates multiple encoded words if necessary. To be on the safe side, I
4606 have increased the buffer size for the ${rfc2047: expansion operator from
4609 33. It is now permitted to omit both strings after an "if" condition; if the
4610 condition is true, the result is "true". As before, when the second string
4611 is omitted, a false condition yields an empty string. This makes it less
4612 cumbersome to write custom ACL and router conditions.
4614 34. Failure to deliver a bounce message always caused it to be frozen, even if
4615 there was an errors_to setting on the router. The errors_to setting is now
4618 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4619 canonical form (fully expanded) before being placed in
4620 $sender_host_address.
4622 36. The table in the code that translates DNS record types into text (T_A to
4623 "A" for instance) was missing entries for NS and CNAME. It is just possible
4624 that this could have caused confusion if both these types were looked up
4625 for the same domain, because the text type is used as part of Exim's
4626 per-process caching. But the chance of anyone hitting this buglet seems
4629 37. The dnsdb lookup has been extended in a number of ways.
4631 (1) There is a new type, "zns", which walks up the domain tree until it
4632 finds some nameserver records. It should be used with care.
4634 (2) There is a new type, "mxh", which is like "mx" except that it returns
4635 just the host names, not the priorities.
4637 (3) It is now possible to give a list of domains (or IP addresses) to be
4638 looked up. The behaviour when one of the lookups defers can be
4639 controlled by a keyword.
4641 (4) It is now possible to specify the separator character for use when
4642 multiple records are returned.
4644 38. The dnslists ACL condition has been extended: it is now possible to supply
4645 a list of IP addresses and/or domains to be looked up in a particular DNS
4648 39. Added log_selector=+queue_time_overall.
4650 40. When running the queue in the test harness, wait just a tad after forking a
4651 delivery process, to get repeatability of debugging output.
4653 41. Include certificate and key file names in error message when GnuTLS fails
4654 to set them up, because the GnuTLS error message doesn't include the name
4655 of the failing file when there is a problem reading it.
4657 42. Allow both -bf and -bF in the same test run.
4659 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4661 44. The "Exiscan patch" is now merged into the mainline Exim source.
4663 45. Sometimes the final signoff response after QUIT could fail to get
4664 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4665 < 0 before doing a fflush(). This bug looks as though it goes back to the
4666 introduction of TLS in release 3.20, but "sometimes" must have been rare
4667 because the tests only now provoked it.
4669 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4670 (this can affect the format of dates).
4672 47. exim_tidydb, when checking for the continued existence of a message for
4673 which it has found a message-specific retry record, was not finding
4674 messages that were in split spool directories. Consequently, it was
4675 deleting retry records that should have stayed in existence.
4677 48. Steve fixed some bugs in eximstats.
4679 49. The SPA authentication driver was not abandoning authentication and moving
4680 on to the next authenticator when an expansion was forced to fail,
4681 contradicting the general specification for all authenticators. Instead it
4682 was generating a temporary error. It now behaves as specified.
4684 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4685 (the order specifies the preference for clients). The order is now AES256,
4686 AES128, 3DES, ARCFOUR128.
4688 51. Small patch to Sieve code - explicitly set From: when generating an
4691 52. Exim crashed if a remote delivery caused a very long error message to be
4692 recorded - for instance if somebody sent an entire SpamAssassin report back
4693 as a large number of 550 error lines. This bug was coincidentally fixed by
4694 increasing the size of one of Exim's internal buffers (big_buffer) that
4695 happened as part of the Exiscan merge. However, to be on the safe side, I
4696 have made the code more robust (and fixed the comments that describe what
4699 53. Now that there can be additional text after "Completed" in log lines (if
4700 the queue_time_overall log selector is set), a one-byte patch to exigrep
4701 was needed to allow it to recognize "Completed" as not the last thing in
4704 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4705 patch that reportedly fixes this has been added. I am not expert enough to
4706 create a test for it. This is what the patch creator wrote:
4708 "I found a little strange behaviour of ldap code when working with
4709 Windows 2003 AD Domain, where users was placed in more than one
4710 Organization Units. When I tried to give exim partial DN, the exit code
4711 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4712 But simultaneously result of request was absolutely normal ldap result,
4713 so I produce this patch..."
4715 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4716 REFERENCE, so I have modified the code to exclude the patch when that macro
4719 55. Some experimental protocols are using DNS PTR records for new purposes. The
4720 keys for these records are domain names, not reversed IP addresses. The
4721 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4722 leaves it alone. Component reversal etc. now happens only for IP addresses.
4725 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4727 57. Double the size of the debug message buffer (to 2048) so that more of very
4728 long debug lines gets shown.
4730 58. The exicyclog utility now does better if the number of log files to keep
4731 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4733 59. Two changes related to the smtp_active_hostname option:
4735 (1) $smtp_active_hostname is now available as a variable.
4736 (2) The default for smtp_banner uses $smtp_active_hostname instead
4737 of $primary_hostname.
4739 60. The host_aton() function is supposed to be passed a string that is known
4740 to be a valid IP address. However, in the case of IPv6 addresses, it was
4741 not checking this. This is a hostage to fortune. Exim now panics and dies
4742 if the condition is not met. A case was found where this could be provoked
4743 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4744 components; fortuitously, this particular loophole had already been fixed
4745 by change 4.50/55 above.
4747 If there are any other similar loopholes, the new check in host_aton()
4748 itself should stop them being exploited. The report I received stated that
4749 data on the command line could provoke the exploit when Exim was running as
4750 exim, but did not say which command line option was involved. All I could
4751 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4752 running as the user.
4755 61. There was a buffer overflow vulnerability in the SPA authentication code
4756 (which came originally from the Samba project). I have added a test to the
4757 spa_base64_to_bits() function which I hope fixes it.
4760 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4761 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4763 63. The daemon start-up calls getloadavg() while still root for those OS that
4764 need the first call to be done as root, but it missed one case: when
4765 deliver_queue_load_max is set with deliver_drop_privilege. This is
4766 necessary for the benefit of the queue runner, because there is no re-exec
4767 when deliver_drop_privilege is set.
4769 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4770 This has been fixed.
4772 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4773 was in use, was not putting the data itself into the right store pool;
4774 consequently, it could be overwritten for a subsequent message in the same
4775 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4778 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4780 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4781 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4782 the function were treating the return as a boolean value, which happened to
4783 work because 0=false and not-0=true, but is not correct code.
4785 68. The host_aton() function was not handling scoped IPv6 addresses (those
4786 with, for example, "%eth0" on the end) correctly.
4788 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4789 negated items (that is, ~something) in unsigned ints. Some compilers
4790 apparently mutter when there is no cast.
4792 70. If an address verification called from an ACL failed, and did not produce a
4793 user-specific message (i.e. there was only a "system" message), nothing was
4794 put in $acl_verify_message. In this situation, it now puts the system
4797 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4798 session; change 4.31/43 added the unwanted input to the log line - except
4799 that it did not do this in the start of session case. It now does.
4801 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4802 This is wrong and can cause the other end to generate a synchronization
4803 error if it is another Exim or anything else that does the synchronization
4804 check. A QUIT command is no longer sent after a timeout.
4806 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4807 during host lookups.
4809 74. The defer_ok option of callout verification was not working if it was used
4810 when verifying addresses in header lines, that is, for this case:
4812 verify = header_sender/callout=defer_ok
4814 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4815 those file descriptors could be used for SMTP connections. If anything
4816 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4817 could be sent to the SMTP client, causing chaos. The daemon now opens
4818 stdin, stdout, and stderr to /dev/null when it puts itself into the
4821 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4822 log by default. The user can override this with suitable Perl magic.
4824 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4825 the log message when discard triggers, was not working for the DATA ACL or
4826 for the non-SMTP ACL.
4828 78. Error message wording change in sieve.c.
4830 79. If smtp_accept_max_per_host was set, the number of connections could be
4831 restricted to fewer than expected, because the daemon was trying to set up
4832 a new connection before checking whether the processes handling previous
4833 connections had finished. The check for completed processes is now done
4834 earlier. On busy systems, this bug wouldn't be noticed because something
4835 else would have woken the daemon, and it would have reaped the completed
4838 80. If a message was submitted locally by a user whose login name contained one
4839 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4840 It caused a spool format error. I have fixed the spool reading code. A
4841 related problem was that the "from" clause in the Received: line became
4842 illegal because of the space(s). It is now covered by ${quote_local_part.
4844 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4847 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4849 83. Patch from Sieve maintainer to fix unterminated string problem in
4850 "vacation" handling.
4852 84. Some minor changes to the Linux configuration files to help with other
4853 OS variants using glibc.
4855 85. One more patch for Sieve to update vacation handling to latest spec.
4858 ----------------------------------------------------
4859 See the note above about the 4.44 and 4.50 releases.
4860 ----------------------------------------------------
4866 1. Change 4.43/35 introduced a bug that caused file counts to be
4867 incorrectly computed when quota_filecount was set in an appendfile
4870 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4871 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4874 3. Give more explanation in the error message when the command for a transport
4875 filter fails to execute.
4877 4. There are several places where Exim runs a non-Exim command in a
4878 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4879 was being done only for the command run by the queryprogram router. It is
4880 now done for all such subprocesses. The other cases are: ${run, transport
4881 filters, and the commands run by the lmtp and pipe transports.
4883 5. Some older OS have a limit of 256 on the maximum number of file
4884 descriptors. Exim was using setrlimit() to set 1000 as a large value
4885 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4886 systems. I've change it so that if it can't get 1000, it tries for 256.
4888 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4889 was an oversight, and furthermore, ever since the addition of extra
4890 controls (e.g. 4.43/32), the checks on when to allow different forms of
4891 "control" were broken. There should now be diagnostics for all cases when a
4892 control that does not make sense is encountered.
4894 7. $recipients is now available in the predata ACL (oversight).
4896 8. Tidy the search cache before the fork to do a delivery from a message
4897 received from the command line. Otherwise the child will trigger a lookup
4898 failure and thereby defer the delivery if it tries to use (for example) a
4899 cached ldap connection that the parent has called unbind on.
4901 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4902 of $address_data from the recipient verification was clobbered by the
4903 sender verification.
4905 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4906 was its contents. (It was OK if the option was not defined at all.)
4908 11. A "Completed" log line is now written for messages that are removed from
4909 the spool by the -Mrm option.
4911 12. $host_address is now set to the target address during the checking of
4912 ignore_target_hosts.
4914 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4915 being passed; this would have caused $sender_host_name to have been used if
4916 matching the list had actually called for a host name (not very likely,
4917 since this list is usually IP addresses). A host name is now passed as
4920 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4921 code that specifies a non-restarting handler (typically sigaction() in
4922 modern systems) in an attempt to fix a rare and obscure crash bug.
4924 15. Narrowed the window for a race in the daemon that could cause it to ignore
4925 SIGCHLD signals. This is not a major problem, because they are used only to
4926 wake it up if nothing else does.
4928 16. A malformed maildirsize file could cause Exim to calculate negative values
4929 for the mailbox size or file count. Odd effects could occur as a result.
4930 The maildirsize information is now recalculated if the size or filecount
4933 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4934 support for a long time. Removed HAVE_SYS_VFS_H.
4936 18. Updated exipick to current release from John Jetmore.
4938 19. Allow an empty sender to be matched against a lookup in an address list.
4939 Previously the only cases considered were a regular expression, or an
4942 20. Exim went into a mad DNS lookup loop when doing a callout where the
4943 host was specified on the transport, if the DNS lookup yielded more than
4946 21. The RFC2047 encoding function was originally intended for short strings
4947 such as real names; it was not keeping to the 75-character limit for
4948 encoded words that the RFC imposes. It now respects the limit, and
4949 generates multiple encoded words if necessary. To be on the safe side, I
4950 have increased the buffer size for the ${rfc2047: expansion operator from
4953 22. Failure to deliver a bounce message always caused it to be frozen, even if
4954 there was an errors_to setting on the router. The errors_to setting is now
4957 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4958 canonical form (fully expanded) before being placed in
4959 $sender_host_address.
4961 24. Updated eximstats to version 1.33
4963 25. Include certificate and key file names in error message when GnuTLS fails
4964 to set them up, because the GnuTLS error message doesn't include the name
4965 of the failing file when there is a problem reading it.
4967 26. Expand error message when OpenSSL has problems setting up cert/key files.
4970 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4971 (this can affect the format of dates).
4973 28. exim_tidydb, when checking for the continued existence of a message for
4974 which it has found a message-specific retry record, was not finding
4975 messages that were in split spool directories. Consequently, it was
4976 deleting retry records that should have stayed in existence.
4978 29. eximstats updated to version 1.35
4979 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4980 1.35 - bugfix such that pie charts by volume are generated correctly
4982 30. The SPA authentication driver was not abandoning authentication and moving
4983 on to the next authenticator when an expansion was forced to fail,
4984 contradicting the general specification for all authenticators. Instead it
4985 was generating a temporary error. It now behaves as specified.
4987 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4988 (the order specifies the preference for clients). The order is now AES256,
4989 AES128, 3DES, ARCFOUR128.
4991 31. Small patch to Sieve code - explicitly set From: when generating an
4994 32. Exim crashed if a remote delivery caused a very long error message to be
4995 recorded - for instance if somebody sent an entire SpamAssassin report back
4996 as a large number of 550 error lines. This bug was coincidentally fixed by
4997 increasing the size of one of Exim's internal buffers (big_buffer) that
4998 happened as part of the Exiscan merge. However, to be on the safe side, I
4999 have made the code more robust (and fixed the comments that describe what
5002 33. Some experimental protocols are using DNS PTR records for new purposes. The
5003 keys for these records are domain names, not reversed IP addresses. The
5004 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5005 leaves it alone. Component reversal etc. now happens only for IP addresses.
5008 34. The host_aton() function is supposed to be passed a string that is known
5009 to be a valid IP address. However, in the case of IPv6 addresses, it was
5010 not checking this. This is a hostage to fortune. Exim now panics and dies
5011 if the condition is not met. A case was found where this could be provoked
5012 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5013 components; fortuitously, this particular loophole had already been fixed
5014 by change 4.50/55 or 4.44/33 above.
5016 If there are any other similar loopholes, the new check in host_aton()
5017 itself should stop them being exploited. The report I received stated that
5018 data on the command line could provoke the exploit when Exim was running as
5019 exim, but did not say which command line option was involved. All I could
5020 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5021 running as the user.
5024 35. There was a buffer overflow vulnerability in the SPA authentication code
5025 (which came originally from the Samba project). I have added a test to the
5026 spa_base64_to_bits() function which I hope fixes it.
5029 36. The daemon start-up calls getloadavg() while still root for those OS that
5030 need the first call to be done as root, but it missed one case: when
5031 deliver_queue_load_max is set with deliver_drop_privilege. This is
5032 necessary for the benefit of the queue runner, because there is no re-exec
5033 when deliver_drop_privilege is set.
5035 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5036 was in use, was not putting the data itself into the right store pool;
5037 consequently, it could be overwritten for a subsequent message in the same
5038 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5041 38. Sometimes the final signoff response after QUIT could fail to get
5042 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5043 < 0 before doing a fflush(). This bug looks as though it goes back to the
5044 introduction of TLS in release 3.20, but "sometimes" must have been rare
5045 because the tests only now provoked it.
5051 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5052 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5053 Now it returns an integer. A number of calls were still expecting a T/F
5054 return. Fortuitously, in all cases, the tests worked in OK situations,
5055 which is the norm. However, things would have gone wrong on any write
5056 failures on the smtp file descriptor. This function is used when sending
5057 messages over SMTP and also when doing verify callouts.
5059 2. When Exim is called to do synchronous delivery of a locally submitted
5060 message (the -odf or -odi options), it no longer closes stderr before doing
5063 3. Implemented the mua_wrapper option.
5065 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5067 5. Implemented the functions header_remove(), header_testname(),
5068 header_add_at_position(), and receive_remove_recipient(), and exported them
5071 6. If an ACL "warn" statement specified the addition of headers, Exim already
5072 inserted X-ACL-Warn: at the start if there was no header name. However, it
5073 was not making this test for the second and subsequent header lines if
5074 there were newlines in the string. This meant that an invalid header could
5075 be inserted if Exim was badly configured.
5077 7. Allow an ACL "warn" statement to add header lines at the start or after all
5078 the Received: headers, as well as at the end.
5080 8. Added the rcpt_4xx retry error code.
5082 9. Added postmaster_mailfrom=xxx to callout verification option.
5084 10. Added mailfrom=xxxx to the callout verification option, for verify=
5087 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5088 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5090 12. Inserted some casts to stop certain compilers warning when using pointer
5091 differences as field lengths or precisions in printf-type calls (mostly
5092 affecting debugging statements).
5094 13. Added optional readline() support for -be (dynamically loaded).
5096 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5097 same clock tick as a message's arrival, so that its received time was the
5098 same as the "first fail" time on the retry record, and that message
5099 remained on the queue past the ultimate address timeout, every queue runner
5100 would try a delivery (because it was past the ultimate address timeout) but
5101 after another failure, the ultimate address timeout, which should have then
5102 bounced the address, did not kick in. This was a "< instead of <=" error;
5103 in most cases the first failure would have been in the next clock tick
5104 after the received time, and all would be well.
5106 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5107 being recognized when the domain list was tested by the match_domain
5108 condition in an expansion string.
5110 16. Added the ${str2b64: operator.
5112 17. Exim was always calling setrlimit() to set a large limit for the number of
5113 processes, without checking whether the existing limit was already
5114 adequate. (It did check for the limit on file descriptors.) Furthermore,
5115 errors from getrlimit() and setrlimit() were being ignored. Now they are
5116 logged to the main and panic logs, but Exim does carry on, to try to do its
5117 job under whatever limits there are.
5119 18. Imported PCRE 5.0.
5121 19. Trivial typo in log message " temporarily refused connection" (the leading
5124 20. If the log selector return_path_on_delivery was set and an address was
5125 redirected to /dev/null, the delivery process crashed because it assumed
5126 that a return path would always be set for a "successful" delivery. In this
5127 case, the whole delivery is bypassed as an optimization, and therefore no
5130 21. Internal re-arrangement: the function for sending a challenge and reading
5131 a response while authentication was assuming a zero-terminated challenge
5132 string. It's now changed to take a pointer and a length, to allow for
5133 binary data in such strings.
5135 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5137 23. Exim was not respecting finduser_retries when seeking the login of the
5138 uid under which it was called; it was always trying 10 times. (The default
5139 setting of finduser_retries is zero.) Also, it was sleeping after the final
5140 failure, which is pointless.
5142 24. Implemented tls_on_connect_ports.
5144 25. Implemented acl_smtp_predata.
5146 26. If the domain in control=submission is set empty, Exim assumes that the
5147 authenticated id is a complete email address when it generates From: or
5148 Sender: header lines.
5150 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5151 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5152 chown and chgrp in /bin and hostname in /usr/bin.
5154 28. Exim was keeping the "process log" file open after each use, just as it
5155 does for the main log. This opens the possibility of it remaining open for
5156 long periods when the USR1 signal hits a daemon. Occasional processlog
5157 errors were reported, that could have been caused by this. Anyway, it seems
5158 much more sensible not to leave this file open at all, so that is what now
5161 29. The long-running daemon process does not normally write to the log once it
5162 has entered its main loop, and it closes the log before doing so. This is
5163 so that log files can straightforwardly be renamed and moved. However,
5164 there are a couple of unusual error situations where the daemon does write
5165 log entries, and I had neglected to close the log afterwards.
5167 30. The text of an SMTP error response that was received during a remote
5168 delivery was being truncated at 512 bytes. This is too short for some of
5169 the long messages that one sometimes sees. I've increased the limit to
5172 31. It is now possible to make retry rules that apply only when a message has a
5173 specific sender, in particular, an empty sender.
5175 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5176 it possible to be selective about when SMTP synchronization is enforced.
5178 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5180 32. Implemented hosts_connection_nolog.
5182 33. Added an ACL for QUIT.
5184 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5187 35. Added mailbox_size and mailbox_filecount to appendfile.
5189 36. Added control = no_multiline_responses to ACLs.
5191 37. There was a bug in the logic of the code that waits for the clock to tick
5192 in the case where the clock went backwards by a substantial amount such
5193 that the microsecond fraction of "now" was more than the microsecond
5194 fraction of "then" (but the whole seconds number was less).
5196 38. Added support for the libradius Radius client library this is found on
5197 FreeBSD (previously only the radiusclient library was supported).
5203 1. When certain lookups returned multiple values in the form name=value, the
5204 quoting of the values was not always being done properly. Specifically:
5205 (a) If the value started with a double quote, but contained no whitespace,
5207 (b) If the value contained whitespace other than a space character (i.e.
5208 tabs or newlines or carriage returns) it was not quoted.
5209 This fix has been applied to the mysql and pgsql lookups by writing a
5210 separate quoting function and calling it from the lookup code. The fix
5211 should probably also be applied to nisplus, ibase and oracle lookups, but
5212 since I cannot test any of those, I have not disturbed their existing code.
5214 2. A hit in the callout cache for a specific address caused a log line with no
5215 reason for rejecting RCPT. Now it says "Previous (cached) callout
5216 verification failure".
5218 3. There was an off-by-one bug in the queryprogram router. An over-long
5219 return line was truncated at 256 instead of 255 characters, thereby
5220 overflowing its buffer with the terminating zero. As well as fixing this, I
5221 have increased the buffer size to 1024 (and made a note to document this).
5223 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5224 when Exim is waiting for an SMTP response from a remote server, Exim
5225 restarts its select() call on the socket, thereby resetting its timeout.
5226 This is not a problem when such interrupts are rare. Somebody set up a cron
5227 job to run exiwhat every 2 minutes, which is less than the normal select()
5228 timeout (5 or 10 minutes). This meant that the select() timeout never
5229 kicked in because it was always reset. I have fixed this by comparing the
5230 time when an interrupt arrives with the time at the start of the first call
5231 to select(). If more time than the timeout has elapsed, the interrupt is
5232 treated as a timeout.
5234 5. Some internal re-factoring in preparation for the addition of Sieve
5235 extensions (by MH). In particular, the "personal" test is moved to a
5236 separate function, and given an option for scanning Cc: and Bcc: (which is
5237 not set for Exim filters).
5239 6. When Exim created an email address using the login of the caller as the
5240 local part (e.g. when creating a From: or Sender: header line), it was not
5241 quoting the local part when it contained special characters such as @.
5243 7. Installed new OpenBSD configuration files.
5245 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5246 try to make them clearer.
5248 9. Callout options, other than the timeout value, were being ignored when
5249 verifying sender addresses in header lines. For example, when using
5251 verify = header_sender/callout=no_cache
5253 the cache was (incorrectly) being used.
5255 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5256 only the Cygwin environment.
5258 11. When return_path_on_delivery was set as a log selector, if different remote
5259 addresses in the same message used different return paths and parallel
5260 remote delivery occurred, the wrong values would sometimes be logged.
5261 (Whenever a remote delivery process finished, the return path value from
5262 the most recently started remote delivery process was logged.)
5264 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5265 lines when AUTH and/or TLS are in use. This is the "received protocol"
5266 field. Exim used to use "asmtp" for authenticated SMTP, without any
5267 indication (in the protocol name) for TLS use. Now it follows the RFC and
5268 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5269 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5270 names appear in log lines as well as in Received: header lines.
5272 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5273 extensions, and comparison tests, and to fix some bugs.
5275 14. Changes to the "personal" filter test:
5277 (1) The test was buggy in that it was just doing the equivalent of
5278 "contains" tests on header lines. For example, if a user's address was
5279 anne@some.where, the "personal" test would incorrectly be true for
5281 To: susanne@some.where
5283 This test is now done by extracting each address from the header in turn,
5284 and checking the entire address. Other tests that are part of "personal"
5285 are now done using regular expressions (for example, to check local parts
5286 of addresses in From: header lines).
5288 (2) The list of non-personal local parts in From: addresses has been
5289 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5290 taken from the Sieve specification recommendations.
5292 (3) If the message contains any header line starting with "List-" it is
5293 treated as non-personal.
5295 (4) The test for "circular" in the Subject: header line has been removed
5296 because it now seems ill-conceived.
5298 15. Minor typos in src/EDITME comments corrected.
5300 16. Installed latest exipick from John Jetmore.
5302 17. If headers_add on a router specified a text string that was too long for
5303 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5304 of string_sprintf() is now avoided.
5306 18. $message_body_size was not set (it was always zero) when running the DATA
5307 ACL and the local_scan() function.
5309 19. For the "mail" command in an Exim filter, no default was being set for
5310 the once_repeat time, causing a random time value to be used if "once" was
5311 specified. (If the value happened to be <= 0, no repeat happened.) The
5312 default is now 0s, meaning "never repeat". The "vacation" command was OK
5313 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5314 (I found it when inspecting the code).
5316 20. There is now an overall timeout for performing a callout verification. It
5317 defaults to 4 times the callout timeout, which applies to individual SMTP
5318 commands during the callout. The overall timeout applies when there is more
5319 than one host that can be tried. The timeout is checked before trying the
5320 next host. This prevents very long delays if there are a large number of
5321 hosts and all are timing out (e.g. when the network connections are timing
5322 out). The value of the overall timeout can be changed by specifying an
5323 additional sub-option for "callout", called "maxwait". For example:
5325 verify = sender/callout=5s,maxwait=20s
5327 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5328 to the end before writing, but this should make it even safer).
5330 22. Exim was forgetting that it had advertised PIPELINING for the second and
5331 subsequent messages on an SMTP connection. It was also not resetting its
5332 memory on STARTTLS and an internal HELO.
5334 23. When Exim logs an SMTP synchronization error within a session, it now
5335 records whether PIPELINING has been advertised or not.
5337 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5338 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5339 rather than long int.
5341 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5343 26. Added the never_mail option to autoreply.
5349 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5350 crash if the getsockname() call failed; this can happen if a connection is
5351 closed very soon after it is established. The problem was simply in the
5352 order in which certain operations were done, causing Exim to try to write
5353 to the SMTP stream before it had set up the file descriptor. The bug has
5354 been fixed by making things happen in the correct order.
5360 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5361 before the connection was closed, thus losing the rejection response.
5363 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5364 some early Solaris releases, but causes trouble in current releases where
5365 socklen_t is defined.
5367 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5370 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5373 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5374 that releasing the top of it at the end releases what was used for sub-
5375 expansions (unless the block got too big). However, discard this block if
5376 the first thing is a variable or header, so that we can use its block when
5377 it is dynamic (useful for very large $message_headers, for example).
5379 6. Lookups now cache *every* query, not just the most recent. A new, separate
5380 store pool is used for this. It can be recovered when all lookup caches are
5381 flushed. Lookups now release memory at the end of their result strings.
5382 This has involved some general refactoring of the lookup sources.
5384 7. Some code has been added to the store_xxx() functions to reduce the amount
5385 of flapping under certain conditions.
5387 8. log_incoming_interface used to affect only the <= reception log lines. Now
5388 it causes the local interface and port to be added to several more SMTP log
5389 lines, for example "SMTP connection from", and rejection lines.
5391 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5393 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5395 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5396 could be overwritten at the end of the current message (or the start of a
5397 new message if it was set in a HELO ACL). The value is now preserved for
5398 the duration of the SMTP connection.
5400 12. If a transport had a headers_rewrite setting, and a matching header line
5401 contained an unqualified address, that address was qualified, even if it
5402 did not match any rewriting rules. The underlying bug was that the values
5403 of the flags that permit the existence of unqualified sender and recipient
5404 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5405 non-local messages, and by -bnq for local messages) were not being
5406 preserved with the message after it was received.
5408 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5409 "next input=" as part of the text comprising the host identity instead of
5410 the correct text. The code was using the same buffer for two different
5411 strings. However, depending on which order the printing function evaluated
5412 its arguments, the bug did not always show up. Under Linux, for example, my
5413 test suite worked just fine.
5415 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5416 doesn't work with some older versions of Perl. It has been changed to "my",
5417 which in any case is probably the better facility to use.
5419 15. A really picky compiler found some instances of statements for creating
5420 error messages that either had too many or two few arguments for the format
5423 16. The size of the buffer for calls to the DNS resolver has been increased
5424 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5425 for addresses that have a lot of PTR records. This alleviates a problem; it
5426 does not fully solve it.
5428 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5429 buffer now truncates the list and logs the incident, which is the same
5430 action as happens when Exim is looking up a host name and its aliases.
5431 Previously in this situation something unpredictable would happen;
5432 sometimes it was "internal error: store_reset failed".
5434 18. If a server dropped the connection unexpectedly when an Exim client was
5435 using GnuTLS and trying to read a response, the client delivery process
5436 crashed while trying to generate an error log message.
5438 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5439 string, for example:
5441 warn message = H1: something\nH2: something
5443 the text was added as a single header line from Exim's point of view
5444 though it ended up OK in the delivered message. However, searching for the
5445 second and subsequent header lines using $h_h2: did not work. This has been
5446 fixed. Similarly, if a system filter added multiple headers in this way,
5447 the routers could not see them.
5449 20. Expanded the error message when iplsearch is called with an invalid key to
5450 suggest using net-iplsearch in a host list.
5452 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5453 ACLs are no longer actually imposed (and a message to that effect is
5456 22. If a "gecos" field in a passwd entry contained escaped characters, in
5457 particular, if it contained a \" sequence, Exim got it wrong when building
5458 a From: or a Sender: header from that name. A second bug also caused
5459 incorrect handling when an unquoted " was present following a character
5460 that needed quoting.
5462 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5463 was not being matched caselessly.
5465 24. Arranged for all hyphens in the exim.8 source to be escaped with
5468 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5469 properly. Recipient callout cache records were still being keyed to include
5470 the sender, even when use_sender was set false. This led to far more
5471 callouts that were necessary. The sender is no longer included in the key
5472 when use_sender is false.
5474 26. Added "control = submission" modifier to ACLs.
5476 27. Added the ${base62d: operator to decode base 62 numbers.
5478 28. dnsdb lookups can now access SRV records.
5480 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5481 the configuration file.
5483 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5484 (-v) mode. This makes the output for a verbose queue run more intelligible.
5486 31. Added a use_postmaster feature to recipient callouts.
5488 32. Added the $body_zerocount variable, containing the number of binary zero
5489 bytes in the message body.
5491 33. The time of last modification of the "new" subdirectory is now used as the
5492 "mailbox time last read" when there is a quota error for a maildir
5495 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5497 35. Added +ignore_unknown as a special item in host lists.
5499 36. Code for decoding IPv6 addresses in host lists is now included, even if
5500 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5501 address was recognized as an IP address, but was then not correctly decoded
5502 into binary, causing unexpected and incorrect effects when compared with
5509 1. Very minor rewording of debugging text in manualroute to say "list of
5510 hosts" instead of "hostlist".
5512 2. If verify=header_syntax was set, and a header line with an unqualified
5513 address (no domain) and a large number of spaces between the end of the
5514 name and the colon was received, the reception process suffered a buffer
5515 overflow, and (when I tested it) crashed. This was caused by some obsolete
5516 code that should have been removed. The fix is to remove it!
5518 3. When running in the test harness, delay a bit after writing a bounce
5519 message to get a bit more predictability in the log output.
5521 4. Added a call to search_tidyup() just before forking a reception process. In
5522 theory, someone could use a lookup in the expansion of smtp_accept_max_
5523 per_host which, without the tidyup, could leave open a database connection.
5525 5. Added the variables $recipient_data and $sender_data which get set from a
5526 lookup success in an ACL "recipients" or "senders" condition, or a router
5527 "senders" option, similar to $domain_data and $local_part_data.
5529 6. Moved the writing of debug_print from before to after the "senders" test
5532 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5533 problems for message scanning, either using a data ACL, or using
5534 local_scan() because the Received: header was not generated till after they
5535 were called (in order to set the time as the time of reception completion).
5536 I have revised the way this works. The header is now generated after the
5537 body is received, but before the ACL or local_scan() are called. After they
5538 are run, the timestamp in the header is updated.
5544 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5545 before starting a queue runner without re-exec. This happened only when
5546 deliver_drop_privilege was set or when the Exim user was set to root. The
5547 effect of the bug was that timeouts during subsequent deliveries caused
5548 crashes instead of being properly handled. The handler is now left at its
5549 default (and expected) setting.
5551 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5552 message, again when deliver_drop_privilege is set or Exim is run as root.
5553 The bug described in (1) was not present in this case, but the tidying up
5554 of the other signals was missing. I have made the two cases consistent.
5556 3. The ignore_target_hosts setting on a manualroute router was being ignored
5557 for hosts that were looked up using the /MX notation.
5559 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5562 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5563 operated on the sender address. After changing the $sender_address to <>
5564 for the sender address verify, Exim was re-instated it as the original
5565 (before rewriting) address, but remembering that it had rewritten it, so it
5566 wasn't rewriting it again. This bug also had the effect of breaking the
5567 sender address verification caching when the sender address was rewritten.
5569 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5570 This has been changed so that if the ip literal address matches
5571 ignore_target_hosts, the router declines.
5573 7. Added expansion conditions match_domain, match_address, and match_local_
5574 part (NOT match_host).
5576 8. The placeholder for the Received: header didn't have a length field set.
5578 9. Added code to Exim itself and to exim_lock to test for a specific race
5579 condition that could lead to file corruption when using MBX delivery. The
5580 issue is with the lockfile that is created in /tmp. If this file is removed
5581 after a process has opened it but before that process has acquired a lock,
5582 there is the potential for a second process to recreate the file and also
5583 acquire a lock. This could lead to two Exim processes writing to the file
5584 at the same time. The added code performs the same test as UW imapd; it
5585 checks after acquiring the lock that its file descriptor still refers to
5586 the same named file.
5588 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5589 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5592 11. Added the smtp_active_hostname option. If used, this will typically be made
5593 to depend on the incoming interface address. Because $interface_address is
5594 not set up until the daemon has forked a reception process, error responses
5595 that can happen earlier (such as "too many connections") no longer contain
5598 12. If an expansion in a condition on a "warn" statement fails because a lookup
5599 defers, the "warn" statement is abandoned, and the next ACL statement is
5600 processed. Previously this caused the whole ACL to be aborted.
5602 13. Added the iplsearch lookup type.
5604 14. Added ident_timeout as a log selector.
5606 15. Added tls_certificate_verified as a log selector.
5608 16. Added a global option tls_require_ciphers (compare the smtp transport
5609 option of the same name). This controls incoming TLS connections.
5611 17. I finally figured out how to make tls_require_ciphers do a similar thing
5612 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5613 before starting the TLS session.
5615 18. Tabs are now shown as \t in -bP output.
5617 19. If the log selector return_path_on_delivery was set, Exim crashed when
5618 bouncing a message because it had too many Received: header lines.
5620 20. If two routers both had headers_remove settings, and the first one included
5621 a superfluous trailing colon, the final name in the first list and the
5622 first name in the second list were incorrectly joined into one item (with a
5623 colon in the middle).
5629 1. Added -C and -D options to the exinext utility, mainly to make it easier
5630 to include in the automated testing, but these could be helpful when
5631 multiple configurations are in use.
5633 2. The exinext utility was not formatting the output nicely when there was
5634 an alternate port involved in the retry record key, nor when there was a
5635 message id as well (for retries that were specific to a specific message
5636 and a specific host). It was also confused by IPv6 addresses, because of
5637 the additional colons they contain. I have fixed the IPv4 problem, and
5638 patched it up to do a reasonable job for IPv6.
5640 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5641 delivery, the log line now contains "pipelined" if PIPELINING was used.
5643 4. An SMTP transport process used to panic and die if the bind() call to set
5644 an explicit outgoing interface failed. This has been changed; it is now
5645 treated in the same way as a connect() failure.
5647 5. A reference to $sender_host_name in the part of a conditional expansion
5648 that was being skipped was still causing a DNS lookup. This no longer
5651 6. The def: expansion condition was not recognizing references to header lines
5652 that used bh_ and bheader_.
5654 7. Added the _cache feature to named lists.
5656 8. The code for checking quota_filecount in the appendfile transport was
5657 allowing one more file than it should have been.
5659 9. For compatibility with Sendmail, the command line option
5667 and sets the incoming protocol and host name (for trusted callers). The
5668 host name and its colon can be omitted when only the protocol is to be set.
5669 Note the Exim already has two private options, -pd and -ps, that refer to
5670 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5671 "s", but I don't think that's a major issue.
5673 10. A number of refactoring changes to the code, none of which should affect
5676 (a) The number of logging options was getting close to filling up the
5677 32-bit word that was used as a bit map. I have split them into two classes:
5678 those that are passed in the argument to log_write(), and those that are
5679 only ever tested independently outside of that function. These are now in
5680 separate 32-bit words, so there is plenty of room for expansion again.
5681 There is no change in the user interface or the logging behaviour.
5683 (b) When building, for example, log lines, the code previously used a
5684 macro that called string_cat() twice, in order to add two strings. This is
5685 not really sufficiently general. Furthermore, there was one instance where
5686 it was actually wrong because one of the argument was used twice, and in
5687 one call a function was used. (As it happened, calling the function twice
5688 did not affect the overall behaviour.) The macro has been replaced by a
5689 function that can join an arbitrary number of extra strings onto a growing
5692 (c) The code for expansion conditions now uses a table and a binary chop
5693 instead of a serial search (which was left over from when there were very
5694 few conditions). Also, it now recognizes conditions like "pam" even when
5695 the relevant support is not compiled in: a suitably worded error message is
5696 given if an attempt is made to use such a condition.
5698 11. Added ${time_interval:xxxxx}.
5700 12. A bug was causing one of the ddress fields not to be passed back correctly
5701 from remote delivery subprocesses. The field in question was not being
5702 subsequently used, so this caused to problems in practice.
5704 13. Added new log selectors queue_time and deliver_time.
5706 14. Might have fixed a bug in maildirsizefile handling that threw up
5707 "unexpected character" debug warnings, and recalculated the data
5708 unnecessarily. In any case, I expanded the warning message to give more
5711 15. Added the message "Restricted characters in address" to the statements in
5712 the default ACL that block characters like @ and % in local parts.
5714 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5715 Three changes have been made:
5717 (a) There was a serious bug; a negative response to MAIL caused the whole
5718 recipient domain to be cached as invalid, thereby blocking all messages
5719 to all local parts at the same domain, from all senders. This bug has
5720 been fixed. The domain is no longer cached after a negative response to
5721 MAIL if the sender used is not empty.
5723 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5726 (c) A new callout option, "use_sender" has been added for people who want
5727 the modified behaviour.
5733 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5736 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5737 indeed breaks things for older releases.
5739 3. Added additional logging to the case where there is a problem reading data
5740 from a filter that is running in a subprocess using a pipe, in order to
5741 try to track down a specific problem.
5743 4. Testing facility fudge: when running in the test harness and attempting
5744 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5745 getting "No route to host". Convert this to a timeout.
5747 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5750 6. Some OS don't have socklen_t but use size_t instead. This affects the
5751 fifth argument of getsockopt() amongst other things. This is now
5752 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5753 can be set for individual OS. I have set it for SunOS5, OSF1, and
5754 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5755 some earlier ones do not.
5757 7. Change 4.30/15 was not doing the test caselessly.
5759 8. The standard form for an IPv6 address literal was being rejected by address
5760 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5761 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5762 this, as well as the form without the "IPv6" on the front (but only when
5763 address literals are enabled, of course).
5765 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5767 10. Exim crashed if a message with an empty sender address specified by -f
5768 encountered a router with an errors_to setting. This could be provoked only
5769 by a command such as
5773 where an empty string was supplied; "<>" did not hit this bug.
5775 11. Installed PCRE release 4.5.
5777 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5778 remained set. It is now erased.
5780 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5781 times from message ids (which are base 36 rather than the normal 62).
5783 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5784 were being counted as actual protocol errors, and logged if the log
5785 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5786 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5787 and DATA following a set of rejected RCPTs do not count as protocol errors.
5788 In other words, Exim assumes they were pipelined, though this may not
5789 actually be the case. Of course, in all cases the client gets an
5790 appropriate error code.
5792 15. If a lookup fails in an ACL condition, a message about the failure may
5793 be available; it is used if testing the ACL cannot continue, because most
5794 such messages specify what the cause of the deferral is. However, some
5795 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5796 that caused an old message to be retained and used if a later statement
5797 caused a defer, replacing the real cause of the deferral.
5799 16. If an IP address had so many PTR records that the DNS lookup buffer
5800 was not large enough to hold them, Exim could crash while trying to process
5801 the truncated data. It now detects and logs this case.
5803 17. Further to 4.21/58, another change has been made: if (and only if) the
5804 first line of a message (the first header line) ends with CRLF, a bare LF
5805 in a subsequent header line has a space inserted after it, so as not to
5806 terminate the header.
5808 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5809 unnecessarily, used atoi() instead of strtol(), and didn't check the
5810 termination when getting file sizes from file names by regex.
5812 19. Completely re-implemented the support for maildirsize files, in the light
5813 of a number of problems with the previous contributed implementation
5814 (4.30/29). In particular:
5816 . If the quota is zero, the maildirsize file is maintained, but no quota is
5819 . If the maildir directory does not exist, it is created before any attempt
5820 to write a maildirsize file.
5822 . The quota value in the file is just a cache; if the quota is changed in
5823 the transport, the new value overrides.
5825 . A regular expression is available for excluding directories from the
5828 20. The autoreply transport checks the characters in options that define the
5829 message's headers; it allows continued headers, but it was checking with
5830 isspace() after an embedded newline instead of explicitly looking for a
5833 21. If all the "regular" hosts to which an address was routed had passed their
5834 expiry times, and had not reached their retry times, the address was
5835 bounced, even if fallback hosts were defined. Now Exim should go on to try
5838 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5839 equivalent code in the SMTP transport. Some hosts send humungous responses
5840 to HELO/EHLO, more than 1024 it seems.
5842 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5843 gives compiler warnings in some environments. I've now done it "properly",
5846 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5847 (because of problems with the built-in one) was declared to return uschar *
5848 instead of char *, causing compiler failure.
5850 25. Fixed a file descriptor leak when processing alias/forward files.
5852 26. Fixed a minor format string issue in dbfn.c.
5854 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5856 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5857 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5860 29. When a sender address is verified, it is cached, to save repeating the test
5861 when there is more than one recipient in a message. However, when the
5862 verification involves a callout, it is possible for different callout
5863 options to be set for different recipients. It is too complicated to keep
5864 track of this in the cache, so now Exim always runs a verification when a
5865 callout is required, relying on the callout cache for the optimization.
5866 The overhead is duplication of the address routing, but this should not be
5869 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5870 to be verified with callout=postmaster, and the main callout worked but the
5871 postmaster check failed, the verification correctly failed. However, if a
5872 subsequent RCPT command asked for sender verification *without* the
5873 postmaster check, incorrect caching caused this verification also to fail,
5876 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5877 it was not caching the DNS options (qualify_single, search_parents) that
5878 were used when the lookup failed. A subsequent lookup with different
5879 options therefore always gave the same answer, though there were cases
5880 where it should not have. (Example: a "domains = !$mx_any" option on a
5881 dnslookup router: the "domains" option is always processed without any
5882 widening, but the router might have qualify_single set.) Now Exim uses the
5883 cached value only when the same options are set.
5885 32. Added John Jetmore's "exipick" utility to the distribution.
5887 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5888 than a timeout (e.g. a certificate is required, and is not provided), an
5889 Exim server now closes the connection immediately. Previously it waited for
5890 the client to close - but if the client is SSL, it seems that they each
5891 wait for each other, leading to a delay before one of them times out.
5893 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5894 maintained 0.8.x compatibility because I don't think many are using it, and
5895 it is clearly obsolete.
5897 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5900 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5901 client certificate was expired. A simple patch fixes this, though I don't
5902 understand the full logic of why the verify callback is called multiple
5905 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5906 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5907 which causes problems with some clients (such as the Certicom SSL Plus
5908 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5909 disables the coutermeasure allowing Eudora to connect."
5911 38. Exim was not checking that a write() to a log file succeeded. This could
5912 lead to Bad Things if a log got too big, in particular if it hit a file
5913 size limit. Exim now panics and dies if it cannot write to a log file, just
5914 as it does if it cannot open a log file.
5916 39. Modified OS/Makefile-Linux so that it now contains
5918 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5920 The two -D definitions ensure that Exim is compiled with large file
5921 support, which makes it possible to handle log files that are bigger than
5924 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5925 instance) a domain was checked against a named list that involved a lookup,
5926 causing $domain_data to be set, then another domain was checked against the
5927 same list, then the first domain was re-checked, the value of $domain_data
5928 after the final check could be wrong. In particular, if the second check
5929 failed, it could be set empty. This bug probably also applied to
5932 41. The strip_trailing_dot option was not being applied to the address given
5933 with the -f command-line option.
5935 42. The code for reading a message's header from the spool was incrementing
5936 $received_count, but never initializing it. This meant that the value was
5937 incorrect (doubled) while delivering a message in the same process in which
5938 it was received. In the most common configuration of Exim, this never
5939 happens - a fresh exec is done - but it can happen when
5940 deliver_drop_privilege is set.
5942 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5943 it now includes up to 150 characters of the unexpected data in the log
5946 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5947 and building data strings. The size of both of these buffers was 10 000
5948 bytes - far larger than anybody would *ever* want, thought I. Needless to
5949 say, somebody hit the limit. I have increased the maximum line length to
5950 20 000 and the maximum data length of concatenated lines to 100 000. I have
5951 also fixed two bugs, because there was no checking on these buffers. Tsk,
5952 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5953 buffer is too small.
5955 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5956 lsearch lookups. Now it does.
5958 46. When parsing a route_list item in a manualroute router, a fixed-length
5959 buffer was used for the list of hosts. I made this 1024 bytes long,
5960 thinking that nobody would ever have a list of hosts that long. Wrong.
5961 Somebody had a whole pile of complicated expansion conditions, and the
5962 string was silently truncated, leading to an expansion error. It turns out
5963 that it is easier to change to an unlimited length (owing to other changes
5964 that have happened since this code was originally written) than to build
5965 structure for giving a limitation error. The length of the item that
5966 expands into the list of hosts is now unlimited.
5968 47. The lsearch lookup could not handle data where the length of text line was
5969 more than 4095 characters. Such lines were truncated, leading to shortened
5970 data being returned. It should now handle lines of any length.
5972 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5973 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5976 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5977 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5978 no longer happens in comment lines. A list of replacements is now placed
5979 at the head of all of the source files, except those whose only change is
5980 to replace PERL_COMMAND in the very first #! line.
5982 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5983 messages on the queue, with a bottom-up merge sort, using code contributed
5984 by Michael Haardt. This should make operations like -bp somewhat faster on
5985 large queues. It won't affect queue runners, except when queue_run_in_order
5988 51. Installed eximstats 1.31 in the distribution.
5990 52. Added support for SRV lookups to the dnslookup router.
5992 53. If an ACL referred to $message_body or $message_body_end, the value was not
5993 reset for any messages that followed in the same SMTP session.
5995 54. The store-handling optimization for building very long strings was not
5996 differentiating between the different store pools. I don't think this
5997 actually made any difference in practice, but I've tidied it.
5999 55. While running the routers to verify a sender address, $sender_address
6000 was still set to the sender address. This is wrong, because when routing to
6001 send a bounce to the sender, it would be empty. Therefore, I have changed
6002 it so that, while verifying a sender address, $sender_address is set to <>.
6003 (There is no change to what happens when verifying a recipient address.)
6005 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6006 target A or AAAA records (if not already returned) without resetting the
6007 qualify_single or search_parents options of the DNS resolver. These are
6008 inappropriate in this case because the targets of MX and SRV records must
6009 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6010 target that, when qualified, matched something in the local domain. These
6011 two options are now turned off when doing these lookups.
6013 57. It seems that at least some releases of Reiserfs (which does not have the
6014 concept of a fixed number of inodes) returns zero and not -1 for the
6015 number of available inodes. This interacted badly with check_spool_inodes,
6016 which assumed that -1 was the "no such thing" setting. What I have done is
6017 to check that the total number of inodes is greater than zero before doing
6018 the test of how many are available.
6020 58. When a "warn" ACL statement has a log_message modifier, the message is
6021 remembered, and not repeated. This is to avoid a lot of repetition when a
6022 message has many recipients that cause the same warning to be written.
6023 However, Exim was preserving the list of already written lines for an
6024 entire SMTP session, which doesn't seem right. The memory is now reset if a
6025 new message is started.
6027 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6028 debugging output unless log_rewrite was also set.
6030 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6031 of (int)(handle) when we know that handle contains (void *)(-1).
6033 61. The Exim daemon panic-logs an error return when it closes the incoming
6034 connection. However "connection reset by peer" seems to be common, and
6035 isn't really an error worthy of noting specially, so that particular error
6038 62. When Exim is trying to find all the local interfaces, it used to panic and
6039 die if the ioctl to get the interface flags failed. However, it seems that
6040 on at least one OS (Solaris 9) it is possible to have an interface that is
6041 included in the list of interfaces, but for which you get a failure error
6042 for this call. This happens when the interface is not "plumbed" into a
6043 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6044 failure of the "get flags" call assumes that the interface is down.
6046 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6047 makes life easier for people who are doing arithmetic on fields extracted
6048 from dates, where you often get leading zeros that should not be
6049 interpreted as octal.
6051 64. Added qualify_domain to the redirect router, to override the global
6054 65. If a pathologically long header line contained very many addresses (the
6055 report of this problem mentioned 10 000) and each of them was rewritten,
6056 Exim could use up a very large amount of memory. (It kept on making new
6057 copies of the header line as it rewrote, and never released the old ones.)
6058 At the expense of a bit more processing, the header rewriting function has
6059 been changed so that it no longer eats memory in this way.
6061 66. The generation of the Received: header has been moved from the time that a
6062 message starts to be received, to the time that it finishes. The timestamp
6063 in the Received: header should now be very close to that of the <= log
6064 line. There are two side-effects of this change:
6066 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6067 logged header lines no longer include the local Received: line, because
6068 it has not yet been created. The same applies to a copy of the message
6069 that is returned to a non-SMTP sender when a message is rejected.
6071 (b) When a filter file is tested using -bf, no additional Received: header
6072 is added to the test message. After some thought, I decided that this
6075 This change does not affect the value of $received_for. It is still set
6076 after address rewriting, but before local_scan() is called.
6078 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6080 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6081 gave an unhelpful panic error message, and a defer error. I have managed to
6082 change this behaviour so that it now rejects any supplied certificate,
6083 which seems right, as the list of acceptable certificates is empty.
6085 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6086 gave an unhelpful defer error. I have not managed to make this reject any
6087 supplied certificates, but the error message it gives is "no certificate
6088 supplied", which is not helpful.
6090 70. exigrep's output now also includes lines that are not associated with any
6091 message, but which match the given pattern. Implemented by a patch from
6092 Martin Sluka, which also tidied up the Perl a bit.
6094 71. Recipient callout verification, like sender verification, was using <> in
6095 the MAIL FROM command. This isn't really the right thing, since the actual
6096 sender may affect whether the remote host accepts the recipient or not. I
6097 have changed it to use the actual sender in the callout; this means that
6098 the cache record is now keyed on a recipient/sender pair, not just the
6099 recipient address. There doesn't seem to be a real danger of callout loops,
6100 since a callout by the remote host to check the sender would use <>.
6101 [SEE ABOVE: changed after hitting problems.]
6103 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6104 temporary errors. However, in the case of such a code being given after
6105 the end of a data transmission (i.e. after ".") Exim was failing to write
6106 a retry record for the message. (Yes, there was some broken host that was
6107 actually sending 8xx at this point.)
6109 73. An unknown lookup type in a host list could cause Exim to panic-die when
6110 the list was checked. (An example that provoked this was putting <; in the
6111 middle of a list instead of at the start.) If this happened during a DATA
6112 ACL check, a -D file could be left lying around. This kind of configuration
6113 error no longer causes Exim to die; instead it causes a defer error. The
6114 incident is still logged to the main and panic logs.
6116 74. Buglet left over from Exim 3 conversion. The message "too many messages
6117 in one connection" was written to the rejectlog but not the mainlog, except
6118 when address rewriting (yes!) was being logged.
6120 75. Added write_rejectlog option.
6122 76. When a system filter was run not as root (that is, when system_filter_user
6123 was set), the values of the $n variables were not being returned to the
6124 main process; thus, they were not subsequently available in the $sn
6127 77. Added +return_path_on_delivery log selector.
6129 78. A connection timeout was being treated differently from recipients deferred
6130 when testing hosts_max_try with a message that was older than the host's
6131 retry timeout. (The host should not be counted, thus allowing all hosts to
6132 be tried at least once before bouncing.) This may have been the cause of an
6133 occasionally reported bug whereby a message would remain on the queue
6134 longer than the retry timeout, but would be bounced if a delivery was
6135 forced. I say "may" because I never totally pinned down the problem;
6136 setting up timeout/retry tests is difficult. See also the next item.
6138 79. The ultimate address timeout was not being applied to errors that involved
6139 a combination of host plus message (for example, a timeout on a MAIL
6140 command). When an address resolved to a number of possible hosts, and they
6141 were not all tried for each delivery (e.g. because of hosts_max_try), a
6142 message could remain on the queue longer than the retry timeout.
6144 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6147 81. Fixed an obscure SMTP outgoing bug which required at least the following
6148 conditions: (a) there was another message waiting for the same server;
6149 (b) the server returned 5xx to all RCPT commands in the first message so
6150 that the message was not completed; (c) the server dropped the connection
6151 or gave a negative response to the RSET that Exim sends to abort the
6152 transaction. The observed case was a dropped connection after DATA that had
6153 been sent in pipelining mode. That is, the server had advertised PIPELINING
6154 but was not implementing it correctly. The effect of the bug was incorrect
6155 behaviour, such as trying another host, and this could lead to a crash.
6161 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6162 and daemon.c were passed as pointers to ints; they should have been
6163 pointers to socklen_t variables (which are typically unsigned ints).
6165 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6168 3. Fixed a really odd bug that affected only the testing scheme; patching a
6169 certain fixed string in the binary changed the value of another string that
6170 happened to be identical to the end of the original first string.
6172 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6173 name", it returns that address as the IP address. On some operating
6174 systems (e.g. Solaris), it also passes back the IP address string as the
6175 "host name". However, on others (e.g. Linux), it passes back an empty
6176 string. Exim wasn't checking for this, and was changing the host name to an
6177 empty string, assuming it had been canonicalized.
6179 5. Although rare, it is permitted to have more than one PTR record for a given
6180 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6181 all the names associated with an address, because they do in Solaris.
6182 However, it seems that they do not in Linux for data that comes from the
6183 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6184 I found this out when I moved to a new Linux workstation and tried to run
6185 the Exim test suite.
6187 To get round this problem I have changed the code so that it now does its
6188 own call to the DNS to look up PTR records when searching for a host name.
6189 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6190 addresses that are only in /etc/hosts are still found.
6192 This behaviour is, however, controlled by an option called host_lookup_
6193 order, which defaults to "bydns:byaddr". If people want to use the other
6194 order, or indeed, just use one or the other means of lookup, they can
6195 specify it in this variable.
6197 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6198 some operating systems, this comes back from gethostbyaddr() as an empty
6199 string, and this is what Exim used to test for. However, it seems that in
6200 other systems, "." is yielded. Exim now tests for this case too.
6202 7. The values of check_spool_space and check_log_space are now held internally
6203 as a number of kilobytes instead of an absolute number of bytes. If a
6204 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6205 kilobyte. This means that much larger values can be stored.
6207 8. Exim monitor: an attempt to get the action menu when not actually pointing
6208 at a message produces an empty menu entitled "No message selected". This
6209 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6210 no entries in it ("Shell widget menu has zero width and/or height"). So I
6211 have added a single, blank menu entry in this case.
6213 9. Added ${quote_local_part.
6215 10. MIME decoding is now applied to the contents of Subject: header lines when
6218 11. Now that a reference to $sender_host_address automatically causes a reverse
6219 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6220 host lookup before query-style lookups in lists that might use this
6221 variable. This has therefore been abolished, and the "net-" prefix is no
6222 longer necessary for query-style lookups.
6224 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6225 have been a typo for LFLAGS, so it has been changed.
6227 13. The install script calls Exim with "-C /dev/null" in order to find the
6228 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6229 to be output. However, since Exim outputs its version number before the
6230 error, it didn't break the script. It just looked ugly. I fixed this by
6231 always allowing "-C /dev/null" if the caller is root.
6233 14. Ignore overlarge ACL variable number when reading spool file - insurance
6234 against a later release with more variables having written the file.
6236 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6237 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6238 this, as well as the form without the "IPv6" on the front.
6240 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6241 OS/Makefile-Darwin file.
6243 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6244 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6246 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6247 variable. However, some OS use ints for inodes. I've added cast to long int
6248 to get rid of the compiler warning.
6250 19. I had forgotten to lock out "/../" in configuration file names when
6251 ALT_CONFIG_PREFIX was set.
6253 20. Routers used for verification do not need to specify transports. However,
6254 if such a router generated a host list, and callout was configured, Exim
6255 crashed, because it could not find a port number from the (non-existent)
6256 transport. It now assumes port 25 in this circumstance.
6258 21. Added the -t option to exigrep.
6260 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6261 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6262 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6263 them is set, LOOKUP_LSEARCH is forced.
6265 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6267 24. Added sender and host information to the "rejected by local_scan()" log
6268 line; previously there was no indication of these.
6270 25. Added .include_if_exists.
6272 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6273 Linux. It turns out that not all file systems support this. Apparently some
6274 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6275 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6276 not supported on the file descriptor, is now ignored when Exim is trying to
6277 sync a directory. This applies only to Linux.
6279 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6281 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6282 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6285 29. Added support for maildirsize files from supplied patch (modified a bit).
6287 30. The use of :fail: followed by an empty string could lead Exim to respond to
6288 sender verification failures with (e.g.):
6290 550 Verification failed for <xxx>
6291 550 Sender verify failed
6293 where the first response line was missing the '-' that indicates it is not
6294 the final line of the response.
6296 31. The loop for finding the name of the user that called Exim had a hardwired
6297 limit of 10; it now uses the value of finduser_retries, which is used for
6298 all other user lookups.
6300 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6303 33. Exim was neglecting to zero errno before one call of strtol() when
6304 expanding a string and expecting an integer value. On some systems this
6305 resulted in spurious "integer overflow" errors. Also, it was casting the
6306 result into an int without checking.
6308 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6309 did not work. The code looks as if it has *never* worked, though it appears
6310 to have been documented since at least release 1.62. I have made it work.
6312 35. The "timeout_DNS" error in retry rules, also documented since at least
6313 1.62, also never worked. As it isn't clear exactly what this means, and
6314 clearly it isn't a major issue, I have abolished the feature by treating it
6315 as "timeout", and writing a warning to the main and panic logs.
6317 36. The display of retry rules for -brt wasn't always showing the error code
6320 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6321 timeout_connect_A, timeout_connect_MX.
6323 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6324 to the empty sender.
6326 39. The daemon was not analysing the content of -oX till after it had closed
6327 stderr and disconnected from the controlling terminal. This meant that any
6328 syntax errors were only noted on the panic log, and the return code from
6329 the command was 0. By re-arranging the code a little, I've made the
6330 decoding happen first, so such errors now appear on stderr, and the return
6331 code is 1. However, the actual setting up of the sockets still happens in
6332 the disconnected process, so errors there are still only recorded on the
6335 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6336 connections (as happens on some IP stacks) was logged at start up time as
6337 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6338 it from "IPv6 and IPv4", which means that two separate sockets are being
6341 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6342 says whether AF_INET or AF_INET6 was passed as an argument.
6344 42. Exiwhat output was messed up when time zones were included in log
6347 43. Exiwhat now gives more information about the daemon's listening ports,
6348 and whether -tls-on-connect was used.
6350 44. The "port" option of the smtp transport is now expanded.
6352 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6353 silently ignored. Now an error message is written to the main and panic
6356 46. There's a new ACL modifier called "logwrite" which writes to a log file
6357 as soon as it is encountered.
6359 47. Added $local_user_uid and $local_user_gid at routing time.
6361 48. Exim crashed when trying to verify a sender address that was being
6364 49. Exim was recognizing only a space character after ".include". It now also
6365 recognizes a tab character.
6367 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6368 extracting the relevant information from the specification. The man page no
6369 longer contains scrambled data for the -d option, and I've added a section
6370 at the front about calling Exim under different names.
6372 51. Added "extra_headers" argument to the "mail" command in filter files.
6374 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6377 53. Installed eximstats 1.29.
6379 54. Added transport_filter_timeout as a generic transport option.
6381 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6382 Cc: header lines. This was required by RFC 822, but it not required by RFC
6385 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6386 incoming messages that did not have them. Now it does so only if the
6387 message originates locally, that is, if there is no associated remote host
6388 address. When Resent- header lines are present, this applies to the Resent-
6389 lines rather than the non-Resent- lines.
6391 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6392 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6394 58. Messages for configuration errors now include the name of the main
6395 configuration file - useful now that there may be more than one file in a
6396 list (.included file names were always shown).
6398 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6399 for those rare installations that do not start the daemon as root or run it
6400 setuid root. I've cut out the call to initgroups() if the daemon is not
6403 60. The Exim user and group can now be bound into the binary as text strings
6404 that are looked up at the start of Exim's processing.
6406 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6408 62. Added $mailstore_basename variable.
6410 63. Installed patch to sieve.c from Michael Haardt.
6412 64. When Exim failed to open the panic log after failing to open the main log,
6413 the original message it was trying to log was written to stderr and debug
6414 output, but if they were not available (the usual case in production), it
6415 was lost. Now it is written to syslog before the two lines that record the
6416 failures to open the logs.
6418 65. Users' Exim filters run in subprocesses under the user's uid. It is
6419 possible for a "deliver" command or an alias in a "personal" command to
6420 provoke an address rewrite. If logging of address rewriting is configured,
6421 this fails because the process is not running as root or exim. There may be
6422 a better way of dealing with this, but for the moment (because 4.30 needs
6423 to be released), I have disabled address rewrite logging when running a
6424 filter in a non-root, non-exim process.
6430 1. The buildconfig auxiliary program wasn't quoting the value set for
6431 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6432 not defined. This bug was masked in 4.22 by the effect that was fixed in
6435 2. Some messages that were rejected after a message id was allocated were
6436 shown as "incomplete" by exigrep. It no longer does this for messages that
6437 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6439 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6440 have allow_domain_literals set, the ID did not get logged in the <= line.
6441 Domain literals are now always recognized in Message-ID: header lines.
6443 4. The first argument for a ${extract expansion item is the key name or field
6444 number. Leading and trailing spaces in this item were not being ignored,
6445 causing some misleading effects.
6447 5. When deliver_drop_privilege was set, single queue runner processes started
6448 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6449 same command in the process it spins off) were not dropping privilege.
6451 6. When the daemon running as "exim" started a queue runner, it always
6452 re-executed Exim in the spun-off process. This is a waste of effort when
6453 deliver_drop_privilege is set. The new process now just calls the
6454 queue-runner function directly.
6460 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6463 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6464 ignored. Though the use of -oP was forcing the writing of a pid file, it
6465 was always written to the default place.
6467 3. If the message "no IP address found for host xxxx" is generated during
6468 incoming verification, it is now followed by identification of the incoming
6469 connection (so you can more easily find what provoked it).
6471 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6473 5. Added some features to "harden" Exim a bit more against certain attacks:
6475 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6476 be put in Local/Makefile. This is like the never_users runtime option,
6477 but it cannot be overridden. The default setting is "root".
6479 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6480 prefix string with which any file named in a -C command line option
6483 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6484 is retained for -C and -D only if the caller of Exim is root. Without
6485 it, the exim user may also use -C and -D and retain privilege.
6487 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6488 command line option is disabled.
6490 6. Macro names set by the -D option must start with an upper case letter, just
6491 like macro names defined in the configuration file.
6493 7. Added "dereference=" facility to LDAP.
6495 8. Two instances of the typo "uknown" in the source files are fixed.
6497 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6498 the Configure-Makefile script screwed up while processing it.
6500 10. Incorporated PCRE 4.4.
6502 11. The SMTP synchronization check was not operating right at the start of an
6503 SMTP session. For example, it could not catch a HELO sent before the client
6504 waited for the greeting. There is now a check for outstanding input at the
6505 point when the greeting is written. Because of the duplex, asynchronous
6506 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6507 way, but not yet received, when the check is performed.
6509 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6510 on TCP/IP sockets, because this apparently causes some broken clients to
6513 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6514 unchanged) from the Cygwin maintainer.
6516 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6517 was supported instead of testing for mbx. Effectively a typo.
6519 15. The spa authenticator server code was not checking that the input it
6520 received was valid base64.
6522 16. The debug output line for the "set" modifier in ACLs was not showing the
6523 name of the variable that was being set.
6525 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6527 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6528 lookup on its own. Something else was needed to trigger the lookup. For
6529 example, a match in host_lookup or the need for a host name in a host list.
6530 Now, if $sender_host_name is referenced and the host name has not yet been
6531 looked up, a lookup is performed. If the lookup fails, the variable remains
6532 empty, and $host_lookup_failed is set to "1".
6534 19. Added "eqi" as a case-independent comparison operator.
6536 20. The saslauthd authentication condition could segfault if neither service
6537 nor realm was specified.
6539 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6540 error message that was logged was misleading, and incoming SMTP
6541 connections were dropped. The message is now more accurate, and temporary
6542 errors are given to SMTP connections.
6544 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6545 (except RSET) with a 421 error, until QUIT is received. However, it was
6546 failing to send a response to QUIT.
6548 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6549 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6550 ACL is now run after the helo_try_verify_hosts code.
6552 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6553 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6554 case-independent, so other case variants are also recognized). Apparently
6555 some systems use these upper case variants.
6557 25. If more than two messages were waiting for the same host, and a transport
6558 filter was specified for the transport, Exim sent two messages over the
6559 same TCP/IP connection, and then failed with "socket operation on non-
6560 socket" when it tried to send the third.
6562 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6564 27. The extern definition of crypt16() in expand.c was not being excluded when
6565 the OS had its own crypt16() function.
6567 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6568 as a preferred synonym for return_size_limit, both as an option and as an
6571 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6573 30. Changed the default configuration ACL to relax the local part checking rule
6574 for addresses that are not in any local domains. For these addresses,
6575 slashes and pipe symbols are allowed within local parts, but the sequence
6576 /../ is explicitly forbidden.
6578 31. SPA server authentication was not clearing the challenge buffer before
6581 32. log_message in a "warn" ACL statement was writing to the reject log as
6582 well as to the main log, which contradicts the documentation and doesn't
6583 seem right (because no rejection is happening). So I have stopped it.
6585 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6586 However, I am unable to do any testing of this.
6588 34. Fixed an infelicity in the appendfile transport. When checking directories
6589 for a mailbox, to see if any needed to be created, it was accidentally
6590 using path names with one or more superfluous leading slashes; tracing
6591 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6593 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6594 contents are added to the log line that is written for every discarded
6595 recipient. (Previously a log_message setting was ignored.)
6597 36. The ${quote: operator now quotes the string if it is empty.
6599 37. The install script runs exim in order to find its version number. If for
6600 some reason other than non-existence or emptiness, which it checks, it
6601 could not run './exim', it was installing it with an empty version number,
6602 i.e. as "exim-". This error state is now caught, and the installation is
6605 38. An argument was missing from the function that creates an error message
6606 when Exim fails to connect to the socket for saslauthd authentication.
6607 This could cause Exim to crash, or give a corrupted message.
6609 39. Added isip, isip4, and isip6 to ${if conditions.
6611 40. The ACL variables $acl_xx are now saved with the message, and can be
6612 accessed later in routers, transports, and filters.
6614 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6615 strings in the file are not string-expanded.
6617 42. If a MAIL command specified a SIZE value that was too large to fit into an
6618 int variable, the check against message_size_limit failed. Such values are
6619 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6620 day this will have to be increased, but I don't think I want to be around
6621 when emails are that large.
6628 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6629 iconv() is not standard in FreeBSD.
6631 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6632 IPv6 enabled. The observed symptom was a segmentation fault on return from
6633 the function os_common_find_running_interfaces() in src/os.c.
6635 3. In the check_special_case() function in daemon.c I had used "errno" as an
6636 argument name, which causes warnings on some systems. This was basically a
6637 typo, since it was named "eno" in the comments!
6639 4. The code that waits for the clock to tick (at a resolution of some fraction
6640 of a second) so as to ensure message-id uniqueness was always waiting for
6641 at least one whole tick, when it could have waited for less. [This is
6642 almost certainly not relevant at current processor speeds, where it is
6643 unlikely to ever wait at all. But we try to future-proof.]
6645 5. The function that sleeps for a time interval that includes fractions of a
6646 second contained a race. It did not block SIGALRM between setting the
6647 timer, and suspending (a couple of lines later). If the interval was short
6648 and the sigsuspend() was delayed until after it had expired, the suspension
6649 never ended. On busy systems this could lead to processes getting stuck for
6652 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6653 process, before it forks any delivery processes. The open lookup caching
6654 mechanism meant that the open file or database connection was passed into
6655 the delivery process. The problem was that delivery processes always tidy
6656 up cached lookup data. This could cause a problem for the next delivery
6657 process started by the queue runner, because the external queue runner
6658 process does not know about the closure. So the next delivery process
6659 still has data in the lookup cache. In the case of a file lookup, there was
6660 no problem because closing a file descriptor in a subprocess doesn't affect
6661 the parent. However, if the lookup was caching a connection to a database,
6662 the connection was closed, and the second delivery process was likely to
6663 see errors such as "PGSQL: query failed: server closed the connection
6664 unexpectedly". The problem has been fixed by closing all cached lookups
6665 in a queue runner before running a delivery process.
6667 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6668 seem to have the "const" qualifier which it has on other OS. I've
6671 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6672 *of the same type* (client or server) with the same public name that an
6673 error should be diagnosed.
6675 9. When Exim looked up a host name for an IP address, but failed to find the
6676 original IP address when looking up the host name (a safety check), it
6677 output the message "<ip address> does not match any IP for NULL", which was
6678 confusing, to say the least. The bug was that the host name should have
6679 appeared instead of "NULL".
6681 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6682 user that is built into the binary, and the -C or -D options is used, root
6683 privilege is dropped before the configuration file is read. In addition,
6684 logging is switched to stderr instead of the normal log files. If the
6685 configuration then re-defines the Exim user, the unprivileged environment
6686 is probably not what is expected, so Exim logs a panic warning message (but
6689 However, if deliver_drop_privilege is set, the unprivileged state may well
6690 be exactly what is intended, so the warning has been cut out in that case,
6691 and Exim is allowed to try to write to its normal log files.
6697 1. smtp_return_error_details was not giving details for temporary sender
6698 or receiver verification errors.
6700 2. Diagnose a configuration error if two authenticators have the same public
6703 3. Exim used not to create the message log file for a message until the first
6704 delivery attempt. This could be confusing when incoming messages were held
6705 for policy or load reasons. The message log file is now created at the time
6706 the message is received, and an initial "Received" line is written to it.
6708 4. The automatically generated man page for command line options had a minor
6709 bug that caused no ill effects; however, a more serious problem was that
6710 the procedure for building the man page automatically didn't always
6711 operate. Consequently, release 4.20 contains an out-of-date version. This
6712 shouldn't happen again.
6714 5. When building Exim with embedded Perl support, the script that builds the
6715 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6716 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6718 6. The freeze_tell option was not being used for messages that were frozen on
6719 arrival, either by an ACL or by local_scan().
6721 7. Added the smtp_incomplete_transaction log selector.
6723 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6724 was accepting AUTH without a new EHLO.
6726 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6727 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6730 10. Logging of TCP/IP connections (when configured) now happens in the main
6731 daemon process instead of the child process, so that the TCP/IP connection
6732 count is more accurate (but it can never be perfect).
6734 11. The use of "drop" in a nested ACL was not being handled correctly in the
6735 outer ACL. Now, if condition failure induced by the nested "drop" causes
6736 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6737 or "require"), the connection is dropped.
6739 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6740 that yield "discard" can now be used with an "accept" or a "discard" verb,
6741 but an error is generated for any others (because I can't see a useful way
6742 to define what should happen).
6744 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6745 are now processed in the same way as lines in the Exim configuration file.
6746 In particular, continuation lines are supported.
6748 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6750 15. Added -ti meaning -t -i.
6752 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6753 domains, and warn by logging if others are found.
6755 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6756 returned from GIFCONF when Exim is trying to find the list of interfaces on
6757 a host. The code in os.c has been modified to copy each ifreq to an aligned
6758 structure in all cases.
6760 Also, in some cases, the returned ifreq's were being copied to a 'struct
6761 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6762 means the last couple of bytes of an IPv6 address could be chopped if the
6763 ifreq contained only a normal sockaddr (14 bytes storage).
6765 18. Named domain lists were not supported in the hosts_treat_as_local option.
6766 An entry such as +xxxx was not recognized, and was treated as a literal
6769 19. Ensure that header lines added by a DATA ACL are included in the reject log
6770 if the ACL subsequently rejects the message.
6772 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6773 MD5 (which is deprecated).
6775 21. When testing a filter file using -bf, Exim was writing a message when it
6776 took the sender from a "From " line in the message, but it was not doing so
6777 when it took $return_path from a Return-Path: header line. It now does.
6779 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6780 with a valid header line field name (a series of printing characters
6781 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6783 23. Changed "disc" in the source to "disk" to conform to the documentation and
6784 the book and for uniformity.
6786 24. Ignore Sendmail's -Ooption=value command line item.
6788 25. When execve() failed while trying to run a command in a pipe transport,
6789 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6790 could be confused with a return value of 69 from the command itself. This
6791 has been changed to 127, the value the shell returns if it is asked to run
6792 a non-existent command. The wording for the related log line suggests a
6793 non-existent command as the problem.
6795 26. If received_header_text expands to an empty string, do not add a Received:
6796 header line to the message. (Well, it adds a token one on the spool, but
6797 marks it "old" so that it doesn't get used or transmitted.)
6799 27. Installed eximstats 1.28 (addition of -nt option).
6801 28. There was no check for failure on the call to getsockname() in the daemon
6802 code. This can fail if there is a shortage of resources on the system, with
6803 ENOMEM, for example. A temporary error is now given on failure.
6805 29. Contrary to the C standard, it seems that in some environments, the
6806 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6807 program. Exim now does this explicitly; it affects the formatting of
6808 timestamps using strftime().
6810 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6811 complaints. I've now initialized all the variables, to avoid this.
6813 32. Header lines added by a system filter were not being "seen" during
6814 transport-time rewrites.
6816 33. The info_callback() function passed to OpenSSL is set up with type void
6817 (*)(SSL *, int, int), as described somewhere. However, when calling the
6818 function (actually a macro) that sets it up, the type void(*)() is
6819 expected. I've put in a cast to prevent warnings from picky compilers.
6821 34. If a DNS black list lookup found a CNAME record, but there were no A
6822 records associated with the domain it pointed at, Exim crashed.
6824 35. If a DNS black list lookup returned more than one A record, Exim ignored
6825 all but the first. It now scans all returned addresses if a particular IP
6826 value is being sought. In this situation, the contents of the
6827 $dnslist_value variable are a list of all the addresses, separated by a
6830 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6831 to accept a host name and all its aliases if the forward lookup for any of
6832 them yielded the IP address of the incoming connection. Now it accepts only
6833 those names whose forward lookup yields the correct IP address. Any other
6834 names are discarded. This closes a loophole whereby a rogue DNS
6835 administrator could create reverse DNS records to break through a
6836 wildcarded host restriction in an ACL.
6838 37. If a user filter or a system filter that ran in a subprocess used any of
6839 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6840 the wrong values were passed to the pipe command ($thisaddress had the
6841 value of $0, $0 had the value of $1, etc). This bug was introduced by
6842 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6844 38. Improved the line breaking for long SMTP error messages from ACLs.
6845 Previously, if there was no break point between 40 and 75 characters, Exim
6846 left the rest of the message alone. Two changes have been made: (a) I've
6847 reduced the minimum length to 35 characters; (b) if it can't find a break
6848 point between 35 and 75 characters, it looks ahead and uses the first one
6849 that it finds. This may give the occasional overlong line, but at least the
6850 remaining text gets split now.
6852 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6853 file descriptors might be low, and that setting 1000 would always raise it.
6854 It turns out that in some environments, the limit is already over 1000 and
6855 that lowering it causes trouble. So now Exim takes care not to decrease it.
6857 40. When delivering a message, the value of $return_path is set to $sender_
6858 address at the start of routing (routers may change the value). By an
6859 oversight, this default was not being set up when an address was tested by
6860 -bt or -bv, which affected the outcome if any router or filter referred to
6863 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6864 log only when "message" or "log_message" are set. However, if one of the
6865 conditions was an address verification, or a call to a nested ACL, the
6866 messages generated by the underlying test were being passed through. This
6867 no longer happens. The underlying message is available in $acl_verify_
6868 message for both "message" and "log_message" expansions, so it can be
6869 passed through if needed.
6871 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6872 new expansion $bh_ to give the encoded byte string without charset
6873 translation. Translation happens only if iconv() is available; HAVE_ICONV
6874 indicates this at build time. HEADERS_CHARSET gives the charset to
6875 translate to; headers_charset can change it in the configuration, and
6876 "headers charset" can change it in an individual filter file.
6878 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6879 that creates RFC 2047 encoded "words" labels them as that charset instead
6880 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6881 expansion operator; (ii) when Exim creates a From: line for a local
6882 message; (iii) when a header line is rewritten to include a "phrase" part.
6884 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6885 buggy, causing it to skip the first lines of messages whose message ID
6886 ended in 'D'. This would not have bitten before Exim release 4.14, because
6887 message IDs were unlikely to end in 'D' before then. The effect was to have
6888 incorrect size information for certain domains.
6890 45. #include "config.h" was missing at the start of the crypt16.c module. This
6891 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6894 46. If there was a timeout during a "random" callout check, Exim treated it as
6895 a failure of the random address, and carried on sending RSET and the real
6896 address. If the delay was just some slowness somewhere, the response to the
6897 original RCPT would be taken as a response to RSET and so on, causing
6898 mayhem of various kinds.
6900 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6901 when I implemented it. It didn't allow for the fact that some option values
6902 may legitimately be negative (e.g. size_addition), and it didn't even do
6903 the right test for positive values.
6905 48. Domain names in DNS records are case-independent. Exim always looks them up
6906 in lower case. Some resolvers return domain names in exactly the case they
6907 appear in the zone file, that is, they may contain uppercase letters. Not
6908 all resolvers do this - some return always lower case. Exim was treating a
6909 change of case by a resolver as a change of domain, similar to a widening
6910 of a domain abbreviation. This triggered its re-routing code and so it was
6911 trying to route what was effectively the same domain again. This normally
6912 caused routing to fail (because the router wouldn't handle the domain
6913 twice). Now Exim checks for this case specially, and just changes the
6914 casing of the domain that it ultimately uses when it transmits the message
6917 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6920 50. If a filter generated a file delivery with a non-absolute name (possible if
6921 no home directory exists for the router), the forbid_file option was not
6924 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6925 the existing equality matching.
6927 52. Exim was using ints instead of ino_t variables in some places where it was
6928 dealing with inode numbers.
6930 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6931 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6932 called TMPDIR, and if it finds it is different, it changes its value.
6934 54. The smtp_printf() function is now made available to local_scan() so
6935 additional output lines can be written before returning. There is also an
6936 smtp_fflush() function to enable the detection of a dropped connection.
6937 The variables smtp_input and smtp_batched_input are exported to
6940 55. Changed the default runtime configuration: the message "Unknown user"
6941 has been removed from the ACL, and instead placed on the localuser router,
6942 using the cannot_route_message feature. This means that any verification
6943 failures that generate their own messages won't get overridden. Similarly,
6944 the "Unrouteable address" message that was in the ACL for unverifiable
6945 relay addresses has also been removed.
6947 56. Added hosts_avoid_esmtp to the smtp transport.
6949 57. The exicyclog script was not checking for the esoteric option
6950 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6951 will work only if exicyclog is run under the appropriate euid.
6953 58. Following a discussion on the list, the rules by which Exim recognises line
6954 endings on incoming messages have been changed. The -dropcr and drop_cr
6955 options are now no-ops, retained only for backwards compatibility. The
6956 following line terminators are recognized: LF CRLF CR. However, special
6957 processing applies to CR:
6959 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6960 nor a local message in the state where . is a terminator.
6962 (ii) If a bare CR is encountered in a header line, an extra space is added
6963 after the line terminator so as not to end the header. The reasoning
6964 behind this is that bare CRs in header lines are most likely either
6965 to be mistakes, or people trying to play silly games.
6967 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6968 was being incorrectly given as 18 bytes larger than it should have been.
6969 This is a VOB (very old bug).
6971 60. This may never have affected anything current, but just in case it has:
6972 When the local host is found other than at the start of a list of hosts,
6973 the local host, those with the same MX, and any that follow, are discarded.
6974 When the list in question was part of a longer list of hosts, the following
6975 hosts (not currently being processed) were also being discarded. This no
6976 longer happens. I'm not sure if this situation could ever has previously
6979 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6982 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6983 header. This is something that is recommended in a new Internet Draft, and
6984 is something that is documented as being done by Sendmail. There are two
6985 possible values. For messages generated by the autoreply transport, Exim
6988 Auto-Submitted: auto-replied
6990 whereas for all other generated messages (e.g. bounces) it adds
6992 Auto-Submitted: auto-generated
6994 63. The "personal" condition in filters now includes a test for the
6995 Auto-Submitted: header. If it contains the string "auto-" the message it
6996 not considered personal.
6998 64. Added rcpt_include_affixes as a generic transport option.
7000 65. Added queue_only_override (default true).
7002 66. Added the syslog_duplication option.
7004 67. If what should have been the first header line of a message consisted of
7005 a space followed by a colon, Exim was mis-interpreting it as a header line.
7006 It isn't of course - it is syntactically invalid and should therefore be
7007 treated as the start of the message body. The misbehaviour could have
7008 caused a number of strange effects, including loss of data in subsequent
7009 header lines, and spool format errors.
7011 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7012 client host had authenticated. This control can now be exercised by an ACL
7013 for more flexibility.
7015 69. By default, callouts do not happen when testing with -bh. There is now a
7016 variant, -bhc, which does actually run the callout code, including
7017 consulting and updating the callout cache.
7019 70. Added support for saslauthd authentication, courtesy of Alexander
7022 71. If statvfs() failed on the spool or log directories while checking their
7023 size for availability, Exim confusingly gave the error "space shortage".
7024 Furthermore, in debugging mode it crashed with a floating point exception.
7025 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7026 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7027 serious problem, Exim now writes to the main and panic logs when this
7028 happens, with details of the failure. It then refuses to accept the
7029 incoming message, giving the message "spool directory problem" or "log
7030 directory problem" with a 421 code for SMTP messages.
7032 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7033 0, 1, and 2 exist, because some OS complain for execs without them (see
7034 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7035 descriptors. However, the code omitted to check that the open succeeded,
7036 causing mysterious errors if for some reason the permissions on /dev/null
7037 got screwed. Now Exim writes a message to the main and panic logs, and
7038 bombs out if it can't open /dev/null.
7040 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7041 interact so that it is all more flexible. It is supposed to remain
7042 backwards compatible. Also added extra_local_interfaces.
7044 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7045 to bomb out with an assertion failure - to the client this appears as a
7046 connection drop. This problem occurs in the part of the code that was taken
7047 from the Samba project. Fortunately, the assertion is in a very simple
7048 function, so I have fixed this by reproducing the function inline in the
7049 one place where it is called, and arranging for authentication to fail
7050 instead of killing the process with assert().
7052 75. The SPA client code was not working when the server requested OEM rather
7053 than Unicode encoding.
7055 76. Added code to make require_files with a specific uid setting more usable in
7056 the case where statting the file as root fails - usually a non-root-mounted
7057 NFS file system. When this happens and the failure is EACCES, Exim now
7058 forks a subprocess and does the per-uid checking as the relevant uid.
7060 77. Added process_log_path.
7062 78. If log_file_path was not explicitly set, a setting of check_log_space or
7063 check_log_inodes was ignored.
7065 79. If a space check for the spool or log partitions fails, the incident is now
7066 logged. Of course, in the latter case the data may get lost...
7068 80. Added the %p formatting code to string_format() so that it can be used to
7069 print addresses in debug_print(). Adjusted all the address printing in the
7070 debugging in store.c to use %p rather than %d.
7072 81. There was a concern that a line of code in smtp_in.c could overflow a
7073 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7074 initially expressed, the concern was not well-founded, because trailing
7075 spaces are removed early. However, if the trailing spaces were followed by
7076 a NULL, they did not get removed, so the overflow was possible. Two fixes
7079 (a) I re-wrote the offending code in a cleaner fashion.
7080 (b) If an incoming SMTP command contains a NULL character, it is rejected
7083 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7084 runs initgroups(), so that if the Exim user is in any additional groups,
7085 they will be used during message reception.
7091 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob