1 Change log file for Exim from version 4.21
2 ------------------------------------------
3 This document describes *changes* to previous versions, that might
4 affect Exim's operation, with an unchanged configuration file. For new
5 options, and new features, see the NewStuff file next to this ChangeLog.
8 Since Exim version 4.90
9 -----------------------
11 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
12 When redis_servers is set to a list of > 1 element, and the Redis servers
13 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
14 case of MOVED into a DEFER case instead, thus moving the query onto the
15 next server in the list. For a cluster of N elements, all N servers must
16 be defined in redis_servers.
18 JH/01 Replace the store_release() internal interface with store_newblock(),
19 which internalises the check required to safely use the old one, plus
20 the allocate and data copy operations duplicated in both (!) of the
23 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
24 modifier. This matches the restriction on the commandline.
26 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
27 Previously only the last row was returned.
29 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
30 we assumed that tags in the header were well-formed, and parsed the
31 element content after inspecting only the first char of the tag.
32 Assumptions at that stage could crash the receive process on malformed
35 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
36 While running the DKIM ACL we operate on the Permanent memory pool so that
37 variables created with "set" persist to the DATA ACL. Also (at any time)
38 DNS lookups that fail create cache records using the Permanent pool. But
39 expansions release any allocations made on the current pool - so a dnsdb
40 lookup expansion done in the DKIM ACL releases the memory used for the
41 DNS negative-cache, and bad things result. Solution is to switch to the
42 Main pool for expansions.
43 While we're in that code, add checks on the DNS cache during store_reset,
44 active in the testsuite.
45 Problem spotted, and debugging aided, by Wolfgang Breyha.
47 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
48 When none of the hosts presented to a transport match an already-open
49 connection, close it and proceed with the list. Previously we would
50 queue the message. Spotted by Lena with Yahoo, probably involving
53 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
54 Previously a spurious "250 OK id=" response was appended to the proper
57 JH/08 The "support for" informational output now, which built with Content
58 Scanning support, has a line for the malware scanner interfaces compiled
59 in. Interface can be individually included or not at build time.
61 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
62 by the template makefile "src/EDITME". The "STREAM" support for an older
63 ClamAV interface method is removed.
65 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
66 rows affected is given instead).
68 JH/11 The runtime Berkeley DB library version is now additionally output by
69 "exim -d -bV". Previously only the compile-time version was shown.
71 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
72 SMTP connection. Previously, when one had more receipients than the
73 first, an abortive onward connection was made. Move to full support for
74 multiple onward connections in sequence, handling cutthrough connection
75 for all multi-message initiating connections.
77 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
78 routers. Previously, a multi-recipient message would fail to match the
79 onward-connection opened for the first recipient, and cause its closure.
81 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
82 a timeout on read on a GnuTLS initiating connection, resulting in the
83 initiating connection being dropped. This mattered most when the callout
84 was marked defer_ok. Fix to keep the two timeout-detection methods
87 JH/15 Relax results from ACL control request to enable cutthrough, in
88 unsupported situations, from error to silently (except under debug)
89 ignoring. This covers use with PRDR, frozen messages, queue-only and
92 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
94 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
95 metadata, resulting in a crash in free().
97 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
98 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
99 Broken also in d185889f4, with init system revamp.
101 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
102 Previously we abruptly closed the connection after reading a malware-
103 found indication; now we go on to read the "scan ok" response line,
106 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
107 ACL. Previously, a crash would result.
109 JH/19 Speed up macro lookups during configuration file read, by skipping non-
110 macro text after a replacement (previously it was only once per line) and
111 by skipping builtin macros when searching for an uppercase lead character.
113 JH/20 DANE support moved from Experimental to mainline. The Makefile control
114 for the build is renamed.
116 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
117 was allocated for every new TLS startup, meaning one per message. Fix
118 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
120 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
121 reported the original. Fix to report (as far as possible) the ACL
122 result replacing the original.
124 JH/23 Fix memory leak during multi-message connections using STARTTLS under
125 OpenSSL. Certificate information is loaded for every new TLS startup,
126 and the resources needed to be freed.
128 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
130 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
133 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
134 DATA response info to the (existing) per-recipient response info for
135 the "C=" log element. It can have useful tracking info from the
136 destination system. Patch from Simon Arlott.
138 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
139 length value. Previously this would segfault.
141 HS/02 Support Avast multiline protoocol, this allows passing flags to
142 newer versions of the scanner.
144 JH/28 Ensure that variables possibly set during message acceptance are marked
145 dead before release of memory in the daemon loop. This stops complaints
146 about them when the debug_store option is enabled. Discovered specifically
147 for sender_rate_period, but applies to a whole set of variables.
148 Do the same for the queue-runner loop, for variables set from spool
149 message files. Do the same for the SMTP per-message loop, for certain
150 variables indirectly set in ACL operations.
152 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
153 as a multi-recipient message from a mailinglist manager). The coding had
154 an arbitrary cutoff number of characters while checking for more input;
155 enforced by writing a NUL into the buffer. This corrupted long / fast
156 input. The problem was exposed more widely when more pipelineing of SMTP
157 responses was introduced, and one Exim system was feeding another.
158 The symptom is log complaints of SMTP syntax error (NUL chars) on the
159 receiving system, and refused recipients seen by the sending system
160 (propating to people being dropped from mailing lists).
161 Discovered and pinpointed by David Carter.
163 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
164 replaced by the ${authresults } expansion.
170 JH/01 Rework error string handling in TLS interface so that the caller in
171 more cases is responsible for logging. This permits library-sourced
172 string to be attached to addresses during delivery, and collapses
173 pairs of long lines into single ones.
175 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
176 during configuration. Wildcards are allowed and expanded.
178 JH/02 Rework error string handling in DKIM to pass more info back to callers.
179 This permits better logging.
181 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
182 do not close it down and have the child transport start it up again on
183 the passed-on TCP connection. Instead, proxy the child (and any
184 subsequent ones) for TLS via a unix-domain socket channel. Logging is
185 affected: the continued delivery log lines do not have any DNSSEC, TLS
186 Certificate or OCSP information. TLS cipher information is still logged.
188 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
189 identical IP addresses on different listening ports. Will also affect
192 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
193 add noisy ifdef guards to special-case this sillyness.
194 Patch from Bernd Kuhls.
196 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
197 than 255 are no longer allowed.
199 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
200 Disable the session-cache too, which might reduce our load. Since we
201 currrectly use a new context for every connection, both as server and
202 client, there is no benefit for these.
203 GnuTLS appears to not support tickets server-side by default (we don't
204 call gnutls_session_ticket_enable_server()) but client side is enabled
205 by default on recent versions (3.1.3 +) unless the PFS priority string
208 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
209 <https://reproducible-builds.org/specs/source-date-epoch/>.
211 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
212 the check for any unsuccessful recipients did not notice the limit, and
213 erroneously found still-pending ones.
215 JH/08 Pipeline CHUNKING command and data together, on kernels that support
216 MSG_MORE. Only in-clear (not on TLS connections).
218 JH/09 Avoid using a temporary file during transport using dkim. Unless a
219 transport-filter is involved we can buffer the headers in memory for
220 creating the signature, and read the spool data file once for the
221 signature and again for transmission.
223 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
224 4.77 as the kernel support then wasn't solid, having issues in 64bit
225 mode. Now, it's been long enough. Add support for FreeBSD also.
227 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
228 case where the routing stage had gathered several addresses to send to
229 a host before calling the transport for the first, we previously failed
230 to close down TLS in the old transport process before passing the TCP
231 connection to the new process. The new one sent a STARTTLS command
232 which naturally failed, giving a failed delivery and bloating the retry
233 database. Investigation and fix prototype from Wolfgang Breyha.
235 JH/12 Fix check on SMTP command input synchronisation. Previously there were
236 false-negatives in the check that the sender had not preempted a response
237 or prompt from Exim (running as a server), due to that code's lack of
238 awareness of the SMTP input buffering.
240 PP/04 Add commandline_checks_require_admin option.
241 Exim drops privileges sanely, various checks such as -be aren't a
242 security problem, as long as you trust local users with access to their
243 own account. When invoked by services which pass untrusted data to
244 Exim, this might be an issue. Set this option in main configuration
245 AND make fixes to the calling application, such as using `--` to stop
248 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
249 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
250 responses to those, into a single TLS record each way (this usually means
251 a single packet). As a side issue, smtp_enforce_sync now works on TLS
254 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
255 affects you only if you're dancing at the edge of the param size limits.
256 If you are, and this message makes sense to you, then: raise the
257 configured limit or use OpenSSL 1.1. Nothing we can do for older
260 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
261 cmdline element to get the documented default one. Previously it was
264 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
265 are unsuable for non-message-specific reasons - eg. connection timeout,
266 banner-time rejection.
268 JH/16 Fix logging of delivery remote port, when specified by router, under
271 PP/06 Repair manualroute's ability to take options in any order, even if one
272 is the name of a transport.
275 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
277 JH/17 Change the list-building routines interface to use the expanding-string
278 triplet model, for better allocation and copying behaviour.
280 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
281 Previously it was constructed the first time a possibly-matching string
282 was met in the configuration file input during startup; now it is done
285 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
286 compatible one, to avoid the (poorly documented) possibility of a config
287 file in the working directory redirecting the DB files, possibly correpting
288 some existing file. CVE-2017-10140 assigned for BDB.
290 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
291 cache-hot. Previously, although the result was properly cached, the
292 initial verify call returned a defer.
294 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
295 the main verify for receipient in uncached-mode.
297 JH/22 Retire historical build files to an "unsupported" subdir. These are
298 defined as "ones for which we have no current evidence of testing".
300 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
301 if present. Previously it was ignored.
303 JH/24 Start using specified-initialisers in C structure init coding. This is
304 a C99 feature (it's 2017, so now considered safe).
306 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
307 if was a fixed-sized field and bitmask ops via macros; it is now more
310 PP/07 GitHub PR 56: Apply MariaDB build fix.
311 Patch provided by Jaroslav Škarvada.
313 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
314 during Coverity cleanups [4.87 JH/47]
315 Diagnosis and fix provided by Michael Fischer v. Mollard.
317 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
318 the right size to place the terminating semicolon on its own folded
319 line, the header hash was calculated to an incorrect value thanks to
320 the (relaxed) space the fold became.
322 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
323 and confused the parent.
325 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
326 which could crash as a result. This could lead to undeliverable messages.
328 JH/28 Logging: "next input sent too soon" now shows where input was truncated
331 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
332 matters on fast-turnover and PID-randomising systems, which were getting
333 out-of-order delivery.
335 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
336 a possibly-overlapping copy. The symptom was that "Remote host closed
337 connection in response to HELO" was logged instead of the actual 4xx
340 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
341 Previously only that bufferd was discarded, resulting in SYMTP command
344 JH/32 DKIM: when a message has multiple signatures matching an identity given
345 in dkim_verify_signers, run the dkim acl once for each. Previously only
346 one run was done. Bug 2189.
348 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
349 "panic the current process" to "deliberately defer". The panic log is
350 still written with the problem list name; the mail and reject logs now
351 get a temp-reject line for the message that was being handled, saying
352 something like "domains check lookup or other defer". The SMTP 451
353 message is still "Temporary local problem".
355 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
356 A crafted sequence of BDAT commands could result in in-use memory beeing
357 freed. CVE-2017-16943.
359 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
360 from SMTP input. Previously it was always done; now only done for DATA
361 and not BDAT commands. CVE-2017-16944.
363 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
364 to the message (such as an overlong header line). Previously this was
365 not done and we did not exit BDAT mode. Followon from the previous item
366 though a different problem.
372 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
373 than -2003 did; needs libidn2 in addition to libidn.
375 JH/02 The path option on a pipe transport is now expanded before use.
377 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
378 Patch provided by "Björn", documentation fix added too.
380 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
381 missing a wire-to-host endian conversion.
383 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
384 close after a BDAT command line could be taken as a following command,
385 giving a synch failure. Fix by only checking for synch immediately
386 before acknowledging the chunk.
388 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
389 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
390 macro. Patches provided by Josh Soref.
392 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
393 Previously we did not; the RFC seems ambiguous and VRFY is not listed
394 by IANA as a service extension. However, John Klensin suggests that we
397 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
398 the dkim code may be unix-mode line endings rather than smtp wire-format
399 CRLF, so prepend a CR to any bare LF.
401 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
402 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
404 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
405 the first were themselves being wrongly included in the feed into dkim
406 processing; with most chunk sizes in use this resulted in an incorrect
407 body hash calculated value.
409 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
410 DKIM signature block, for verification. Although advised against by
411 standards it is specifically not ruled illegal.
413 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
415 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
416 missing a body hash (the bh= tag).
418 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
419 It seems that HAProxy sends the Proxy Protocol information in clear and
420 only then does a TLS startup, so do the same.
422 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
423 TCP connections (such as for Spamd) unless the daemon successfully set
424 Fast Open mode on its listening sockets. This fixes breakage seen on
425 too-old kernels or those not configured for Fast Open, at the cost of
426 requiring both directions being enabled for TFO, and TFO never being used
427 by non-daemon-related Exim processes.
429 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
430 endings, at least on the first header line. Try to canonify any that get
431 past that check, despite the cost.
433 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
434 now limited to an arbitrary five deep, while parsing addresses with the
435 strip_excess_angle_brackets option enabled.
437 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
438 instead leave the unprompted TLS handshake in socket buffer for the
439 TLS library to consume.
441 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
443 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
445 JH/16 Drop variables when they go out of scope. Memory management drops a whole
446 region in one operation, for speed, and this leaves assigned pointers
447 dangling. Add checks run only under the testsuite which checks all
448 variables at a store-reset and panics on a dangling pointer; add code
449 explicitly nulling out all the variables discovered. Fixes one known
450 bug: a transport crash, where a dangling pointer for $sending_ip_address
451 originally assigned in a verify callout, is re-used.
453 PP/06 Drop '.' from @INC in various Perl scripts.
455 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
457 PP/08 Reduce a number of compilation warnings under clang; building with
458 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
459 should be warning-free.
461 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
463 HS/01 Fix portability problems introduced by PP/08 for platforms where
464 realloc(NULL) is not equivalent to malloc() [SunOS et al].
466 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
467 chunk. This allows us to accept broken chunked messages. We need a more
468 general solution here.
470 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
471 already-broken messages in the queue.
473 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
475 JH/19 Fix reference counting bug in routing-generated-address tracking.
481 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
482 supports it and a size is available (ie. the sending peer gave us one).
484 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
485 years of being deprecated). The replacements are the ACLs
486 acl_smtp_mime and acl_not_smtp_mime.
488 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
489 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
490 or even an in-clear connection were permitted. Now, if the host lookup
491 was dnssec and dane was requested then the host is only used if the
492 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
493 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
494 if one fails this test.
495 This means that a poorly-configured remote DNS will make it incommunicado;
496 but it protects against a DNS-interception attack on it.
498 JH/04 Bug 1810: make continued-use of an open smtp transport connection
499 non-noisy when a race steals the message being considered.
501 JH/05 If main configuration option tls_certificate is unset, generate a
502 self-signed certificate for inbound TLS connections.
504 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
505 in rewrites and routers.
507 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
508 and logged a warning sing 4.83; now they are a configuration file error.
510 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
511 (lacking @domain). Apply the same qualification processing as RCPT.
513 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
515 JH/10 Support ${sha256:} applied to a string (as well as the previous
518 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
519 a cutthrough deliver is pending, as we always want to make a connection.
520 This also avoids re-routing the message when later placing the cutthrough
521 connection after a verify cache hit.
522 Do not update it with the verify result either.
524 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
525 when routing results in more than one destination address.
527 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
528 signing (which inhibits the cutthrough capability). Previously only
529 the presence of an option was tested; now an expansion evaluating as
530 empty is permissible (obviously it should depend only on data available
531 when the cutthrough connection is made).
533 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
534 the relevant preceding SMTP command did not note the pipelining mode.
536 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
537 Previously they were not counted.
539 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
540 as one having no matching records. Previously we deferred the message
541 that needed the lookup.
543 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
544 distinguished as "(=".
546 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
547 for missing MX records. Previously it only worked for missing A records.
549 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
551 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
552 after the data-go-ahead and data-ack. Patch from Jason Betts.
554 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
555 even for a "none" policy. Patch from Tony Meyer.
557 JH/22 Fix continued use of a connection for further deliveries. If a port was
558 specified by a router, it must also match for the delivery to be
561 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
562 When one of the recipients of a message was unsuitable for the connection
563 (has no matching addresses), we lost track of needing to mark it
564 deferred. As a result mail would be lost.
566 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
568 JH/25 Decoding ACL controls is now done using a binary search; the source code
569 takes up less space and should be simpler to maintain. Merge the ACL
570 condition decode tables also, with similar effect.
572 JH/26 Fix problem with one_time used on a redirect router which returned the
573 parent address unchanged. A retry would see the parent address marked as
574 delivered, so not attempt the (identical) child. As a result mail would
577 JH/27 Fix a possible security hole, wherein a process operating with the Exim
578 UID can gain a root shell. Credit to http://www.halfdog.net/ for
579 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
582 JH/28 Enable {spool,log} filesystem space and inode checks as default.
583 Main config options check_{log,spool}_{inodes,space} are now
584 100 inodes, 10MB unless set otherwise in the configuration.
586 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
587 Previously it only applied to the main-section connection policy
590 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
592 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
593 by me. Added RFC7919 DH primes as an alternative.
595 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
596 are not in the system include path.
598 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
599 GnuTLS, when a session startup failed (eg because the client disconnected)
600 Exim did stdio operations after fclose. This was exposed by a recent
601 change which nulled out the file handle after the fclose.
603 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
604 signed directly by the cert-signing cert, rather than an intermediate
605 OCSP-signing cert. This is the model used by LetsEncrypt.
607 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
609 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
610 an incoming connection.
612 HS/02 Bug 1802: Do not half-close the connection after sending a request
615 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
616 fallback to "prime256v1".
618 JH/34 SECURITY: Use proper copy of DATA command in error message.
619 Could leak key material. Remotely exploitable. CVE-2016-9963.
625 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
626 and 3.4.4 - once the server is enabled to respond to an OCSP request
627 it does even when not requested, resulting in a stapling non-aware
628 client dropping the TLS connection.
630 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
631 support variable-length bit vectors. No functional change.
633 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
634 The I= interface field on outgoing lines is now after the H= remote
635 host field, same as incoming lines. There is a separate
636 outgoing_interface log selector which allows you to disable the
639 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
640 If not running log_selector +smtp_connection the mainlog would be held
641 open indefinitely after a "too many connections" event, including to a
642 deleted file after a log rotate. Leave the per net connection logging
643 leaving it open for efficiency as that will be quickly detected by the
644 check on the next write.
646 HS/01 Bug 1671: Fix post transport crash.
647 Processing the wait-<transport> messages could crash the delivery
648 process if the message IDs didn't exist for some reason. When
649 using 'split_spool_directory=yes' the construction of the spool
650 file name failed already, exposing the same netto behaviour.
652 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
653 mime_regex ACL conditions.
655 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
656 to DSN fail messages (bounces): remote IP, remote greeting, remote response
657 to HELO, local diagnostic string.
659 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
660 log line to debug. Even when configured with a tls authenticator many
661 client connections are expected to not authenticate in this way, so
662 an authenticate fail is not an error.
664 HS/02 Add the Exim version string to the process info. This way exiwhat
665 gives some more detail about the running daemon.
667 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
668 matter for fast-change records such as DNSBLs.
670 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
671 retry record, even if constant. There may be multiple transports with
672 different interface settings and the retry behaviour needs to be kept
675 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
677 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
679 JH/10 Bug 840: fix log_defer_output option of pipe transport
681 JH/11 Bug 830: use same host for all RCPTS of a message, even under
682 hosts_randomize. This matters a lot when combined with mua_wrapper.
684 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
685 ${quote_pgsql:<string>} operator.
687 JH/13 Bug 1708: avoid misaligned access in cached lookup.
689 JH/14 Change header file name for freeradius-client. Relevant if compiling
690 with Radius support; from the Gentoo tree and checked under Fedora.
692 JH/15 Bug 1712: Introduce $prdr_requested flag variable
694 JH/16 Bug 1714: Permit an empty string as expansion result for transport
695 option transport_filter, meaning no filtering.
697 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
699 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
700 defaults to "*" (all hosts). The variable is now available when not built
701 with TLS, default unset, mainly to enable keeping the testsuite sane.
702 If a server certificate is not supplied (via tls_certificate) an error is
703 logged, and clients will find TLS connections fail on startup. Presumably
704 they will retry in-clear.
705 Packagers of Exim are strongly encouraged to create a server certificate
706 at installation time.
708 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
709 with the $config_file variable.
711 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
712 in transport context, after the attempt, and per-recipient. The latter type
713 is per host attempted. The event data is the error message, and the errno
714 information encodes the lookup type (A vs. MX) used for the (first) host,
715 and the trailing two digits of the smtp 4xx response.
717 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
718 to write to mainlog (or rejectlog, paniclog) in the window between file
719 creation and permissions/ownership being changed. Particularly affects
720 installations where exicyclog is run as root, rather than exim user;
721 result is that the running daemon panics and dies.
723 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
725 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
726 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
727 "pri" and "weight". Note that the previous implicit priority given by the
728 list order is no longer honoured.
730 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
733 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
734 by defining SUPPORT_SOCKS.
736 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
737 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
738 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
739 variables are renamed to proxy_{local,external}_{address,port}.
741 JH/27 Move Internationalisation support from Experimental to mainline, enabled
742 for a build by defining SUPPORT_I18N
744 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
745 of the query string, and make ${quote_redis:} do that quoting.
747 JH/29 Move Events support from Experimental to mainline, enabled by default
748 and removable for a build by defining DISABLE_EVENT.
750 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
752 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
753 cached by the daemon.
755 JH/32 Move Redis support from Experimental to mainline, enabled for a build
756 by defining LOOKUP_REDIS. The libhiredis library is required.
758 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
759 keys are given for lookup.
761 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
762 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
763 only supported when built with TLS support. The PolarSSL SHA routines
764 are still used when the TLS library is too old for convenient support.
766 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
767 openssl_options), for security. OpenSSL forces this from version 1.1.0
768 server-side so match that on older versions.
770 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
771 allocation for $value could be released as the expansion processing
772 concluded, but leaving the global pointer active for it.
774 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
775 and to use the domains and local_parts ACL conditions.
777 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
778 incorrectly not doubled on cutthrough transmission, hence seen as a
779 body-termination at the receiving system - resulting in truncated mails.
780 Commonly the sender saw a TCP-level error, and retransmitted the message
781 via the normal store-and-forward channel. This could result in duplicates
782 received - but deduplicating mailstores were liable to retain only the
783 initial truncated version.
785 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
787 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
789 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
790 we're in there, support oversigning also; bug 1309.
792 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
794 HS/04 Add support for keep_environment and add_environment options.
796 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
797 either intentional arithmetic overflow during PRNG, or testing config-
800 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
801 delivery resulted in actual delivery. Cancel cutthrough before DATA
804 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
805 rejects a recipient: pass the reject to the originator.
807 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
808 Many were false-positives and ignorable, but it's worth fixing the
811 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
812 for the new environment-manipulation done at startup. Move the routines
813 from being local to tls.c to being global via the os.c file.
815 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
816 an extract embedded as result-arg for a map, the first arg for extract
817 is unavailable so we cannot tell if this is a numbered or keyed
818 extraction. Accept either.
824 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
827 JH/02 The smtp transport option "multi_domain" is now expanded.
829 JH/03 The smtp transport now requests PRDR by default, if the server offers
832 JH/04 Certificate name checking on server certificates, when exim is a client,
833 is now done by default. The transport option tls_verify_cert_hostnames
834 can be used to disable this per-host. The build option
835 EXPERIMENTAL_CERTNAMES is withdrawn.
837 JH/05 The value of the tls_verify_certificates smtp transport and main options
838 default to the word "system" to access the system default CA bundle.
839 For GnuTLS, only version 3.0.20 or later.
841 JH/06 Verification of the server certificate for a TLS connection is now tried
842 (but not required) by default. The verification status is now logged by
843 default, for both outbound TLS and client-certificate supplying inbound
846 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
849 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
850 Status Notification (bounce) messages are now MIME format per RFC 3464.
851 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
852 under the control of the dsn_advertise_hosts option, and routers may
853 have a dsn_lasthop option.
855 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
856 default, modifiable by a malware= option. The list separator for
857 the options can now be changed in the usual way. Bug 68.
859 JH/10 The smtp_receive_timeout main option is now expanded before use.
861 JH/11 The incoming_interface log option now also enables logging of the
862 local interface on delivery outgoing connections.
864 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
865 if the interface and destination host and port all match.
867 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
870 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
871 Patch from Andrew Lewis.
873 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
874 now supports optional time-restrictions, weighting, and priority
875 modifiers per server. Patch originally by <rommer@active.by>.
877 JH/16 The spamd_address main option now supports a mixed list of local
878 and remote servers. Remote servers can be IPv6 addresses, and
879 specify a port-range.
881 JH/17 Bug 68: The spamd_address main option now supports an optional
882 timeout value per server.
884 JH/18 Bug 1581: Router and transport options headers_add/remove can
885 now have the list separator specified.
887 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
890 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
893 JH/21 Support for the A6 type of dns record is withdrawn.
895 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
896 rather than the verbs used.
898 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
899 from 255 to 1024 chars.
901 JH/24 Verification callouts now attempt to use TLS by default.
903 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
904 are generic router options now. The defaults didn't change.
906 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
907 Original patch from Alexander Shikoff, worked over by JH.
909 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
910 files if ZCAT_COMMAND is not executable.
912 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
914 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
916 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
917 Normally benign, it bites when the pair was led to by a CNAME;
918 modern usage is to not canonicalize the domain to a CNAME target
919 (and we were inconsistent anyway for A-only vs AAAA+A).
921 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
923 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
924 when evaluating $sender_host_dnssec.
926 JH/31 Check the HELO verification lookup for DNSSEC, adding new
927 $sender_helo_dnssec variable.
929 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
931 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
933 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
935 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
936 documented as working, but never had. Support all but $spam_report.
938 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
939 added for tls authenticator.
941 HS/03 Add perl_taintmode main config option
947 TL/01 When running the test suite, the README says that variables such as
948 no_msglog_check are global and can be placed anywhere in a specific
949 test's script, however it was observed that placement needed to be near
950 the beginning for it to behave that way. Changed the runtest perl
951 script to read through the entire script once to detect and set these
952 variables, reset to the beginning of the script, and then run through
953 the script parsing/test process like normal.
955 TL/02 The BSD's have an arc4random API. One of the functions to induce
956 adding randomness was arc4random_stir(), but it has been removed in
957 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
958 function when detected.
960 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
961 cause callback expansion.
963 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
964 syntax errors in an expansion can be treated as a string instead of
965 logging or causing an error, due to the internal use of bool_lax
966 instead of bool when processing it.
968 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
969 server certificates when making smtp deliveries.
971 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
973 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
975 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
977 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
978 Merged patch from Sebastian Wiedenroth.
980 JH/05 Fix results-pipe from transport process. Several recipients, combined
981 with certificate use, exposed issues where response data items split
982 over buffer boundaries were not parsed properly. This eventually
983 resulted in duplicates being sent. This issue only became common enough
984 to notice due to the introduction of connection certificate information,
985 the item size being so much larger. Found and fixed by Wolfgang Breyha.
987 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
988 size buffer was used, resulting in syntax errors when an expansion
991 JH/07 Add support for directories of certificates when compiled with a GnuTLS
992 version 3.3.6 or later.
994 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
995 is EXPERIMENTAL_EVENT, the main-configuration and transport options
996 both become "event_action", the variables become $event_name, $event_data
997 and $event_defer_errno. There is a new variable $verify_mode, usable in
998 routers, transports and related events. The tls:cert event is now also
999 raised for inbound connections, if the main configuration event_action
1002 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1003 early OCSP support, but no stapling (appears to be less than 1.0.0).
1005 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1006 server certificate names available under the smtp transport option
1007 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1010 JH/10 Time-related extraction expansions from certificates now use the main
1011 option "timezone" setting for output formatting, and are consistent
1012 between OpenSSL and GnuTLS compilations. Bug 1541.
1014 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1015 encoded parameter in the incoming message. Bug 1558.
1017 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1018 include certificate info, eximon was claiming there were spoolfile
1021 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1023 JH/14 Log delivery-related information more consistently, using the sequence
1024 "H=<name> [<ip>]" wherever possible.
1026 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1027 are problematic for Debian distribution, omit them from the release
1030 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1032 JH/16 Fix string representation of time values on 64bit time_t architectures.
1035 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1036 output list separator was used.
1041 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1042 checkers that were complaining about end of non-void function with no
1045 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1046 This was a regression introduced in 4.83 by another bugfix.
1048 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1050 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1051 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1057 TF/01 Correctly close the server side of TLS when forking for delivery.
1059 When a message was received over SMTP with TLS, Exim failed to clear up
1060 the incoming connection properly after forking off the child process to
1061 deliver the message. In some situations the subsequent outgoing
1062 delivery connection happened to have the same fd number as the incoming
1063 connection previously had. Exim would try to use TLS and fail, logging
1064 a "Bad file descriptor" error.
1066 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1067 utilities have not been installed.
1069 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1070 temporary space as the ACL may create new global variables.
1072 TL/01 LDAP support uses per connection or global context settings, depending
1073 upon the detected version of the libraries at build time.
1075 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1076 to extract and use the src ip:port in logging and expansions as if it
1077 were a direct connection from the outside internet. PPv2 support was
1078 updated based on HAProxy spec change in May 2014.
1080 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1082 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1083 Properly escape header and check for NULL return.
1085 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1088 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1090 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1091 characters in header names, implemented as a verify condition.
1092 Contributed by Michael Fischer v. Mollard.
1094 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1095 results permerror and temperror. Previous values are deprecated but
1096 still accepted. In a future release, err_perm and err_temp will be
1097 completely removed, which will be a backward incompatibility if the
1098 ACL tests for either of these two old results. Patch contributed by
1099 user bes-internal on the mailing list.
1101 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1103 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1104 selectors, in both main and reject logs.
1106 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1109 JH/07 Add malware type "sock" for talking to simple daemon.
1111 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1113 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1114 routers/transports under cutthrough routing.
1116 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1117 numbers. Touch up "bool" conditional to keep the same definition.
1119 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1121 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1123 JH/12 Expand items in router/transport headers_add or headers_remove lists
1124 individually rather than the list as a whole. Bug 1452.
1126 Required for reasonable handling of multiple headers_ options when
1127 they may be empty; requires that headers_remove items with embedded
1128 colons must have them doubled (or the list-separator changed).
1130 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1131 view the policy declared in the DMARC record. Currently, $dmarc_status
1132 is a combined value of both the record presence and the result of the
1135 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1137 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1138 dnslookup router and the smtp transport (applying to the forward
1141 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1142 of ldap servers used for a specific lookup. Patch provided by Heiko
1145 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1146 New variable $lookup_dnssec_authenticated for observability.
1148 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1149 Patch submitted by Lars Timman.
1151 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1153 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1154 Requires trusted mode and valid format message id, aborts otherwise.
1155 Patch contributed by Heiko Schlichting.
1157 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1158 certextract with support for various fields. Bug 1358.
1160 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1161 is requested by default, modifiable by smtp transport option
1164 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1165 operate on certificate variables to give certificate fingerprints
1166 Also new ${sha256:cert_variable}.
1168 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1170 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1171 Christian Aistleitner.
1173 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1175 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1176 file. Patch from Wolfgang Breyha.
1178 JH/25 Expand the coverage of the delivery $host and $host_address to
1179 client authenticators run in verify callout. Bug 1476.
1181 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1182 align with daemon_smtp_ports. Bug 72.
1184 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1185 support and error reporting did not work properly.
1187 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1188 and is readable. Patch from Andrew Colin Kissa.
1190 TL/14 Enhance documentation of ${run expansion and how it parses the
1191 commandline after expansion, particularly in the case when an
1192 unquoted variable expansion results in an empty value.
1194 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1196 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1197 by renaming away from it. Addresses GH issue 15, reported by
1200 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1202 TL/15 SECURITY: prevent double expansion in math comparison functions
1203 (can expand unsanitized data). Not remotely exploitable.
1210 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1212 PP/02 Make -n do something, by making it not do something.
1213 When combined with -bP, the name of an option is not output.
1215 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1218 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1219 $sender_host_name and config options to manage this, and basic check
1222 PP/05 DSCP support for outbound connections and control modifier for inbound.
1224 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1225 (Only plugin which currently uses this is kerberos4, which nobody should
1226 be using, but we should make it available and other future plugins might
1227 conceivably use it, even though it would break NAT; stuff *should* be
1228 using channel bindings instead).
1230 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1231 name; added for Sendmail compatibility; requires admin caller.
1232 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1233 just ignore it); requires trusted caller.
1234 Also parse but ignore: -Ac -Am -X<logfile>
1237 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1239 TL/02 Add +smtp_confirmation as a default logging option.
1241 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1242 Patch by Magnus Holmgren from 2007-02-20.
1244 TL/04 Bugzilla 1281 - Spec typo.
1245 Bugzilla 1283 - Spec typo.
1246 Bugzilla 1290 - Spec grammar fixes.
1248 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1250 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1252 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1253 mailing list by by Dmitry Isaikin.
1255 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1257 JH/02 Support "G" suffix to numbers in ${if comparisons.
1259 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1261 NM/01 Bugzilla 1197 - Spec typo
1262 Bugzilla 1196 - Spec examples corrections
1264 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1266 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1267 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1270 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1271 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1273 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1274 "acl {{name}{arg}...}", and optional args on acl condition
1277 JH/05 Permit multiple router/transport headers_add/remove lines.
1279 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1281 JH/07 Avoid using a waiting database for a single-message-only transport.
1282 Performance patch from Paul Fisher. Bugzilla 1262.
1284 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1287 JH/09 Add $headers_added variable, with content from use of ACL modifier
1288 add_header (but not yet added to the message). Bugzilla 199.
1290 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1291 Pulled from Bugzilla 817 by Wolfgang Breyha.
1293 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1295 (nb: this is the same fix as in Exim 4.80.1)
1297 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1300 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1303 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1305 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1306 advertise SMTP AUTH mechanism to us, instead of a generic
1307 protocol violation error. Also, make Exim more robust to bad
1308 data from the Dovecot auth socket.
1310 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1312 When a queue runner is handling a message, Exim first routes the
1313 recipient addresses, during which it prunes them based on the retry
1314 hints database. After that it attempts to deliver the message to
1315 any remaining recipients. It then updates the hints database using
1318 So if a recipient address works intermittently, it can get repeatedly
1319 deferred at routing time. The retry hints record remains fresh so the
1320 address never reaches the final cutoff time.
1322 This is a fairly common occurrence when a user is bumping up against
1323 their storage quota. Exim had some logic in its local delivery code
1324 to deal with this. However it did not apply to per-recipient defers
1325 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1327 This change adds a proper retry rule check during routing so that the
1328 final cutoff time is checked against the message's age. We only do
1329 this check if there is an address retry record and there is not a
1330 domain retry record; this implies that previous attempts to handle
1331 the address had the retry_use_local_parts option turned on. We use
1332 this as an approximation for the destination being like a local
1333 delivery, as in LMTP.
1335 I suspect this new check makes the old local delivery cutoff check
1336 redundant, but I have not verified this so I left the code in place.
1338 TF/02 Correct gecos expansion when From: is a prefix of the username.
1340 Test 0254 submits a message to Exim with the header
1344 When I ran the test suite under the user fanf2, Exim expanded
1345 the header to contain my full name, whereas it should have added
1346 a Resent-Sender: header. It erroneously treats any prefix of the
1347 username as equal to the username.
1349 This change corrects that bug.
1351 GF/01 DCC debug and logging tidyup
1352 Error conditions log to paniclog rather than rejectlog.
1353 Debug lines prefixed by "DCC: " to remove any ambiguity.
1355 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1357 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1358 Bug spotted by Jeremy Harris; was flawed since initial commit.
1359 Would have resulted in OCSP responses post-SNI triggering an Exim
1360 NULL dereference and crash.
1362 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1364 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1365 Bug detection, analysis and fix by Samuel Thibault.
1366 Bugzilla 1331, Debian bug #698092.
1368 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1370 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1371 Server implementation by Todd Lyons, client by JH.
1372 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1373 config variable "prdr_enable" controls whether the server
1374 advertises the facility. If the client requests PRDR a new
1375 acl_data_smtp_prdr ACL is called once for each recipient, after
1376 the body content is received and before the acl_smtp_data ACL.
1377 The client is controlled by both of: a hosts_try_prdr option
1378 on the smtp transport, and the server advertisement.
1379 Default client logging of deliveries and rejections involving
1380 PRDR are flagged with the string "PRDR".
1382 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1383 fclose(). Diagnosis by Todd Lyons.
1385 PP/17 Update configure.default to handle IPv6 localhost better.
1386 Patch by Alain Williams (plus minor tweaks).
1389 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1390 This is now consistent with GnuTLS, and is now documented: the
1391 previous undocumented portable approach to treating the option as
1392 unset was to force an expansion failure. That still works, and
1393 an empty string is now equivalent.
1395 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1396 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1397 not performing validation itself.
1399 PP/20 Added force_command boolean option to pipe transport.
1400 Patch from Nick Koston, of cPanel Inc.
1402 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1405 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1407 PP/21 Fix eximon continuous updating with timestamped log-files.
1408 Broken in a format-string cleanup in 4.80, missed when I repaired the
1409 other false fix of the same issue.
1410 Report and fix from Heiko Schlichting.
1413 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1414 Report from Prashanth Katuri.
1416 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1417 It's SecureTransport, so affects any MacOS clients which use the
1418 system-integrated TLS libraries, including email clients.
1420 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1421 using a MIME ACL for non-SMTP local injection.
1422 Report and assistance in diagnosis by Warren Baker.
1424 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1426 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1428 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1429 last id that failed so it may be referenced in subsequent ACL's.
1431 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1434 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1435 ldap library initialization, allowing self-signed CA's to be
1436 used. Also properly sets require_cert option later in code by
1437 using NULL (global ldap config) instead of ldap handle (per
1438 session). Bug diagnosis and testing by alxgomz.
1440 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1441 the src/util/ subdirectory.
1443 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1444 renamed to Transport Post Delivery Action by Jeremy Harris, as
1447 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1448 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1449 redis_servers = needs to be configured which will be used by the redis
1450 lookup. Patch from Warren Baker, of The Packet Hub.
1452 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1454 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1455 hostname or reverse DNS when processing a host list. Used suggestions
1456 from multiple comments on this bug.
1458 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1460 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1461 Missed a few lines, added it to make the runtest require no keyboard
1464 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1465 contains upper case chars. Make router use caseful_local_part.
1467 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1468 support when GnuTLS has been built with p11-kit.
1474 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1476 This, or similar/improved, will also be change PP/11 of 4.82.
1482 PP/01 Handle short writes when writing local log-files.
1483 In practice, only affects FreeBSD (8 onwards).
1484 Bugzilla 1053, with thanks to Dmitry Isaikin.
1486 NM/01 Bugzilla 949 - Documentation tweak
1488 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1491 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1493 PP/02 Implemented gsasl authenticator.
1495 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1497 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1498 `pkg-config foo` for cflags/libs.
1500 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1501 with rest of GSASL and with heimdal_gssapi.
1503 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1504 `pkg-config foo` for cflags/libs for the TLS implementation.
1506 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1507 properties get this fed in as external SSF. A number of robustness
1508 and debugging improvements to the cyrus_sasl authenticator.
1510 PP/08 cyrus_sasl server now expands the server_realm option.
1512 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1513 Patch by Jeremy Harris.
1515 PP/10 Added dbmjz lookup type.
1517 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1519 PP/12 MAIL args handles TAB as well as SP, for better interop with
1520 non-compliant senders.
1521 Analysis and variant patch by Todd Lyons.
1523 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1524 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1525 Patch from Dirk Mueller <dmueller@suse.com>
1527 PP/13 tls_peerdn now print-escaped for spool files.
1528 Observed some $tls_peerdn in wild which contained \n, which resulted
1529 in spool file corruption.
1531 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1532 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1533 or write after TLS renegotiation, which otherwise led to messages
1536 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1537 as a tracking header (ie: a signed header comes before the signature).
1538 Patch from Wolfgang Breyha.
1540 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1541 comma-sep list; embedded commas doubled.
1543 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1545 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1547 Report and patch from Dmitry Banschikov.
1549 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1550 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1551 failures. We appear to now support TLS1.1+ with Exim.
1553 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1554 lets Exim select keys and certificates based upon TLS SNI from client.
1555 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1556 before an outbound SMTP session. New log_selector, +tls_sni.
1558 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1559 NULL dereference. Report and patch from Alun Jones.
1561 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1562 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1563 Not seeing resolver debug output on NetBSD, but suspect this is a
1564 resolver implementation change.
1566 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1567 Left warnings. Added "eximon gdb" invocation mode.
1569 PP/21 Defaulting "accept_8bitmime" to true, not false.
1571 PP/22 Added -bw for inetd wait mode support.
1573 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1574 locate the relevant includes and libraries. Made this the default.
1576 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1577 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1579 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1580 This may cause build issues on older platforms.
1582 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1583 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1584 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1585 Added SNI support via GnuTLS too.
1586 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1588 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1590 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1591 Applied second patch from Janne, implementing suggestion to default
1592 multiple-strings-in-record handling to match SPF spec.
1594 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1596 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1597 read-only, out of scope).
1598 Patch from Wolfgang Breyha, report from Stuart Northfield.
1600 PP/29 Fix three issues highlighted by clang analyser static analysis.
1601 Only crash-plausible issue would require the Cambridge-specific
1602 iplookup router and a misconfiguration.
1603 Report from Marcin Mirosław.
1605 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1607 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1608 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1609 As part of this, removing so much warning spew let me fix some minor
1610 real issues in debug logging.
1612 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1613 assignment on my part. Fixed.
1615 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1616 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1617 Janne Snabb (who went above and beyond: thank you).
1619 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1620 string otherwise requires a connection and a bunch more work and it's
1621 relatively easy to get wrong. Should also expose TLS library linkage
1624 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1625 64-bit ${eval} (JH/03).
1627 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1628 GNU libc to support some of the 64-bit stuff, should not lead to
1629 conflicts. Defined before os.h is pulled in, so if a given platform
1630 needs to override this, it can.
1632 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1633 protection layer was required, which is not implemented.
1634 Bugzilla 1254, patch from Wolfgang Breyha.
1636 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1637 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1638 tls_dhparam take prime identifiers. Also unbreak combination of
1639 OpenSSL+DH_params+TLSSNI.
1641 PP/39 Disable SSLv2 by default in OpenSSL support.
1647 PP/01 Solaris build fix for Oracle's LDAP libraries.
1648 Bugzilla 1109, patch from Stephen Usher.
1650 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1652 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1655 TF/02 Fix a couple more cases where we did not log the error message
1656 when unlink() failed. See also change 4.74-TF/03.
1658 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1659 lock up or crash if it happened to be inside a call to libc when it
1660 got a SIGUSR1 from exiwhat.
1662 The SIGUSR1 handler appends the current process status to the process
1663 log which is later printed by exiwhat. It used to use the general
1664 purpose logging code to do this, but several functions it calls are
1665 not safe for signals.
1667 The new output code in the SIGUSR1 handler is specific to the process
1668 log, and simple enough that it's easy to inspect for signal safety.
1669 Removing some special cases also simplifies the general logging code.
1670 Removing the spurious timestamps from the process log simplifies
1673 TF/04 Improved ratelimit ACL condition.
1675 The /noupdate option has been deprecated in favour of /readonly which
1676 has clearer semantics. The /leaky, /strict, and /readonly update modes
1677 are mutually exclusive. The update mode is no longer included in the
1678 database key; it just determines when the database is updated. (This
1679 means that when you upgrade Exim will forget old rate measurements.)
1681 Exim now checks that the per_* options are used with an update mode that
1682 makes sense for the current ACL. For example, when Exim is processing a
1683 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1684 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1685 must specify per_mail/readonly. If you omit the update mode it defaults to
1686 /leaky where that makes sense (as before) or /readonly where required.
1688 The /noupdate option is now undocumented but still supported for
1689 backwards compatibility. It is equivalent to /readonly except that in
1690 ACLs where /readonly is required you may specify /leaky/noupdate or
1691 /strict/noupdate which are treated the same as /readonly.
1693 A useful new feature is the /count= option. This is a generalization
1694 of the per_byte option, so that you can measure the throughput of other
1695 aggregate values. For example, the per_byte option is now equivalent
1696 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1698 The per_rcpt option has been generalized using the /count= mechanism
1699 (though it's more complicated than the per_byte equivalence). When it is
1700 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1701 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1702 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1703 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1704 Note that using per_rcpt with a non-readonly update mode in more than
1705 one ACL will cause the recipients to be double-counted. (The per_mail
1706 and per_byte options don't have this problem.)
1708 The handling of very low rates has changed slightly. If the computed rate
1709 is less than the event's count (usually one) then this event is the first
1710 after a long gap. In this case the rate is set to the same as this event's
1711 count, so that the first message of a spam run is counted properly.
1713 The major new feature is a mechanism for counting the rate of unique
1714 events. The new per_addr option counts the number of different
1715 recipients that someone has sent messages to in the last time period. It
1716 behaves like per_rcpt if all the recipient addresses are different, but
1717 duplicate recipient addresses do not increase the measured rate. Like
1718 the /count= option this is a general mechanism, so the per_addr option
1719 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1720 example, measure the rate that a client uses different sender addresses
1721 with the options per_mail/unique=$sender_address. There are further
1722 details in the main documentation.
1724 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1726 TF/06 Removed a few PCRE remnants.
1728 TF/07 Automatically extract Exim's version number from tags in the git
1729 repository when doing development or release builds.
1731 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1732 Bugzilla 879. Patch from Paul Fisher.
1734 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1735 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1738 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1740 PP/05 Variable $av_failed, true if the AV scanner deferred.
1741 Bugzilla 1078. Patch from John Horne.
1743 PP/06 Stop make process more reliably on build failure.
1744 Bugzilla 1087. Patch from Heiko Schlittermann.
1746 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1747 Bugzilla 1089. Patch from Heiko Schlittermann.
1749 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1750 Bugzilla 1131. Patch from Holger Weiß.
1752 PP/09 Handle IPv6 addresses with SPF.
1753 Bugzilla 860. Patch from Wolfgang Breyha.
1755 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1757 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1760 PP/11 match_* no longer expand right-hand-side by default.
1761 New compile-time build option, EXPAND_LISTMATCH_RHS.
1762 New expansion conditions, "inlist", "inlisti".
1764 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1766 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1768 PP/14 fix log_write() format string regression from TF/03.
1769 Bugzilla 1152. Patch from Dmitry Isaikin.
1775 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1777 PP/02 Harmonised TLS library version reporting; only show if debugging.
1778 Layout now matches that introduced for other libraries in 4.74 PP/03.
1780 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1782 PP/04 New "dns_use_edns0" global option.
1784 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1787 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1788 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1790 TK/01 Updated PolarSSL code to 0.14.2.
1791 Bugzilla 1097. Patch from Andreas Metzler.
1793 PP/07 Catch divide-by-zero in ${eval:...}.
1794 Fixes bugzilla 1102.
1796 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1799 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1800 format-string attack -- SECURITY: remote arbitrary code execution.
1802 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1803 time unintentionally subject to list matching rules, letting the header
1804 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1805 arbitrary string expansion). This allowed for information disclosure.
1807 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1808 INT_MIN/-1 -- value coerced to INT_MAX.
1814 NM/01 Workaround for PCRE version dependency in version reporting
1817 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1818 This fixes portability to compilers other than gcc, notably
1819 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1821 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1822 makefiles for portability to HP-UX and POSIX correctness.
1824 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1825 Also via indented variable definition in the Makefile.
1826 (Debugging by Oliver Heesakkers).
1828 PP/02 Restore caching of spamd results with expanded spamd_address.
1829 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1831 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1832 Improves build reliability. Fix from: Frank Elsner
1834 NM/02 Fix wide character breakage in the rfc2047 coding
1835 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1837 NM/03 Allow underscore in dnslist lookups
1838 Fixes bug 1026. Patch from Graeme Fowler
1840 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1841 Code patches from Adam Ciarcinski of NetBSD.
1843 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1846 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1847 is logged, to avoid truncation. Patch from John Horne.
1849 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1850 Patch from Jakob Hirsch.
1852 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1853 SQL string expansion failure details.
1854 Patch from Andrey Oktyabrski.
1856 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1857 Patch from Simon Arlott.
1859 PP/09 New lookups functionality failed to compile on old gcc which rejects
1860 extern declarations in function scope.
1861 Patch from Oliver Fleischmann
1863 PP/10 Use sig_atomic_t for flags set from signal handlers.
1864 Check getgroups() return and improve debugging.
1865 Fixed developed for diagnosis in bug 927 (which turned out to be
1868 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1869 Patch from Mark Zealey.
1871 PP/12 Bugzilla 1056: Improved spamd server selection.
1872 Patch from Mark Zealey.
1874 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1875 Based on patch from Heiko Schlittermann.
1877 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1878 Patch from Uwe Doering, sign-off by Michael Haardt.
1880 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1881 variable declaration deep within a block. Bug and patch from
1884 PP/15 lookups-Makefile IRIX compatibility coercion.
1886 PP/16 Make DISABLE_DKIM build knob functional.
1888 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1889 Patch by Simon Arlott
1891 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1892 variable argument macros. Our copy now differs from upstream.
1898 TF/01 Failure to get a lock on a hints database can have serious
1899 consequences so log it to the panic log.
1901 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1902 controlled using the smtp_confirmation log selector.
1904 TF/03 Include the error message when we fail to unlink a spool file.
1906 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1907 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1908 for maintaining out-of-tree patches for some time.
1910 PP/01 Bugzilla 139: Documentation and portability issues.
1911 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1912 Handle per-OS dynamic-module compilation flags.
1914 PP/02 Let /dev/null have normal permissions.
1915 The 4.73 fixes were a little too stringent and complained about the
1916 permissions on /dev/null. Exempt it from some checks.
1917 Reported by Andreas M. Kirchwitz.
1919 PP/03 Report version information for many libraries, including
1920 Exim version information for dynamically loaded libraries. Created
1921 version.h, now support a version extension string for distributors
1922 who patch heavily. Dynamic module ABI change.
1924 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1925 privilege escalation vulnerability whereby the Exim run-time user
1926 can cause root to append content of the attacker's choosing to
1929 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1932 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1933 If dropping privileges for untrusted macros, we disabled normal logging
1934 on the basis that it would fail; for the Exim run-time user, this is not
1935 the case, and it resulted in successful deliveries going unlogged.
1936 Fixed. Reported by Andreas Metzler.
1942 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1943 only prepend for the Resent-* case. Fixes regression introduced in
1944 Exim 4.70 by NM/22 for Bugzilla 607.
1946 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1947 increasing numbers of administrators be bitten by this.
1949 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1951 PP/03 Bugzilla 994: added openssl_options main configuration option.
1953 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1955 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1957 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1959 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1960 without a peer certificate, leading to a segfault because of an
1961 assumption that peers always have certificates. Be a little more
1962 paranoid. Problem reported by Martin Tscholak.
1964 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1965 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1966 NB: ClamAV planning to remove STREAM in "middle of 2010".
1967 CL also introduces -bmalware, various -d+acl logging additions and
1968 more caution in buffer sizes.
1970 PP/09 Implemented reverse_ip expansion operator.
1972 PP/10 Bugzilla 937: provide a "debug" ACL control.
1974 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1976 PP/12 Bugzilla 973: Implement --version.
1978 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1980 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1982 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1984 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1985 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1986 ignore trailing whitespace.
1988 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1990 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1993 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1994 Notification from Dr Andrew Aitchison.
1996 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1997 ExtendedDetectionInfo response format.
1998 Notification from John Horne.
2000 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2003 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2004 XSL and documented dependency on system catalogs, with examples of how
2007 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2010 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2011 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2012 configuration file which is writeable by the Exim user or group.
2014 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2015 of configuration files to cover files specified with the -C option if
2016 they are going to be used with root privileges, not just the default
2019 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2020 option (effectively making it always true).
2022 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2023 files to be used while preserving root privileges.
2025 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2026 that rogue child processes cannot use them.
2028 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2029 run-time user, instead of root.
2031 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2032 Exim run-time user without dropping privileges.
2034 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2035 result string, instead of calling string_vformat() twice with the same
2038 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2039 for other users. Others should always drop root privileges if they use
2040 -C on the command line, even for a whitelisted configure file.
2042 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2044 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2050 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2051 $data_path, and $header_path variables; fixed documentation bugs and
2054 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2055 exipick to access non-standard spools, including the "frozen" queue
2058 NM/01 Bugzilla 965: Support mysql stored procedures.
2059 Patch from Alain Williams
2061 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2063 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2064 Patch from Andreas Metzler
2066 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2067 Patch from Kirill Miazine
2069 NM/05 Bugzilla 671: Added umask to procmail example.
2071 JJ/03 installed exipick 20100323.0, fixing doc bug
2073 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2074 directory. Notification and patch from Dan Rosenberg.
2076 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2078 TK/02 Improve log output when DKIM signing operation fails.
2080 MH/01 Treat the transport option dkim_domain as a colon separated
2081 list, not as a single string, and sign the message with each element,
2082 omitting multiple occurences of the same signer.
2084 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2085 Bugzilla 985, 986. Patch by Simon Arlott
2087 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2088 Patch by Simon Arlott
2090 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2091 MBX locking. Notification from Dan Rosenberg.
2097 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2099 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2101 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2103 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2105 NM/04 Bugzilla 915: Flag broken perl installation during build.
2111 TK/01 Added patch by Johannes Berg that expands the main option
2112 "spamd_address" if it starts with a dollar sign.
2114 TK/02 Write list of recipients to X-Envelope-Sender header when building
2115 the mbox-format spool file for content scanning (suggested by Jakob
2118 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2119 (http://www.dcc-servers.net/) support via dccifd. Activated by
2120 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2122 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2123 by Mark Daniel Reidel <mr@df.eu>.
2125 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2126 When building exim an external PCRE library is now needed -
2127 PCRE is a system library on the majority of modern systems.
2128 See entry on PCRE_LIBS in EDITME file.
2130 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2131 conversation. Added nologin parameter to request.
2132 Patch contributed by Kirill Miazine.
2134 TF/01 Do not log submission mode rewrites if they do not change the address.
2136 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2138 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2139 log files in place. Contributed by Roberto Lima.
2141 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2143 TF/03 Bugzilla 615: When checking the local_parts router precondition
2144 after a local_part_suffix or local_part_prefix option, Exim now
2145 does not use the address's named list lookup cache, since this
2146 contains cached lookups for the whole local part.
2148 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2149 Robert Millan. Documentation is in experimental-spec.txt.
2151 TF/04 Bugzilla 668: Fix parallel build (make -j).
2153 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2155 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2156 Patch provided by Jan Srzednicki.
2158 TF/05 Leading white space used to be stripped from $spam_report which
2159 wrecked the formatting. Now it is preserved.
2161 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2162 that they are available at delivery time.
2164 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2166 TF/08 TLS error reporting now respects the incoming_interface and
2167 incoming_port log selectors.
2169 TF/09 Produce a more useful error message if an SMTP transport's hosts
2170 setting expands to an empty string.
2172 NM/06 Bugzilla 744: EXPN did not work under TLS.
2173 Patch provided by Phil Pennock.
2175 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2176 Patch provided by Richard Godbee.
2178 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2179 acl_smtp_notquit, added index entry.
2181 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2182 Patch provided by Eugene Bujak.
2184 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2185 accept(). Patch provided by Maxim Dounin.
2187 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2188 Patch provided by Phil Pennock.
2190 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2192 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2193 Patch provided by Brad "anomie" Jorsch.
2195 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2196 Patch provided by Dean Brooks.
2198 TK/05 Add native DKIM support (does not depend on external libraries).
2200 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2201 Patch provided by Graeme Fowler.
2203 NM/16 Bugzilla 851: Documentation example syntax fix.
2205 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2207 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2210 NM/19 Bugzilla 745: TLS version reporting.
2211 Patch provided by Phil Pennock.
2213 NM/20 Bugzilla 167: bool: condition support.
2214 Patch provided by Phil Pennock.
2216 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2217 clients. Patch provided by Phil Pennock.
2219 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2220 Patch provided by Brad "anomie" Jorsch.
2222 NM/23 Bugzilla 687: Fix misparses in eximstats.
2223 Patch provided by Heiko Schlittermann.
2225 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2226 Patch provided by Heiko Schlittermann.
2228 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2229 plus update to original patch.
2231 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2233 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2234 Patch provided by David Brownlee.
2236 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2238 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2240 NM/30 Bugzilla 888: TLS documentation bugfixes.
2242 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2244 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2245 Unlike the original bugzilla I have changed all shell scripts in src tree.
2247 NM/33 Bugzilla 898: Transport filter timeout fix.
2248 Patch by Todd Rinaldo.
2250 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2251 Patch by Serge Demonchaux.
2253 NM/35 Bugzilla 39: Base64 decode bug fixes.
2254 Patch by Jakob Hirsch.
2256 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2258 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2260 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2262 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2268 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2269 ALT-N's libdkim that I have put here:
2270 http://duncanthrax.net/exim-experimental/
2272 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2273 were called 'true' and it seems that C99 defines that as a reserved
2274 keyword to be used with 'bool' variable types. That means you could
2275 not include C99-style headers which use bools without triggering
2276 build errors in sieve.c.
2278 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2279 as mailq or other aliases. Changed the --help handling significantly
2280 to do whats expected. exim_usage() emits usage/help information.
2282 SC/01 Added the -bylocaldomain option to eximstats.
2284 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2286 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2288 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2294 PH/01 Another patch from the Sieve maintainer.
2296 PH/02 When an IPv6 address is converted to a string for single-key lookup
2297 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2298 dots are used instead of colons so that keys in lsearch files need not
2299 contain colons. This was done some time before quoting was made available
2300 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2301 (notated using the quote facility) so as to distinguish them from IPv4
2302 keys. This meant that lookups for IP addresses in host lists did not work
2303 for iplsearch lookups.
2305 This has been fixed by arranging for IPv6 addresses to be expressed with
2306 colons if the lookup type is iplsearch. This is not incompatible, because
2307 previously such lookups could never work.
2309 The situation is now rather anomalous, since one *can* have colons in
2310 ordinary lsearch keys. However, making the change in all cases is
2311 incompatible and would probably break a number of configurations.
2313 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2316 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2317 conversion specification without a maximum field width, thereby enabling
2318 a rogue spamd server to cause a buffer overflow. While nobody in their
2319 right mind would setup Exim to query an untrusted spamd server, an
2320 attacker that gains access to a server running spamd could potentially
2321 exploit this vulnerability to run arbitrary code as the Exim user.
2323 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2324 $primary_hostname instead of what libspf2 thinks the hosts name is.
2326 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2327 a directory entry by the name of the lookup key. Previously, if a
2328 symlink pointed to a non-existing file or a file in a directory that
2329 Exim lacked permissions to read, a lookup for a key matching that
2330 symlink would fail. Now it is enough that a matching directory entry
2331 exists, symlink or not. (Bugzilla 503.)
2333 PH/03 The body_linecount and body_zerocount variables are now exported in the
2336 PH/04 Added the $dnslist_matched variable.
2338 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2339 This means they are set thereafter only if the connection becomes
2342 PH/06 Added the client_condition to authenticators so that some can be skipped
2343 by clients under certain conditions.
2345 PH/07 The error message for a badly-placed control=no_multiline_responses left
2346 "_responses" off the end of the name.
2348 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2350 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2351 (without spaces) instead of just copying the configuration text.
2353 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2355 PH/11 Added $max_received_linelength.
2357 PH/12 Added +ignore_defer and +include_defer to host lists.
2359 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2360 way in which PCRE > 7.0 is built.
2362 PH/14 Implemented queue_only_load_latch.
2364 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2365 MAIL command. The effect was to mangle the value on 64-bit systems.
2367 PH/16 Another patch from the Sieve maintainer.
2369 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2371 PH/18 If a system quota error occurred while trying to create the file for
2372 a maildir delivery, the message "Mailbox is full" was not appended to the
2373 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2374 only to a quota excession during the actual writing of the file.
2376 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2377 characters?) which causes problems in log lines. The DN values are now
2378 passed through string_printing() before being added to log lines.
2380 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2381 and InterBase are left for another time.)
2383 PH/21 Added message_body_newlines option.
2385 PH/22 Guard against possible overflow in moan_check_errorcopy().
2387 PH/23 POSIX allows open() to be a macro; guard against that.
2389 PH/24 If the recipient of an error message contained an @ in the local part
2390 (suitably quoted, of course), incorrect values were put in $domain and
2391 $local_part during the evaluation of errors_copy.
2397 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2398 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2401 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2402 issue a MAIL command.
2404 PH/02 In an ACL statement such as
2406 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2408 if a client was not listed at all, or was listed with a value other than
2409 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2410 the condition was not true (as it should be), so access was not denied.
2411 The bug was that the ! inversion was incorrectly passed on to the second
2412 item. This has been fixed.
2414 PH/03 Added additional dnslists conditions == and =& which are different from
2415 = and & when the dns lookup returns more than one IP address.
2417 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2418 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2420 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2421 FSYNC, which compiles an option called disable_fsync that allows for
2422 bypassing fsync(). The documentation is heavily laced with warnings.
2424 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2426 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2427 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2428 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2429 including adding "make clean"; (3) Added -fPIC when compiling the test
2430 dynamically loaded module, to get rid of a warning.
2432 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2433 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2434 = 0s. The bug is otherwise harmless.
2436 PH/07 There was a bug in the dovecot authenticator such that the value of
2437 $auth1 could be overwritten, and so not correctly preserved, after a
2438 successful authentication. This usually meant that the value preserved by
2439 the server_setid option was incorrect.
2441 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2443 PH/09 Installed PCRE release 7.0.
2445 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2446 run for batched SMTP input. It is now run at the start of every message
2447 in the batch. While fixing this I discovered that the process information
2448 (output by running exiwhat) was not always getting set for -bs and -bS
2449 input. This is fixed, and it now also says "batched" for BSMTP.
2451 PH/11 Added control=no_pipelining.
2453 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2454 patch, slightly modified), and move the expansion of helo_data till after
2455 the connection is made in the smtp transport (so it can use these
2458 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2460 PH/14 Added log_selector = +pid.
2462 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2464 PH/16 Add ${if forany and ${if forall.
2466 PH/17 Added dsn_from option to vary the From: line in DSNs.
2468 PH/18 Flush SMTP output before performing a callout, unless control =
2469 no_callout_flush is set.
2471 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2472 was true (the default) a successful delivery failed to delete the retry
2473 item, thus causing premature timeout of the address. The bug is now
2476 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2478 PH/21 Long custom messages for fakedefer and fakereject are now split up
2479 into multiline responses in the same way that messages for "deny" and
2480 other ACL rejections are.
2482 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2483 with slight modification.
2485 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2486 draft, changing the syntax and factoring some duplicate code".
2488 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2489 for deliveries of the second and subsequent messages over the same SMTP
2492 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2493 ${reduce, with only minor "tidies".
2495 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2497 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2498 expansion side effects.
2500 PH/27 When a message times out after an over-quota error from an Exim-imposed
2501 quota, the bounce message says "mailbox is full". This message was not
2502 being given when it was a system quota that was exceeded. It now should
2505 MH/03 Made $recipients available in local_scan(). local_scan() already has
2506 better access to the recipient list through recipients_list[], but
2507 $recipients can be useful in postmaster-provided expansion strings.
2509 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2510 in the case of a MAIL command with additional options following the
2511 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2512 were accidentally chopped off.
2514 PH/29 SMTP synchronization checks are implemented when a command is read -
2515 there is a check that no more input is waiting when there shouldn't be
2516 any. However, for some commands, a delay in an ACL can mean that it is
2517 some time before the response is written. In this time, more input might
2518 arrive, invalidly. So now there are extra checks after an ACL has run for
2519 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2520 pipelining has not been advertised.
2522 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2524 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2525 returned in the SMTP response when smtp_return_error_details was false.
2526 This has been fixed.
2528 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2529 instead of the C I/O that was originally supplied, because problems were
2530 reported on Solaris.
2532 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2533 Exim which did not show up earlier: it was assuming that a call to
2534 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2535 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2536 was a macro that became an assignment, so it seemed to work. This has
2537 changed to a proper function call with a void return, hence the compile
2538 error. Exim's code has been fixed.
2540 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2543 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2545 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2547 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2548 because it turns out to be too big or there is a timeout) there is no
2549 "Completed" line in the log. When some messages of this type were
2550 selected by exigrep, they were listed as "not completed". Others were
2551 picked up by some special patterns. I have improved the selection
2552 criteria to be more general.
2554 PH/38 The host_find_failed option in the manualroute router can now be set
2555 to "ignore", to completely ignore a host whose IP address cannot be
2556 found. If all hosts are ignored, the behaviour is controlled by the new
2557 host_all_ignored option.
2559 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2560 homing or because of multiple MX records with /mx) generated more than
2561 one IP address, and the following item turned out to be the local host,
2562 all the secondary addresses of the first item were incorrectly removed
2563 from the list, along with the local host and any following hosts (which
2564 is what is supposed to happen).
2566 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2567 whoever called Exim into the -H file. In the case of the daemon it was
2568 behaving confusingly. When first started, it used values for whoever
2569 started the daemon, but after a SIGHUP it used the Exim user (because it
2570 calls itself on a restart). I have changed the code so that it now always
2573 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2574 message are rejected with the same error (e.g. no authentication or bad
2575 sender address), and a DATA command is nevertheless sent (as can happen
2576 with PIPELINING or a stupid MUA), the error message that was given to the
2577 RCPT commands is included in the rejection of the DATA command. This is
2578 intended to be helpful for MUAs that show only the final error to their
2581 PH/42 Another patch from the Sieve maintainer.
2583 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2584 Eximstats - Fixed some broken HTML links and added missing column headers
2586 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2587 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2589 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2591 PH/43 Yet another patch from the Sieve maintainer.
2593 PH/44 I found a way to check for a TCP/IP connection going away before sending
2594 the response to the final '.' that terminates a message, but only in the
2595 case where the client has not sent further data following the '.'
2596 (unfortunately, this is allowed). However, in many cases there won't be
2597 any further data because there won't be any more messages to send. A call
2598 to select() can be used: if it shows that the input is "ready", there is
2599 either input waiting, or the socket has been closed. An attempt to read
2600 the next input character can distinguish the two cases. Previously, Exim
2601 would have sent an OK response which the client would never have see.
2602 This could lead to message repetition. This fix should cure that, at
2603 least in a lot of common cases.
2605 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2606 advertised in response to EHLO.
2612 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2613 fixed by 4.65/MH/01 (is this a record?) are fixed:
2615 (i) An empty string was always treated as zero by the numeric comparison
2616 operators. This behaviour has been restored.
2618 (ii) It is documented that the numeric comparison operators always treat
2619 their arguments as decimal numbers. This was broken in that numbers
2620 starting with 0 were being interpreted as octal.
2622 While fixing these problems I realized that there was another issue that
2623 hadn't been noticed. Values of message_size_limit (both the global option
2624 and the transport option) were treated as octal if they started with 0.
2625 The documentation was vague. These values are now always treated as
2626 decimal, and I will make that clear in the documentation.
2632 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2633 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2636 MH/01 Don't check that the operands of numeric comparison operators are
2637 integers when their expansion is in "skipping" mode (fixes bug
2638 introduced by 4.64-PH/07).
2640 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2641 child addresses, Exim now panics and dies. Previously, because the count
2642 is held in a short int, deliveries were likely to be lost. As such a
2643 large number of recipients for a single message is ridiculous
2644 (performance will be very, very poor), I have chosen to impose a limit
2645 rather than extend the field.
2651 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2652 leftover -K file (the existence of which was triggered by #402).
2653 While we were at it, introduced process PID as part of the -K
2654 filename. This should rule out race conditions when creating
2657 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2658 processing considerably. Previous code took too long for large mails,
2659 triggering a timeout which in turn triggers #401.
2661 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2662 in the DK code in transports.c. sendfile() is not really portable,
2663 hence the _LINUX specificness.
2665 TF/01 In the add_headers option to the mail command in an Exim filter,
2666 there was a bug that Exim would claim a syntax error in any
2667 header after the first one which had an odd number of characters
2670 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2671 callout verification, Exim cached a "reject" for the entire domain. This
2672 is correct for most verifications, but it is not correct for a recipient
2673 verification with use_sender or use_postmaster set, because in that case
2674 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2675 case of MAIL FROM:<> rejection from other early rejections (e.g.
2676 rejection of HELO). When verifying a recipient using a non-null MAIL
2677 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2678 Whatever the result of the callout, the value of the domain cache is
2679 left unchanged (for any other kind of callout, getting as far as trying
2680 RCPT means that the domain itself is ok).
2682 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2685 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2686 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2687 session, the daemon ignores ECONNECT errors and logs others; it now
2688 ignores EPIPE as well.
2690 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2691 (quoted-printable decoding).
2693 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2694 later the small subsequent patch to fix an introduced bug.
2696 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2698 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2700 PH/08 An error is now given if message_size_limit is specified negative.
2702 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2703 to be given (somewhat) arbitrary names.
2705 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2708 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2709 miscellaneous code fixes
2711 PH/10 Added the log_reject_target ACL modifier to specify where to log
2714 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2715 hostname. This is wrong, because it relates to the incoming message (and
2716 probably the interface on which it is arriving) and not to the outgoing
2717 callout (which could be using a different interface). This has been
2718 changed to use the value of the helo_data option from the smtp transport
2719 instead - this is what is used when a message is actually being sent. If
2720 there is no remote transport (possible with a router that sets up host
2721 addresses), $smtp_active_hostname is used.
2723 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2724 tweaks were necessary in order to get it to work (see also 21 below):
2725 (a) The code assumed that strncpy() returns a negative number on buffer
2726 overflow, which isn't the case. Replaced with Exim's string_format()
2728 (b) There were several signed/unsigned issues. I just did the minimum
2729 hacking in of casts. There is scope for a larger refactoring.
2730 (c) The code used strcasecmp() which is not a standard C function.
2731 Replaced with Exim's strcmpic() function.
2732 (d) The code set only $1; it now sets $auth1 as well.
2733 (e) A simple test gave the error "authentication client didn't specify
2734 service in request". It would seem that Dovecot has changed its
2735 interface. Fortunately there's a specification; I followed it and
2736 changed what the client sends and it appears to be working now.
2738 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2741 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2742 address A is aliased to B and C, where B exists and C does not. Without
2743 -v the output is "A verified" because verification stops after a
2744 successful redirection if more than one address is generated. However,
2745 with -v the child addresses are also verified. Exim was outputting "A
2746 failed to verify" and then showing the successful verification for C,
2747 with its parentage. It now outputs "B failed to verify", showing B's
2748 parentage before showing the successful verification of C.
2750 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2751 look up a TXT record in a specific list after matching in a combined
2754 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2755 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2756 they consult the DNS. I had assumed they would set it the way they
2757 wanted; and indeed my experiments on Linux seem to show that in some
2758 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2759 To be on the safe side, however, I have now made the interface to
2760 host_find_byname() similar to host_find_bydns(), with an argument
2761 containing the DNS resolver options. The host_find_byname() function now
2762 sets these options at its start, just as host_find_bydns() does. The smtp
2763 transport options dns_qualify_single and dns_search_parents are passed to
2764 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2765 of host_find_byname() use the default settings of RES_DEFNAMES
2766 (qualify_single) but not RES_DNSRCH (search_parents).
2768 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2769 spool_read_header() do less string testing, by means of a preliminary
2770 switch on the second character of optional "-foo" lines. (This is
2771 overdue, caused by the large number of possibilities that now exist.
2772 Originally there were few.) While I was there, I also converted the
2773 str(n)cmp tests so they don't re-test the leading "-" and the first
2774 character, in the hope this might squeeze out yet more improvement.
2776 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2777 flag allowing group syntax was set by the header_syntax check but not
2778 turned off, possible causing trouble later; (2) The flag was not being
2779 set at all for the header_verify test, causing "group"-style headers to
2780 be rejected. I have now set it in this case, and also caused header_
2781 verify to ignore an empty address taken from a group. While doing this, I
2782 came across some other cases where the code for allowing group syntax
2783 while scanning a header line wasn't quite right (mostly, not resetting
2784 the flag correctly in the right place). These bugs could have caused
2785 trouble for malformed header lines. I hope it is now all correct.
2787 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2788 with the "reply" argument non-NULL. The code, however (which originally
2789 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2790 but it didn't always do it. This confused somebody who was copying the
2791 code for some other use. I have removed all the tests.
2793 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2794 feature that was used to support insecure browsers during the U.S. crypto
2795 embargo. It requires special client support, and Exim is probably the
2796 only MTA that supported it -- and would never use it because real RSA is
2797 always available. This code has been removed, because it had the bad
2798 effect of slowing Exim down by computing (never used) parameters for the
2799 RSA_EXPORT functionality.
2801 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2802 authenticator to fail if there's a tab character in the incoming data
2803 (there should never be unless someone is messing about, as it's supposed
2804 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2805 if the connection is using TLS or if the remote IP is the same as the
2806 local IP, and the "valid-client-cert option" if a client certificate has
2809 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2810 authenticators. This can be used for authorization after authentication
2811 succeeds. (In the case of plaintext, it servers for both authentication
2814 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2815 if any retry times were supplied.
2817 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2818 connection, where there is no client IP address to check. In this
2819 situation, the verify now always succeeds.
2821 PH/25 Applied John Jetmore's -Mset patch.
2823 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2825 PH/27 In a string expansion for a processed (not raw) header when multiple
2826 headers of the same name were present, leading whitespace was being
2827 removed from all of them, but trailing whitespace was being removed only
2828 from the last one. Now trailing whitespace is removed from each header
2829 before concatenation. Completely empty headers in a concatenation (as
2830 before) are ignored.
2832 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2833 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2835 PH/29 [Removed. This was a change that I later backed out, and forgot to
2836 correct the ChangeLog entry (that I had efficiently created) before
2837 committing the later change.]
2839 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2840 address errors (4xx response to RCPT) over the same connection as other
2841 messages routed to the same hosts. Such deliveries are always "forced",
2842 so retry times are not inspected. This resulted in far too many retries
2843 for the affected addresses. The effect occurred only when there were more
2844 hosts than the hosts_max_try setting in the smtp transport when it had
2845 the 4xx errors. Those hosts that it had tried were not added to the list
2846 of hosts for which the message was waiting, so if all were tried, there
2847 was no problem. Two fixes have been applied:
2849 (i) If there are any address or message errors in an SMTP delivery, none
2850 of the hosts (tried or untried) are now added to the list of hosts
2851 for which the message is waiting, so the message should not be a
2852 candidate for sending over the same connection that was used for a
2853 successful delivery of some other message. This seems entirely
2854 reasonable: after all the message is NOT "waiting for some host".
2855 This is so "obvious" that I'm not sure why it wasn't done
2856 previously. Hope I haven't missed anything, but it can't do any
2857 harm, as the worst effect is to miss an optimization.
2859 (ii) If, despite (i), such a delivery is accidentally attempted, the
2860 routing retry time is respected, so at least it doesn't keep
2861 hammering the server.
2863 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2864 in ${readsocket because some servers need this prod.
2866 PH/32 Added some extra debug output when updating a wait-xxx database.
2868 PH/33 The hint "could be header name not terminated by colon", which has been
2869 given for certain expansion errors for a long time, was not being given
2870 for the ${if def:h_colon_omitted{... case.
2872 PH/34 The spec says: "With one important exception, whenever a domain list is
2873 being scanned, $domain contains the subject domain." There was at least
2874 one case where this was not true.
2876 PH/35 The error "getsockname() failed: connection reset by peer" was being
2877 written to the panic log as well as the main log, but it isn't really
2878 panic-worthy as it just means the connection died rather early on. I have
2879 removed the panic log writing for the ECONNRESET error when getsockname()
2882 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2883 runs only) independently of the message's sender address. This meant
2884 that, if the 4xx error was in fact related to the sender, a different
2885 message to the same recipient with a different sender could confuse
2886 things. In particular, this can happen when sending to a greylisting
2887 server, but other circumstances could also provoke similar problems.
2888 I have changed the default so that the retry time for these errors is now
2889 based a combination of the sender and recipient addresses. This change
2890 can be overridden by setting address_retry_include_sender=false in the
2893 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2894 remote server are returned as part of bounce messages. This was not
2895 happening for LMTP over a pipe (the lmtp transport), but now it is the
2896 same for both kinds of LMTP.
2898 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2899 in header lines that were in fact CNAMEs. This is no longer the case.
2901 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2902 and queue runs started by the daemon processed all messages. This has
2903 been fixed so that -R and -S can now usefully be given with -q<time>.
2905 PH/40 Import PCRE release 6.7 (fixes some bugs).
2907 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2909 PH/42 Give an error if -q is specified more than once.
2911 PH/43 Renamed the variables $interface_address and $interface_port as
2912 $received_ip_address and $received_port, to make it clear that these
2913 values apply to message reception, and not to the outgoing interface when
2914 a message is delivered. (The old names remain recognized, of course.)
2916 PH/44 There was no timeout on the connect() call when using a Unix domain
2917 socket in the ${readsocket expansion. There now is.
2919 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2920 be meaningful with "accept".
2922 SC/01 Eximstats V1.43
2923 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2925 SC/02 Eximstats V1.44
2926 Use a glob alias rather than an array ref in the generated
2927 parser. This improves both readability and performance.
2929 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2930 Collect SpamAssassin and rejection statistics.
2931 Don't display local sender or destination tables unless
2932 there is data to show.
2933 Added average volumes into the top table text output.
2935 SC/04 Eximstats V1.46
2936 Collect data on the number of addresses (recipients)
2937 as well as the number of messages.
2939 SC/05 Eximstats V1.47
2940 Added 'Message too big' to the list of mail rejection
2941 reasons (thanks to Marco Gaiarin).
2943 SC/06 Eximstats V1.48
2944 Mainlog lines which have GMT offsets and are too short to
2945 have a flag are now skipped.
2947 SC/07 Eximstats V1.49 (Alain Williams)
2948 Added the -emptyok flag.
2950 SC/08 Eximstats V1.50
2951 Fixes for obtaining the IP address from reject messages.
2953 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2954 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2955 whitespace changes from 4.64-PH/27
2957 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2960 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2961 are found, allow negative numbers in numeric criteria)
2963 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2965 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2966 to match changes made in 4.64-PH/43
2968 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2970 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2971 "domain" is actually an IP address, and force a failure. This locks out
2972 those revolvers/nameservers that support "A-for-A" lookups, in
2973 contravention of the specifications.
2975 PH/48 When a host name was looked up from an IP address, and the subsequent
2976 forward lookup of the name timed out, the host name was left in
2977 $sender_host_name, contrary to the specification.
2979 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2980 restricted to single-key lookups, Exim was not diagnosing an error if
2981 * or *@ was used with a query-style lookup.
2983 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2985 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2986 long ago, but noone interested enough thought of it. Let's just say that
2987 the "1.1" means that there are some new functions that weren't there at
2988 some point in the past.
2990 PH/51 Error processing for expansion failure of helo_data from an smtp
2991 transport during callout processing was broken.
2993 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2994 tested/used via the -bh/-bhc/-bs options.
2996 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2997 bug, fixed in subsequent PCRE releases).
2999 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3000 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3002 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3008 SC/01 Use a glob alias rather than an array ref in eximstats generated
3009 parser. This improves both readability and performance.
3011 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3012 Don't display local sender or destination tables in eximstats unless
3013 there is data to show.
3014 Added average volumes into the eximstats top table text output.
3016 SC/03 Collect data on the number of addresses (recipients) as well
3017 as the number of messages in eximstats.
3019 TF/01 Correct an error in the documentation for the redirect router. Exim
3020 does (usually) call initgroups() when daemonizing.
3022 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3023 with consistent privilege compared to when running as a daemon.
3025 TF/03 Note in the spec that $authenticated_id is not set for local
3026 submissions from trusted users.
3028 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3029 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3031 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3032 by adding some example configuration directives to the default
3033 configuration file. A little bit of work is required to uncomment the
3034 directives and define how usernames and passwords are checked, but
3035 there is now a framework to start from.
3037 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3038 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3039 without this. I don't know how relevant this is to other LDAP libraries.
3041 PH/02 Add the verb name to the "unknown ACL verb" error.
3043 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3045 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3047 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3048 directory not expanded when it should be if an expanded home directory
3049 was set for the address (which is overridden by the transport).
3051 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3054 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3055 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3056 because it is too late at that time, and has no effect.
3058 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3059 security issue with \' (bugzilla #107). I could not use the
3060 PQescapeStringConn() function, because it needs a PGconn value as one of
3063 PH/08 When testing addresses using -bt, indicate those final addresses that
3064 are duplicates that would not cause an additional delivery. At least one
3065 person was confused, thinking that -bt output corresponded to deliveries.
3066 (Suppressing duplicates isn't a good idea as you lose the information
3067 about possibly different redirections that led to the duplicates.)
3069 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3070 systems where poll() doesn't work, in particular OS X.
3072 PH/10 Added more information to debugging output for retry time not reached.
3074 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3075 operations in malware.c.
3077 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3080 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3081 syslog_duplication set false, log lines that would normally be written
3082 both the the main log and to the reject log were not written to syslog at
3085 PH/14 In the default configuration, change the use of "message" in ACL warn
3086 statements to "add_header".
3088 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3089 not followed by a command (e.g. "seen endif").
3091 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3092 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3095 PH/17 Added extra conditions to the default value of delay_warning_condition
3099 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3100 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3101 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3104 The Auto-Submitted: and various List- headers are standardised, whereas I
3105 don't think Precedence: ever was.
3107 PH/18 Refactored debugging code in route_finduser() to show more information,
3108 in particular, the error code if getpwnam() issues one.
3110 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3111 This is apparently needed in addition to the PH/07 change above to avoid
3112 any possible encoding problems.
3114 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3115 but not after initializing Perl.
3117 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3118 output them only if debugging. By default they are written stderr,
3119 apparently, which is not desirable.
3121 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3124 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3127 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3129 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3130 authenticated or an ident call has been made. Suppress the default
3131 values for $authenticated_id and $authenticated_sender (but permit -oMai
3132 and -oMas) when testing with -bh.
3134 PH/24 Re-jigged the order of the tests in the default configuration so that the
3135 tests for valid domains and recipients precede the DNS black list and CSA
3136 tests, on the grounds that those ones are more expensive.
3138 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3139 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3140 This bug exists in every version of Exim that I still have, right back to
3143 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3144 However, an attempt to turn on case-sensitivity in a regex key by
3145 including (?-i) didn't work because the subject string was already
3146 lowercased, and the effects were non-intuitive. It turns out that a
3147 one-line patch can be used to allow (?-i) to work as expected.
3153 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3154 other effects) broke the use of negated acl sub-conditions.
3156 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3159 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3160 "Deny" causes Exim to reject the incoming connection with a 554 error.
3161 Unfortunately, if there is a major crisis, such as a disk failure,
3162 tcp-wrappers gives "deny", whereas what one would like would be some
3163 kind of temporary error. A kludge has been added to help with this.
3164 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3165 554 error is used if errno is still zero or contains ENOENT (which occurs
3166 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3169 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3171 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3172 errors. Otherwise a message that provokes a temporary error (when other
3173 messages do not) can cause a whole host to time out.
3175 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3176 addresses were routed directly to files or pipes from a redirect router.
3177 File deliveries just didn't batch; pipe deliveries might have suffered
3180 PH/06 A failure to get a lock for a hints database would erroneously always say
3181 "Failed to get write lock", even when it was really a read lock.
3183 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3184 of 0600. This has been changed to use the value of the lockfile_mode
3185 option (which defaults to 0600).
3187 PH/08 Applied small patch from the Sieve maintainer.
3189 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3190 folder from quota calculations, a direct delivery into this folder messed
3191 up the contents of the maildirsize file. This was because the regex was
3192 used only to exclude .Trash (or whatever) when the size of the mailbox
3193 was calculated. There was no check that a delivery was happening into an
3194 excluded directory. This bug has been fixed by ignoring all quota
3195 processing for deliveries into excluded directories.
3197 PH/10 Added the maildirfolder_create_regex option to appendfile.
3203 PH/01 The code for finding all the local interface addresses on a FreeBSD
3204 system running IPv6 was broken. This may well have applied to all BSD
3205 systems, as well as to others that have similar system calls. The broken
3206 code found IPv4 interfaces correctly, but gave incorrect values for the
3207 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3208 that it would not match correctly against @[] and not recognize the IPv6
3211 PH/02 The ipliteral router was not recognizing addresses of the form user@
3212 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3214 PH/03 Added disable_ipv6.
3216 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3217 decoded form, because it is most often used to construct To: headers
3218 lines in autoreplies, and the decoded form may well be syntactically
3219 invalid. However, $reply_address has leading white space removed, and all
3220 newlines turned into spaces so that the autoreply transport does not
3223 PH/05 If group was specified without a user on a router, and no group or user
3224 was specified on a transport, the group from the router was ignored.
3226 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3227 for visible compile-time settings that can be used to change these
3228 numbers, for those that want even more. Backwards compatibility with old
3229 spool files has been maintained. However, going back to a previous Exim
3230 release will lost any variables that are in spool files.
3232 PH/07 Two small changes when running in the test harness: increase delay when
3233 passing a TCP/IP connection to a new process, in case the original
3234 process has to generate a bounce, and remove special handling of
3235 127.0.0.2 (sic), which is no longer necessary.
3237 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3238 be the same on different OS.
3240 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3243 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3244 whether --show-vars was specified or not
3246 JJ/02 exipick: Added support for new ACL variable spool format introduced
3249 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3250 syntactically invalid From: or Reply-to: line, and a filter used this to
3251 generate an autoreply, and therefore failed to obtain an address for the
3252 autoreply, Exim could try to deliver to a non-existent relative file
3253 name, causing unrelated and misleading errors. What now happens is that
3254 it logs this as a hard delivery error, but does not attempt to create a
3257 PH/11 The exinext utility has a -C option for testing purposes, but although
3258 the given file was scanned by exinext itself; it wasn't being passed on
3259 when Exim was called.
3261 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3262 an end-of-file indication when reading a command response.
3264 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3265 compiled. In many other places in Exim, IPv6 addresses are always
3266 recognized, so I have changed this. It also means that IPv4 domain
3267 literals of the form [IPV4:n.n.n.n] are now always recognized.
3269 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3270 used if the router is not running as root, for example, when verifying at
3271 ACL time, or when using -bh. The debugging output from this situation was
3272 non-existent - all you got was a failure to exec. I have made two
3275 (a) Failures to set uid/gid, the current directory, or a process leader
3276 in a subprocess such as that created by queryprogram now generate
3277 suitable debugging output when -d is set.
3279 (b) The queryprogram router detects when it is not running as root,
3280 outputs suitable debugging information if -d is set, and then runs
3281 the subprocess without attempting to change uid/gid.
3283 PH/15 Minor change to Makefile for building test_host (undocumented testing
3286 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3287 additional section of a DNS packet that returns MX or SRV records.
3288 Instead, it always explicitly searches for A/AAAA records. This avoids
3289 major problems that occur when a DNS server includes only records of one
3290 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3291 fixed another bug: if SRV records were looked up and the corresponding
3292 address records were *not* found in the additional section, the port
3293 values from the SRV records were lost.
3295 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3296 using the correct key (the original address) when searching the retry
3297 rules in order to find which one to use for generating the retry hint.
3299 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3300 adding the default one. Similarly, if it contains a Reply-To: header, the
3301 errors_reply_to option, if set, is not used.
3303 PH/19 When calculating a retry time, Exim used to measure the "time since
3304 failure" by looking at the "first failed" field in the retry record. Now
3305 it does not use this if it is later than than the arrival time of the
3306 message. Instead it uses the arrival time. This makes for better
3307 behaviour in cases where some deliveries succeed, thus re-setting the
3308 "first failed" field. An example is a quota failure for a huge message
3309 when small messages continue to be delivered. Without this change, the
3310 "time since failure" will always be short, possible causing more frequent
3311 delivery attempts for the huge message than are intended.
3312 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3314 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3315 $1, $2, $3) because the numerical variables can be reset during some
3316 expansion items (e.g. "match"), thereby losing the authentication data.
3318 PH/21 Make -bV show the size of off_t variables so that the test suite can
3319 decide whether to run tests for quotas > 2G.
3321 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3322 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3323 filecount value is greater than 2G or if a quota value is greater than 2G
3324 on a system where the size of off_t is not greater than 4, a panic error
3327 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3328 never match. The debug and -bh output now contains an explicit error
3329 message indicating a malformed IPv4 address or mask.
3331 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3332 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3333 PH/23 above applies.
3335 PH/25 Do not write to syslog when running in the test harness. The only
3336 occasion when this arises is a failure to open the main or panic logs
3337 (for which there is an explicit test).
3339 PH/26 Added the /no_tell option to "control=freeze".
3341 PH/27 If a host name lookup failed very early in a connection, for example, if
3342 the IP address matched host_lookup and the reverse lookup yielded a name
3343 that did not have a forward lookup, an error message of the form "no IP
3344 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3345 could be logged. Now it outputs the IP address instead of "NULL".
3347 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3348 allows the sender and the authenticated sender to be set when
3349 submitting a message from within Exim. Since child_open_exim() is
3350 documented for local_scan(), the new function should be too.
3352 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3353 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3354 results in an empty string is now treated as unset.
3356 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3358 PH/31 Added sender_verify_fail logging option.
3360 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3361 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3362 tidied the source and removed it altogether.
3364 PH/33 When a queue run was abandoned because the load average was too high, a
3365 log line was always written; now it is written only if the queue_run log
3366 selector is set. In addition, the log line for abandonment now contains
3367 information about the queue run such as the pid. This is always present
3368 in "start" and "stop" lines but was omitted from the "abandon" line.
3370 PH/34 Omit spaces between a header name and the colon in the error message that
3371 is given when verify = headers_syntax fails (if there are lots of them,
3372 the message gets confusing).
3374 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3375 names, as there are now some PTR records that contain slashes. This check
3376 is only to protect against broken name servers that fall over on strange
3377 characters, so the fact that it applies to all lookups doesn't matter.
3379 PH/36 Now that the new test suite is complete, we can remove some of the
3380 special code in Exim that was needed for the old test suite. For example,
3381 sorting DNS records because real resolvers return them in an arbitrary
3382 order. The new test suite's fake resolver always returns records in the
3385 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3386 bounces) except when queue_only is set, to avoid logging races between
3387 the different processes.
3389 PH/38 Panic-die if .include specifies a non-absolute path.
3391 PH/39 A tweak to the "H" retry rule from its user.
3393 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3394 a label. They prevented compilation on older perls.
3396 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3397 a warning to be raised on newish perls.
3399 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3400 on queue. Changes to match documented behaviour of showing count of
3401 messages matching specified criteria.
3403 PH/40 Changed the default ident timeout from 30s to 5s.
3405 PH/41 Added support for the use of login_cap features, on those BSD systems
3406 that have them, for controlling the resources used by pipe deliveries.
3408 PH/42 The content-scanning code uses fopen() to create files in which to put
3409 message data. Previously it was not paying any attention to the mode of
3410 the files. Exim runs with umask(0) because the rest of the code creates
3411 files with open(), and sets the required mode explicitly. Thus, these
3412 files were ending up world-writeable. This was not a big issue, because,
3413 being within the spool directory, they were not world-accessible. I have
3414 created a function called modefopen, which takes an additional mode
3415 argument. It sets umask(777), creates the file, chmods it to the required
3416 mode, then resets the umask. All the relevant calls to fopen() in the
3417 content scanning code have been changed to use this function.
3419 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3420 to 24 hours. This avoids potential overflow problems when processing G
3421 and H retry rules. I suspect nobody ever tinkers with this value.
3423 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3425 PH/45 When the plaintext authenticator is running as a client, the server's
3426 challenges are checked to ensure they are valid base64 strings. By
3427 default, the authentication attempt is cancelled if an invalid string is
3428 received. Setting client_ignore_invalid_base64 true ignores these errors.
3429 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3430 they are received. Thus, the responses can be made to depend on the
3431 challenges. If an invalid string is ignored, an empty string is placed in
3434 PH/46 Messages that are created by the autoreply transport now contains a
3435 References: header, in accordance with RFCs 2822 and 3834.
3437 PH/47 Added authenticated_sender_force to the smtp transport.
3439 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3441 PH/49 Installed latest patch from the Sieve maintainer.
3443 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3444 was also set, the appendfile transport was unnecessarily scanning a
3445 directory of message files (e.g. for maildir delivery) to find the count
3446 of files (along with the size), even though it did not need this
3447 information. It now does the scan only if it needs to find either the
3448 size of the count of files.
3450 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3452 PH/52 Two bugs concerned with error handling when the smtp transport is
3455 (i) Exim was not creating retry information for temporary errors given
3456 for individual recipients after the DATA command when the smtp transport
3457 was used in LMTP mode. This meant that they could be retried too
3458 frequently, and not timed out correctly.
3460 (ii) Exim was setting the flag that allows error details to be returned
3461 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3462 recipients that were returned after the DATA command.
3464 PH/53 This is related to PH/52, but is more general: for any failing address,
3465 when detailed error information was permitted to be returned to the
3466 sender, but the error was temporary, then after the final timeout, only
3467 "retry timeout exceeded" was returned. Now it returns the full error as
3468 well as "retry timeout exceeded".
3470 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3471 do this, and (what is worse) MTAs that accept it.
3473 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3474 will now be deprecated.
3476 PH/56 New os.c-cygwin from the Cygwin maintainer.
3478 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3479 formats (previously only available in exim formats via -bpr, -bpru,
3480 and -bpra. Now also available in native and exiqgrep formats)
3482 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3483 with very large, slow to parse queues
3485 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3487 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3489 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3490 responses to authentication challenges, though it was showing the
3491 challenges; (ii) I've removed the CR characters from the debug output for
3494 PH/58 Allow for the insertion of a newline as well as a space when a string
3495 is turned into more than one encoded-word during RFC 2047 encoding. The
3496 Sieve code now uses this.
3498 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3499 data_4xx, lost_connection, tls_required.
3501 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3502 message was being sent as an SMTP response.
3504 PH/61 Add -l and -k options to exicyclog.
3506 PH/62 When verifying, if an address was redirected to one new address, so that
3507 verification continued, and the new address failed or deferred after
3508 having set something in $address_data, the value of $address_data was not
3509 passed back to the ACL. This was different to the case when no
3510 redirection occurred. The value is now passed back in both cases.
3512 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3513 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3514 use login_cap.h, so on its own it isn't the distinguishing feature. The
3515 new name refers directly to the setclassresources() function.
3517 PH/65 Added configuration files for NetBSD3.
3519 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3521 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3522 is preferred over IPv4.
3524 PH/68 The bounce_return_message and bounce_return_body options were not being
3525 honoured for bounces generated during the reception of non-SMTP messages.
3526 In particular, this applied to messages rejected by the ACL. This bug has
3527 been fixed. However, if bounce_return_message is true and bounce_return_
3528 body is false, the headers that are returned for a non-SMTP message
3529 include only those that have been read before the error was detected.
3530 (In the case of an ACL rejection, they have all been read.)
3532 PH/69 The HTML version of the specification is now built in a directory called
3533 spec_html instead of spec.html, because the latter looks like a path with
3534 a MIME-type, and this confuses some software.
3536 PH/70 Catch two compiler warnings in sieve.c.
3538 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3539 function verify_get_ident() calls ip_connect() to connect a socket, but
3540 if the "connect()" function timed out, ip_connect() used to close the
3541 socket. However, verify_get_ident() also closes the socket later, and in
3542 between Exim writes to the log, which may get opened at this point. When
3543 the socket was closed in ip_connect(), the log could get the same file
3544 descriptor number as the socket. This naturally causes chaos. The fix is
3545 not to close the socket in ip_connect(); the socket should be closed by
3546 the function that creates it. There was only one place in the code where
3547 this was missing, in the iplookup router, which I don't think anybody now
3548 uses, but I've fixed it anyway.
3550 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3551 well as to direct DNS lookups. Otherwise the handling of names in host
3552 lists is inconsistent and therefore confusing.
3558 PH/01 Two changes to the default runtime configuration:
3560 (1) Move the checks for relay_from_hosts and authenticated clients from
3561 after to before the (commented out) DNS black list checks.
3563 (2) Add control=submission to the relay_from_hosts and authenticated
3564 clients checks, on the grounds that messages accepted by these
3565 statements are most likely to be submissions.
3567 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3569 (1) Generate an error if the third argument for the ${prvs expansion is
3572 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3575 (3) Reset the variables that are obtained from the first argument of
3576 ${prvscheck and used in the second argument before leaving the code,
3577 because their memory is reclaimed, so using them afterwards may do
3580 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3581 one (it's much easier than Tom thought :-).
3583 (5) Because of (4), we can now allow for the use of $prvscheck_result
3584 inside the third argument.
3586 PH/03 For some reason, the default setting of PATH when running a command from
3587 a pipe transport was just "/usr/bin". I have changed it to
3590 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3591 anything to be listed in the output from -bV.
3593 PH/05 When a filter generated an autoreply, the entire To: header line was
3594 quoted in the delivery log line, like this:
3596 => >A.N.Other <ano@some.domain> <original@ddress> ...
3598 This has been changed so that it extracts the operative address. There
3599 may be more than one such address. If so, they are comma-separated, like
3602 => >ano@some.domain,ona@other.domain <original@ddress> ...
3604 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3605 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3606 not being looked up in the rDNS to get a host name, Exim was showing the
3607 IP address twice in Received: lines, even though the IP addresses were
3608 identical. For example:
3610 Received: from [1.2.3.4] (helo=[1.2.3.4])
3612 However, if the real host name was known, it was omitting the HELO data
3613 if it matched the actual IP address. This has been tidied up so that it
3614 doesn't show the same IP address twice.
3616 PH/07 When both +timestamp and +memory debugging was on, the value given by
3617 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3618 called by the debug printing, thereby overwriting the timestamp buffer.
3619 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3621 PH/08 When the original message was included in an autoreply transport, it
3622 always said "this is a copy of the message, including all the headers",
3623 even if body_only or headers_only was set. It now gives an appropriate
3626 PH/09 Applied a patch from the Sieve maintainer which:
3628 o fixes some comments
3629 o adds the (disabled) notify extension core
3630 o adds some debug output for the result of if/elsif tests
3631 o points to the current vacation draft in the documentation
3632 and documents the missing references header update
3636 o fixes a bug in processing the envelope test (when testing
3637 multiple envelope elements, the last element determined the
3640 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3641 Electronic Mail") by including:
3643 Auto-submitted: auto-generated
3645 in the messages that it generates (bounce messages and others, such as
3646 warnings). In the case of bounce messages for non-SMTP messages, there was
3647 also a typo: it was using "Auto_submitted" (underscore instead of
3648 hyphen). Since every message generated by Exim is necessarily in response
3649 to another message, thes have all been changed to:
3651 Auto-Submitted: auto-replied
3653 in accordance with these statements in the RFC:
3655 The auto-replied keyword:
3657 - SHOULD be used on messages sent in direct response to another
3658 message by an automatic process,
3660 - MUST NOT be used on manually-generated messages,
3662 - MAY be used on Delivery Status Notifications (DSNs) and Message
3663 Disposition Notifications (MDNs),
3665 - MUST NOT be used on messages generated by automatic or periodic
3666 processes, except for messages which are automatic responses to
3669 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3670 to the default Received: header definition.
3672 PH/12 Added log selector acl_warn_skipped (default on).
3674 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3675 variables because (a) they are in the wrong storage pool and (b) even if
3676 they were copied, it wouldn't work properly because of the caching.
3678 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3679 checking when decoding. Apparently there are clients that generate
3680 overlong encoded strings. Why am I not surprised?
3682 PH/15 If the first argument of "${if match_address" was not empty, but did not
3683 contain an "@" character, Exim crashed. Now it writes a panic log message
3684 and treats the condition as false.
3686 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3688 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3689 extension "envelope-auth". The code is finished and in agreement with
3690 other implementations, but there is no documentation so far and in fact,
3691 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3692 not changing the active code.
3694 Print executed "if" and "elsif" statements when debugging is used. This
3695 helps a great deal to understand what a filter does.
3697 Document more things not specified clearly in RFC3028. I had all this
3698 sorted out, when out of a sudden new issues came to my mind. Oops."
3700 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3703 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3704 canonical form (as documented). However, after a host name lookup from
3705 the IP address, check_host() was doing a simple string comparison with
3706 addresses acquired from the DNS when checking that the found name did
3707 have the original IP as one of its addresses. Since any found IPv6
3708 addresses are likely to be in abbreviated form, the comparison could
3709 fail. Luckily, there already exists a function for doing the comparison
3710 by converting both addresses to binary, so now that is used instead of
3711 the text comparison.
3713 PH/20 There was another similar case to PH/19, when a complete host name was
3714 given in a host list; looking up its IP address could give an abbreviated
3715 form, whereas the current host's name might or might not be abbreviated.
3716 The same fix has been applied.
3722 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3723 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3726 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3727 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3729 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3731 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3732 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3733 Also, for "Auto-Submitted", treat anything other than "no" as
3734 non-personal, in accordance with RFC 3834. (Previously it treated
3735 anything starting "auto-" as non-personal.)
3737 TF/01 The control=submission/name=... option had a problem with syntax
3738 errors if the name included a slash character. The /name= option
3739 now slurps the rest of the string, so it can include any characters
3740 but it must come last in the list of options (after /sender_retain
3743 PH/05 Some modifications to the interface to the fake nameserver for the new
3751 TK/01 Added the "success_on_redirect" address verification option. See
3752 NewStuff for rationale and an example.
3754 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3756 PH/02 Patch to exigrep to allow it to work on syslog lines.
3758 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3759 fread() to skip over the body file's header line, because in Cygwin the
3760 header line is locked and is inaccessible.
3762 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3763 co-exist for some time) to make it clear that it is the Exim ID that is
3764 referenced, not the Message-ID: header line.
3766 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3767 string_format() function, because snprintf() does not exist on all
3770 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3771 ${stat: expansion item.
3773 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3774 protocol synchronization error", to keep the pedants happy.
3776 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3777 well as for IRIX systems, when gcc is being used. See the host.c source
3780 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3782 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3785 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3786 transport and to the smtp transport in LMTP mode.
3788 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3790 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3791 run a filter in a subprocess. This could lead to confusion in subsequent
3792 lookups in the parent process. There should also be a search_tidyup() at
3793 the end of the subprocess.
3795 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3796 only if the host matched helo_try_verify_hosts, which caused the
3797 verification to occur when the EHLO/HELO command was issued. The ACL just
3798 tested the remembered result. Now, if a previous verification attempt has
3799 not happened, "verify = helo" does it there and then.
3801 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3803 TK/03 Fix log output including CR from clamd.
3805 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3806 exist provoked a memory error which could cause a segfault.
3808 PH/15 Installed PCRE 6.2
3810 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3812 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3813 of the problem. Specifically, suggested +O2 rather than +O1 for the
3816 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3818 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3819 record, the port was not correctly logged when the outgoing_port log
3820 selector was set (it logged the transort's default port).
3822 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3823 fallback_hosts, and "hosts" in the smtp transport.
3825 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3826 host errors such as "Connection refused".
3828 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3829 authentication with radiusclient 0.4.9:
3831 - Error returned from rc_read_config was caught wrongly
3832 - Username/password not passed on to radius server due to wrong length.
3834 The presumption is that some radiusclient API changes for 4.51/PH/17
3835 were not taken care of correctly. The code is still untested by me (my
3836 Linux distribution still has 0.3.2 of radiusclient), but it was
3837 contributed by a Radius user.
3839 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3840 expanding the "port" option of the smtp transport.
3842 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3843 while reading a MIME header. Thanks to Tom Hughes for a patch.
3845 PH/24 Include config.h inside local_scan.h so that configuration settings are
3848 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3849 that in an ACL for RCPT (for example), you can examine exactly what was
3852 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3853 commands, but it was not correctly comparing the address with the actual
3854 client host address. Thus, it would show the EHLO address in Received:
3855 header lines when this was not necessary.
3857 PH/27 Added the % operator to ${eval:}.
3859 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3860 it should be ignoring failures (because with -C, for example, it has lost
3861 privilege). It wasn't ignoring creation failures other than "already
3864 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3867 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3868 because an input line was too long, either on its own, or by virtue of
3869 too many continuations, the temporary file was not being removed, and the
3870 return code was incorrect.
3872 PH/31 Missing "BOOL" in function definition in filtertest.c.
3874 PH/32 Applied Sieve patches from the maintainer.
3876 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3878 PH/33 Added "verify = not_blind".
3880 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3881 Local/Makefile (with some defaults set). These are used in built scripts
3882 such as exicyclog, but they have never been used in the exim_install
3883 script (though there are many overriding facilities there). I have
3884 arranged that the exim_install script now takes note of these two
3887 PH/35 Installed configuration files for Dragonfly.
3889 PH/36 When a locally submitted message by a trusted user did not contain a
3890 From: header, and the sender address was obtained from -f or from an SMTP
3891 MAIL command, and the trusted user did not use -F to supply a sender
3892 name, $originator_name was incorrectly used when constructing a From:
3893 header. Furthermore, $originator_name was used for submission mode
3894 messages from external hosts without From: headers in a similar way,
3895 which is clearly wrong.
3897 PH/37 Added control=suppress_local_fixups.
3899 PH/38 When log_selector = +received_sender was set, and the addition of the
3900 sender made the log line's construction buffer exactly full, or one byte
3901 less than full, an overflow happened when the terminating "\n" was
3904 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3905 when the result of a list match is failure because a DNS lookup failed.
3907 PH/40 RM_COMMAND is now used in the building process.
3909 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3910 the "build-* directories that it finds.
3912 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3913 address in a domain literal was a prefix of an interface address.
3915 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3916 when verifying a sender address, unless rewrite_headers is false.
3918 PH/44 Wrote a long comment about why errors_to addresses are verified as
3919 recipients, not senders.
3921 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3922 the ratelimit ACL was added.
3924 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3926 PH/46 Added extra information about PostgreSQL errors to the error string.
3928 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3929 avoiding the need to install special zones in a real server. This is
3930 backwards compatible; if it can't find the fake resolver, it drops back.
3931 Thus, both old and new test suites can be run.
3933 TF/02 Added util/ratelimit.pl
3935 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3936 clock is set back in time.
3938 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3939 Candler <B.Candler@pobox.com>.
3941 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3942 OK for addresses that are the result of redirections.
3944 PH/48 A number of further additions for the benefit of the new test suite,
3945 including a fake gethostbyname() that interfaces to the fake DNS resolver
3948 TF/06 The fix for widen_domains has also been applied to qualify_single and
3949 search_parents which are the other dnslookup options that can cause
3952 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3955 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3957 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3958 service" if it encountered an MX record with an empty target hostname.
3959 The message is now "an MX or SRV record indicated no SMTP service".
3961 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3962 if the DNS of the sending site is misconfigured. This is quite a
3963 common situation. This change restores the behaviour of treating a
3964 helo verification defer as a failure.
3966 PH/51 If self=fail was set on a router, the bounce message did not include the
3967 actual error message.
3973 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3975 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3976 command itself ended in error, the underlying message about the transport
3977 filter timeout was being overwritten with the pipe command error. Now the
3978 underlying error message should be appended to the second error message.
3980 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3982 PH/02 Reduce the amount of output that "make" produces by default. Full output
3983 can still be requested.
3985 PH/03 The warning log line about a condition test deferring for a "warn" verb
3986 was being output only once per connection, rather than after each
3987 occurrence (because it was using the same function as for successful
3988 "warn" verbs). This seems wrong, so I have changed it.
3990 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3991 it should not have, which might have caused a crash in the right
3992 circumstances, but probably never did.
3994 PH/04 Installed a modified version of Tony Finch's patch to make submission
3995 mode fix the return path as well as the Sender: header line, and to
3996 add a /name= option so that you can make the user's friendly name appear
3999 TF/03 Added the control = fakedefer ACL modifier.
4001 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4002 Mark Lowes for thorough testing.
4004 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4006 TK/03 Merged latest SRS patch from Miles Wilton.
4008 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4009 with the definition in sysexits.h (which is #included earlier).
4010 Fortunately, Exim does not actually use EX_OK. The code used to try to
4011 preserve the sysexits.h value, by assuming that macro definitions were
4012 scanned for macro replacements. I have been disabused of this notion,
4013 so now the code just undefines EX_OK before #including unistd.h.
4015 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4016 in the smtp transport. When a block could not be written in a single
4017 write() function, the timeout was being re-applied to each part-write.
4018 This seems wrong - if the receiver was accepting one byte at a time it
4019 would take for ever. The timeout is now adjusted when this happens. It
4020 doesn't have to be particularly precise.
4022 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4023 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4025 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4026 without a domain if the check to <postmaster@domain> fails.
4028 SC/01 Eximstats: added -xls and the ability to specify output files
4029 (patch written by Frank Heydlauf).
4031 SC/02 Eximstats: use FileHandles for outputting results.
4033 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4035 SC/04 Eximstats: fixed display of large numbers with -nvr option
4037 SC/05 Eximstats: fixed merging of reports with empty tables.
4039 SC/06 Eximstats: added the -include_original_destination flag
4041 SC/07 Eximstats: removed tabs and trailing whitespace.
4043 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4045 TK/06 MBOX spool code: Add real "From " MBOX separator line
4046 so the .eml file is really in mbox format (even though
4047 most programs do not really care). Patch from Alex Miller.
4049 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4050 The latter is generated from $received_to and is only set if the
4051 message has one envelope recipient. SA can use these headers,
4052 obviously out-of-the-box. Patch from Alex Miller.
4054 PH/08 The ${def test on a variable was returning false if the variable's
4055 value was "0", contrary to what the specification has always said!
4056 The result should be true unless the variable is empty.
4058 PH/09 The syntax error of a character other than { following "${if
4059 def:variable_name" (after optional whitespace) was not being diagnosed.
4060 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4061 accidental colon was present, for example, could give incorrect results.
4063 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4064 result is used (not including appendfile, where other changes are about
4067 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4068 This involved changing a lot of size variables from int to off_t. It
4069 should work with maildirs and everything.
4071 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4072 spamd dying while we are connected to it.
4074 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4077 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4078 function for BDB 4.3.
4080 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4082 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4083 the sources for PCRE have been re-arranged and more files are now
4086 PH/15 The code I had for printing potentially long long variables in PH/11
4087 above was not the best (it lost precision). The length of off_t variables
4088 is now inspected at build time, and an appropriate printing format (%ld
4089 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4090 to be "long long int" or "long int". This is needed for the internal
4091 formatting function string_vformat().
4093 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4094 the configuration file to be ":syslog", then the script "guesses" where
4095 the logs files are, rather than using the compiled in default. In our
4096 case the guess is not the same as the compiled default, so the script
4097 suddenly stopped working when I started to use syslog. The patch checks
4098 to see if log_file_path is "". If so, it attempts to read it from exim
4099 with no configuration file to get the compiled in version, before it
4100 falls back to the previous guessing code."
4102 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4103 implementing BATV in an Exim configuration. See NewStuff for the gory
4106 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4107 Makefile that are specific to HP-UX.
4109 PH/18 If the "use_postmaster" option was set for a recipient callout together
4110 with the "random" option, the postmaster address was used as the MAIL
4111 FROM address for the random test, but not for the subsequent recipient
4112 test. It is now used for both.
4114 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4115 patch removes a few documentation additions to RFC 3028, because the
4116 latest draft now contains them. It adds the new en;ascii-case comparator
4117 and a new error check for 8bit text in MIME parts. Comparator and
4118 require names are now matched exactly. I enabled the subaddress
4119 extension, but it is not well tested yet (read: it works for me)."
4121 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4122 rework some of the code of TK/09 above to avoid the hardwired use of
4123 "%lld" and "long long". Replaced the call to snprintf() with a call to
4126 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4127 records point to non-existent hosts", "retry timeout exceeded", and
4128 "retry time not reached for any host after a long failure period".
4130 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4131 experimental DomainKeys support:
4133 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4134 (2) On an error such as an illegally used "control", the wrong name for
4135 the control was given.
4137 These problems did NOT occur unless DomainKeys support was compiled.
4139 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4141 PH/24 Added ${if match_ip condition.
4143 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4144 databases so that it will be absolutely obvious if a crash occurs in the
4145 DB library. This is a regular occurrence (often caused by mis-matched
4148 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4149 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4150 picked up on a user's system that detects such things. There doesn't seem
4151 to be a gcc warning option for this - only an attribute that has to be
4152 put on the function's prototype. It seems that in Fedora Core 4 they have
4153 set this on a number of new functions. No doubt there will be more in due
4156 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4157 specify a transport. However, if an address that was verified by such a
4158 router was the subject of a callout, Exim crashed because it tried to
4159 read the rcpt_include_affixes from the non-existent transport. Now it
4160 just assumes that the setting of that option is false. This bug was
4161 introduced by 4.51/PH/31.
4163 PH/28 Changed -d+all to exclude +memory, because that information is very
4164 rarely of interest, but it makes the output a lot bigger. People tend to
4165 do -d+all out of habit.
4167 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4168 code in os-type was giving problems when libc.so lives in lib64, like on
4171 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4172 aren't the modern standard, and it seems that some systems' include files
4173 don't always have them. Exim was already checking for some of the newer
4174 ones like T_AAAA, and defining it itself. I've added checks for all the
4175 record types that Exim uses.
4177 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4178 not automatically generating a new one, as it is supposed to. This
4179 prevented TLS from working. If the file did exist, but contained invalid
4180 data, a new version was generated, as expected. It was only the case of a
4181 non-existent file that was broken.
4183 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4184 with a change in libdomainkeys > 0.64.
4186 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4187 from DNS. If the selector record carries the flag, it now has
4188 precedence over the domain-wide flag.
4190 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4192 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4193 the use of a port name that isn't defined in /etc/services) occurred, the
4194 message was deferred as in a normal delivery, and thus remained on the
4195 spool, instead of being failed because of the mua_wrapper setting. This
4196 is now fixed, and I tidied up some of the mua_wrapper messages at the
4199 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4200 the messages in a hash of arrays rather than using individual hashes.
4201 This is a bit cleaner and results in dramatic memory savings, albeit
4202 at a slight CPU cost.
4204 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4205 as requested by Marc Sherman.
4207 SC/10 Eximstats: added histograms for user specified patterns as requested
4210 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4212 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4213 fopen() in the content-scanning modules that did not already have it.
4219 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4220 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4222 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4224 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4226 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4227 to test Sieve filters that use "vacation".
4229 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4230 that changes the way the GnuTLS parameters are stored in the cache file.
4231 The new format can be generated externally. For backward compatibility,
4232 if the data in the cache doesn't make sense, Exim assumes it has read an
4233 old-format file, and it generates new data and writes a new file. This
4234 means that you can't go back to an older release without removing the
4237 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4238 work if there are any delivery delays because "one_time" forces the
4239 parent to be marked "delivered", so its unseen clone is never tried
4240 again. For this reason, Exim now forbids the simultaneous setting of
4243 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4244 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4245 entry if you want to know the details. The fix, however, neglected to
4246 consider the case when local delivery batching is involved. The test for
4247 "previously delivered" was not happening when checking to see if an
4248 address could be batched with a previous (undelivered) one; under
4249 certain circumstances this could lead to multiple deliveries to the same
4252 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4253 in its include files, and this causes problems building Exim.
4255 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4256 header_syntax) but Exim was just ignoring anything given after a slash.
4257 In particular, this caused confusion with an attempt to use "verify =
4258 reverse_host_lookup/defer_ok". An error is now given when options are
4259 supplied for verify items that do not have them. (Maybe reverse_host_
4260 lookup should have a defer_ok option, but that's a different point.)
4262 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4263 defined by RFC 821) to 2048, because there were problems with some AUTH
4264 commands, and RFC 1869 says the size should be increased for extended
4265 SMTP commands that take arguments.
4267 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4270 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4271 "unknown" error; now it says that the functionality isn't in the binary.
4273 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4274 an address' error message when a string expansion fails (syntax or
4275 whatever). Otherwise the password may appear in the log. Following change
4276 PH/42 below, there is no longer a chance of it appearing in a bounce
4279 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4281 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4282 of its addresses was ever tried. (Bugzilla bug #2.)
4284 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4285 the result incorrectly in the debug output. (It correctly added a newline
4286 to what was transported.)
4288 TF/01 Added $received_time.
4290 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4291 commented out examples of how to interface to a virus scanner and to
4292 SpamAssassin. Also added commented examples of av_scanner and
4293 spamd_address settings.
4295 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4296 and controls are allowed in which ACLs. There were a couple of minor
4297 errors. Some of the entries in the conditions table (which is a table of
4298 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4299 negation of where the condition IS allowed.
4301 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4303 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4304 header file does not have a version number, so I've had to invent a new
4305 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4306 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4307 radiusclient), but it was contributed by a Radius user.
4309 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4310 files or directories, for OpenSSL.
4312 PH/19 When an Exim process that is running as root has to create an Exim log
4313 file, it does so in a subprocess that runs as exim:exim so as to get the
4314 ownership right at creation (otherwise, other Exim processes might see
4315 the file with the wrong ownership). There was no test for failure of this
4316 fork() call, which would lead to the process getting stuck as it waited
4317 for a non-existent subprocess. Forks do occasionally fail when resources
4318 run out. I reviewed all the other calls to fork(); they all seem to check
4321 PH/20 When checking for unexpected SMTP input at connect time (before writing
4322 the banner), Exim was not dealing correctly with a non-positive return
4323 from the read() function. If the client had disconnected by this time,
4324 the result was a log entry for a synchronization error with an empty
4325 string after "input=" when read() returned zero. If read() returned -1
4326 (an event I could not check), uninitialized data bytes were printed.
4327 There were reports of junk text (parts of files, etc) appearing after
4330 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4332 PH/22 Added support for macro redefinition, and (re)definition in between
4333 driver and ACL definitions.
4335 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4336 forgetting to use the resulting value; it was using the unexpanded value.
4338 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4339 hadn't been configured. The fix is from Juergen Kreileder, who
4340 understands it better than I do:
4342 "Here's what I see happening with three configured cyrus_sasl
4343 authenticators configured (plain, login, cram-md5):
4345 On startup auth_cyrus_sasl_init() gets called for each of these.
4346 This means three calls to sasl_listmech() without a specified mech_list.
4347 => SASL tests which mechs of all available mechs actually work
4348 => three warnings about OTP not working
4349 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4351 With the patch, sasl_listmech() also gets called three times. But now
4352 SASL's mech_list option is set to the server_mech specified in the the
4353 authenticator. Or in other words, the answer from sasl_listmech()
4354 gets limited to just the mech you're testing for (which is different
4356 => the return list contains just 'plain' or 'login', 'cram-md5' or
4357 nothing depending on the value of ob->server_mech.
4359 I've just tested the patch: Authentication still works fine,
4360 unavailable mechs specified in the exim configuration are still
4361 caught, and the auth.log warnings about OTP are gone."
4363 PH/25 When debugging is enabled, the contents of the command line are added
4364 to the debugging output, even when log_selector=+arguments is not
4367 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4368 answer is "GNU", and only if the return is "GNU/something" is the answer
4371 PH/27 $acl_verify_message is now set immediately after the failure of a
4372 verification in an ACL, and so is available in subsequent modifiers. In
4373 particular, the message can be preserved by coding like this:
4375 warn !verify = sender
4376 set acl_m0 = $acl_verify_message
4378 Previously, $acl_verify_message was set only while expanding "message"
4379 and "log_message" when a very denied access.
4381 PH/28 Modified OS/os.c-Linux with
4383 -#ifndef OS_LOAD_AVERAGE
4384 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4386 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4387 nomenclature these days.)
4389 PH/29 Installed patch from the Sieve maintainer that adds the options
4390 sieve_useraddress and sieve_subaddress to the redirect router.
4392 PH/30 In these circumstances:
4393 . Two addresses routed to the same list of hosts;
4394 . First host does not offer TLS;
4395 . First host accepts first address;
4396 . First host gives temporary error to second address;
4397 . Second host offers TLS and a TLS session is established;
4398 . Second host accepts second address.
4399 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4400 and peerdn, if requested) that were in fact used only for the second
4403 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4404 attention to any local part prefix or suffix that was matched by the
4405 router that accepted the address. It now behaves in the same way as it
4406 does for delivery: the affixes are removed from the local part unless
4407 rcpt_include_affixes is set on the transport.
4409 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4410 timeout during the DATA phase of an incoming message.
4412 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4413 applied a patch sanctioned by the Sieve maintainer.
4415 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4416 the uid or gid is negative. A case of a negative gid caused this to be
4417 noticed. The fix allows for either to be negative.
4419 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4420 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4423 PH/36 The change PH/12 above was broken. Fixed it.
4425 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4426 the grounds that routing the same address twice would always produce the
4427 same answer. This might have been true once, but it is certainly no
4428 longer true now. Routing a child address may depend on the previous
4429 routing that produced that child. Some complicated redirection strategies
4430 went wrong when messages had multiple recipients, and made Exim's
4431 behaviour dependent on the order in which the addresses were given.
4433 I have moved the duplicate checking until after the routing is complete.
4434 Exim scans the addresses that are assigned to local and remote
4435 transports, and removes any duplicates. This means that more work will be
4436 done, as duplicates will always all be routed, but duplicates are
4437 presumably rare, so I don't expect this is of any significance.
4439 For deliveries to pipes, files, and autoreplies, the duplicate checking
4440 still happens during the routing process, since they are not going to be
4443 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4444 It corrects a timeout issue with spamd. This is Ian's comment: "The
4445 background is that sometimes spamd either never reads data from a
4446 connection it has accepted, or it never writes response data. The exiscan
4447 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4448 blindly assumes that writes won't block so it may never time out."
4450 PH/39 Allow G after quota size as well as K and M.
4452 PH/40 The value set for $authenticated_id in an authenticator may not contain
4453 binary zeroes or newlines because the value is written to log lines and
4454 to spool files. There was no check on this. Now the value is run through
4455 the string_printing() function so that such characters are converted to
4456 printable escape sequences.
4458 PH/41 $message_linecount is a new variable that contains the total number of
4459 lines in the message. Compare $body_linecount, which is the count for the
4462 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4463 bounce and delay warning messages, except in certain special cases, which
4466 (a) An SMTP error message from a remote host;
4467 (b) A message specified in a :fail: redirection;
4468 (c) A message specified in a "fail" command in a system filter;
4469 (d) A message specified in a FAIL return from the queryprogram router;
4470 (e) A message specified by the cannot_route_message router option.
4472 In these cases only, Exim does include the error details in bounce and
4473 warning messages. There are also a few cases where bland messages such
4474 as "unrouteable address" or "local delivery error" are given.
4476 PH/43 $value is now also set for the "else" part of a ${run expansion.
4478 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4479 being worked on, but at least Exim now implements the latest version to
4482 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4483 process to complete was treated as a delivery failure, a timeout while
4484 writing the message to the pipe was logged, but erroneously treated as a
4485 successful delivery. Such timeouts include transport filter timeouts. For
4486 consistency with the overall process timeout, these timeouts are now
4487 treated as errors, giving rise to delivery failures by default. However,
4488 there is now a new Boolean option for the pipe transport called
4489 timeout_defer, which, if set TRUE, converts the failures into defers for
4490 both kinds of timeout. A transport filter timeout is now identified in
4493 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4494 systems where "make" and "gmake" are different, calling "gmake" at top
4495 level broke things. I've arranged for the value of $(MAKE) to be passed
4496 from the Makefile to this script so that it can call the same version of
4500 A note about Exim versions 4.44 and 4.50
4501 ----------------------------------------
4503 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4504 changes of various kinds. As a consequence, a big documentation update was
4505 needed. This delayed the release for rather longer than seemed good, especially
4506 in the light of a couple of (minor) security issues. Therefore, the changes
4507 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4508 release. So 4.44 and 4.50 are in effect two different branches that both start
4511 I have left the 4.50 change log unchanged; it contains all the changes since
4512 4.43. The change log for 4.44 is below; many of its items are identical to
4513 those for 4.50. This seems to be the most sensible way to preserve the
4514 historical information.
4520 1. Minor wording change to the doc/README.SIEVE file.
4522 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4523 computation of the current number of files was incorrect.
4525 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4526 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4529 4. Give more explanation in the error message when the command for a transport
4530 filter fails to execute.
4532 5. There are several places where Exim runs a non-Exim command in a
4533 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4534 was being done only for the command run by the queryprogram router. It is
4535 now done for all such subprocesses. The other cases are: ${run, transport
4536 filters, and the commands run by the lmtp and pipe transports.
4538 6. Added CONFIGURE_GROUP build-time option.
4540 7. Some older OS have a limit of 256 on the maximum number of file
4541 descriptors. Exim was using setrlimit() to set 1000 as a large value
4542 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4543 systems. I've change it so that if it can't get 1000, it tries for 256.
4545 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4546 was an oversight, and furthermore, ever since the addition of extra
4547 controls (e.g. 4.43/32), the checks on when to allow different forms of
4548 "control" were broken. There should now be diagnostics for all cases when a
4549 control that does not make sense is encountered.
4551 9. Added the /retain_sender option to "control=submission".
4553 10. $recipients is now available in the predata ACL (oversight).
4555 11. Tidy the search cache before the fork to do a delivery from a message
4556 received from the command line. Otherwise the child will trigger a lookup
4557 failure and thereby defer the delivery if it tries to use (for example) a
4558 cached ldap connection that the parent has called unbind on.
4560 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4561 of $address_data from the recipient verification was clobbered by the
4562 sender verification.
4564 13. The value of address_data from a sender verification is now available in
4565 $sender_address_data in subsequent conditions in the ACL statement.
4567 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4569 15. Added a new option "connect=<time>" to callout options, to set a different
4572 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4573 was its contents. (It was OK if the option was not defined at all.)
4575 17. A "Completed" log line is now written for messages that are removed from
4576 the spool by the -Mrm option.
4578 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4579 information about exactly what failed.
4581 19. Added -dd to debug only the daemon process.
4583 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4584 handles timeouts, both on the server side and network timeouts. Renamed the
4585 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4587 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4588 It is now set to "smtps".
4590 22. $host_address is now set to the target address during the checking of
4591 ignore_target_hosts.
4593 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4594 being passed; this would have caused $sender_host_name to have been used if
4595 matching the list had actually called for a host name (not very likely,
4596 since this list is usually IP addresses). A host name is now passed as
4599 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4600 code that specifies a non-restarting handler (typically sigaction() in
4601 modern systems) in an attempt to fix a rare and obscure crash bug.
4603 25. Narrowed the window for a race in the daemon that could cause it to ignore
4604 SIGCHLD signals. This is not a major problem, because they are used only to
4605 wake it up if nothing else does.
4607 26. A malformed maildirsize file could cause Exim to calculate negative values
4608 for the mailbox size or file count. Odd effects could occur as a result.
4609 The maildirsize information is now recalculated if the size or filecount
4612 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4613 support for a long time. Removed HAVE_SYS_VFS_H.
4615 28. Installed the latest version of exipick from John Jetmore.
4617 29. In an address list, if the pattern was not a regular expression, an empty
4618 subject address (from a bounce message) matched only if the pattern was an
4619 empty string. Non-empty patterns were not even tested. This was the wrong
4620 because it is perfectly reasonable to use an empty address as part of a
4621 database query. An empty address is now tested by patterns that are
4622 lookups. However, all the other forms of pattern expect the subject to
4623 contain a local part and a domain, and therefore, for them, an empty
4624 address still always fails if the pattern is not itself empty.
4626 30. Exim went into a mad DNS loop when attempting to do a callout where the
4627 host was specified on an smtp transport, and looking it up yielded more
4628 than one IP address.
4630 31. Re-factored the code for checking spool and log partition space into a
4631 function that finds that data and another that does the check. The former
4632 is then used to implement four new variables: $spool_space, $log_space,
4633 $spool_inodes, and $log_inodes.
4635 32. The RFC2047 encoding function was originally intended for short strings
4636 such as real names; it was not keeping to the 75-character limit for
4637 encoded words that the RFC imposes. It now respects the limit, and
4638 generates multiple encoded words if necessary. To be on the safe side, I
4639 have increased the buffer size for the ${rfc2047: expansion operator from
4642 33. It is now permitted to omit both strings after an "if" condition; if the
4643 condition is true, the result is "true". As before, when the second string
4644 is omitted, a false condition yields an empty string. This makes it less
4645 cumbersome to write custom ACL and router conditions.
4647 34. Failure to deliver a bounce message always caused it to be frozen, even if
4648 there was an errors_to setting on the router. The errors_to setting is now
4651 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4652 canonical form (fully expanded) before being placed in
4653 $sender_host_address.
4655 36. The table in the code that translates DNS record types into text (T_A to
4656 "A" for instance) was missing entries for NS and CNAME. It is just possible
4657 that this could have caused confusion if both these types were looked up
4658 for the same domain, because the text type is used as part of Exim's
4659 per-process caching. But the chance of anyone hitting this buglet seems
4662 37. The dnsdb lookup has been extended in a number of ways.
4664 (1) There is a new type, "zns", which walks up the domain tree until it
4665 finds some nameserver records. It should be used with care.
4667 (2) There is a new type, "mxh", which is like "mx" except that it returns
4668 just the host names, not the priorities.
4670 (3) It is now possible to give a list of domains (or IP addresses) to be
4671 looked up. The behaviour when one of the lookups defers can be
4672 controlled by a keyword.
4674 (4) It is now possible to specify the separator character for use when
4675 multiple records are returned.
4677 38. The dnslists ACL condition has been extended: it is now possible to supply
4678 a list of IP addresses and/or domains to be looked up in a particular DNS
4681 39. Added log_selector=+queue_time_overall.
4683 40. When running the queue in the test harness, wait just a tad after forking a
4684 delivery process, to get repeatability of debugging output.
4686 41. Include certificate and key file names in error message when GnuTLS fails
4687 to set them up, because the GnuTLS error message doesn't include the name
4688 of the failing file when there is a problem reading it.
4690 42. Allow both -bf and -bF in the same test run.
4692 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4694 44. The "Exiscan patch" is now merged into the mainline Exim source.
4696 45. Sometimes the final signoff response after QUIT could fail to get
4697 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4698 < 0 before doing a fflush(). This bug looks as though it goes back to the
4699 introduction of TLS in release 3.20, but "sometimes" must have been rare
4700 because the tests only now provoked it.
4702 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4703 (this can affect the format of dates).
4705 47. exim_tidydb, when checking for the continued existence of a message for
4706 which it has found a message-specific retry record, was not finding
4707 messages that were in split spool directories. Consequently, it was
4708 deleting retry records that should have stayed in existence.
4710 48. Steve fixed some bugs in eximstats.
4712 49. The SPA authentication driver was not abandoning authentication and moving
4713 on to the next authenticator when an expansion was forced to fail,
4714 contradicting the general specification for all authenticators. Instead it
4715 was generating a temporary error. It now behaves as specified.
4717 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4718 (the order specifies the preference for clients). The order is now AES256,
4719 AES128, 3DES, ARCFOUR128.
4721 51. Small patch to Sieve code - explicitly set From: when generating an
4724 52. Exim crashed if a remote delivery caused a very long error message to be
4725 recorded - for instance if somebody sent an entire SpamAssassin report back
4726 as a large number of 550 error lines. This bug was coincidentally fixed by
4727 increasing the size of one of Exim's internal buffers (big_buffer) that
4728 happened as part of the Exiscan merge. However, to be on the safe side, I
4729 have made the code more robust (and fixed the comments that describe what
4732 53. Now that there can be additional text after "Completed" in log lines (if
4733 the queue_time_overall log selector is set), a one-byte patch to exigrep
4734 was needed to allow it to recognize "Completed" as not the last thing in
4737 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4738 patch that reportedly fixes this has been added. I am not expert enough to
4739 create a test for it. This is what the patch creator wrote:
4741 "I found a little strange behaviour of ldap code when working with
4742 Windows 2003 AD Domain, where users was placed in more than one
4743 Organization Units. When I tried to give exim partial DN, the exit code
4744 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4745 But simultaneously result of request was absolutely normal ldap result,
4746 so I produce this patch..."
4748 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4749 REFERENCE, so I have modified the code to exclude the patch when that macro
4752 55. Some experimental protocols are using DNS PTR records for new purposes. The
4753 keys for these records are domain names, not reversed IP addresses. The
4754 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4755 leaves it alone. Component reversal etc. now happens only for IP addresses.
4758 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4760 57. Double the size of the debug message buffer (to 2048) so that more of very
4761 long debug lines gets shown.
4763 58. The exicyclog utility now does better if the number of log files to keep
4764 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4766 59. Two changes related to the smtp_active_hostname option:
4768 (1) $smtp_active_hostname is now available as a variable.
4769 (2) The default for smtp_banner uses $smtp_active_hostname instead
4770 of $primary_hostname.
4772 60. The host_aton() function is supposed to be passed a string that is known
4773 to be a valid IP address. However, in the case of IPv6 addresses, it was
4774 not checking this. This is a hostage to fortune. Exim now panics and dies
4775 if the condition is not met. A case was found where this could be provoked
4776 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4777 components; fortuitously, this particular loophole had already been fixed
4778 by change 4.50/55 above.
4780 If there are any other similar loopholes, the new check in host_aton()
4781 itself should stop them being exploited. The report I received stated that
4782 data on the command line could provoke the exploit when Exim was running as
4783 exim, but did not say which command line option was involved. All I could
4784 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4785 running as the user.
4788 61. There was a buffer overflow vulnerability in the SPA authentication code
4789 (which came originally from the Samba project). I have added a test to the
4790 spa_base64_to_bits() function which I hope fixes it.
4793 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4794 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4796 63. The daemon start-up calls getloadavg() while still root for those OS that
4797 need the first call to be done as root, but it missed one case: when
4798 deliver_queue_load_max is set with deliver_drop_privilege. This is
4799 necessary for the benefit of the queue runner, because there is no re-exec
4800 when deliver_drop_privilege is set.
4802 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4803 This has been fixed.
4805 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4806 was in use, was not putting the data itself into the right store pool;
4807 consequently, it could be overwritten for a subsequent message in the same
4808 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4811 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4813 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4814 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4815 the function were treating the return as a boolean value, which happened to
4816 work because 0=false and not-0=true, but is not correct code.
4818 68. The host_aton() function was not handling scoped IPv6 addresses (those
4819 with, for example, "%eth0" on the end) correctly.
4821 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4822 negated items (that is, ~something) in unsigned ints. Some compilers
4823 apparently mutter when there is no cast.
4825 70. If an address verification called from an ACL failed, and did not produce a
4826 user-specific message (i.e. there was only a "system" message), nothing was
4827 put in $acl_verify_message. In this situation, it now puts the system
4830 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4831 session; change 4.31/43 added the unwanted input to the log line - except
4832 that it did not do this in the start of session case. It now does.
4834 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4835 This is wrong and can cause the other end to generate a synchronization
4836 error if it is another Exim or anything else that does the synchronization
4837 check. A QUIT command is no longer sent after a timeout.
4839 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4840 during host lookups.
4842 74. The defer_ok option of callout verification was not working if it was used
4843 when verifying addresses in header lines, that is, for this case:
4845 verify = header_sender/callout=defer_ok
4847 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4848 those file descriptors could be used for SMTP connections. If anything
4849 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4850 could be sent to the SMTP client, causing chaos. The daemon now opens
4851 stdin, stdout, and stderr to /dev/null when it puts itself into the
4854 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4855 log by default. The user can override this with suitable Perl magic.
4857 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4858 the log message when discard triggers, was not working for the DATA ACL or
4859 for the non-SMTP ACL.
4861 78. Error message wording change in sieve.c.
4863 79. If smtp_accept_max_per_host was set, the number of connections could be
4864 restricted to fewer than expected, because the daemon was trying to set up
4865 a new connection before checking whether the processes handling previous
4866 connections had finished. The check for completed processes is now done
4867 earlier. On busy systems, this bug wouldn't be noticed because something
4868 else would have woken the daemon, and it would have reaped the completed
4871 80. If a message was submitted locally by a user whose login name contained one
4872 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4873 It caused a spool format error. I have fixed the spool reading code. A
4874 related problem was that the "from" clause in the Received: line became
4875 illegal because of the space(s). It is now covered by ${quote_local_part.
4877 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4880 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4882 83. Patch from Sieve maintainer to fix unterminated string problem in
4883 "vacation" handling.
4885 84. Some minor changes to the Linux configuration files to help with other
4886 OS variants using glibc.
4888 85. One more patch for Sieve to update vacation handling to latest spec.
4891 ----------------------------------------------------
4892 See the note above about the 4.44 and 4.50 releases.
4893 ----------------------------------------------------
4899 1. Change 4.43/35 introduced a bug that caused file counts to be
4900 incorrectly computed when quota_filecount was set in an appendfile
4903 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4904 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4907 3. Give more explanation in the error message when the command for a transport
4908 filter fails to execute.
4910 4. There are several places where Exim runs a non-Exim command in a
4911 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4912 was being done only for the command run by the queryprogram router. It is
4913 now done for all such subprocesses. The other cases are: ${run, transport
4914 filters, and the commands run by the lmtp and pipe transports.
4916 5. Some older OS have a limit of 256 on the maximum number of file
4917 descriptors. Exim was using setrlimit() to set 1000 as a large value
4918 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4919 systems. I've change it so that if it can't get 1000, it tries for 256.
4921 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4922 was an oversight, and furthermore, ever since the addition of extra
4923 controls (e.g. 4.43/32), the checks on when to allow different forms of
4924 "control" were broken. There should now be diagnostics for all cases when a
4925 control that does not make sense is encountered.
4927 7. $recipients is now available in the predata ACL (oversight).
4929 8. Tidy the search cache before the fork to do a delivery from a message
4930 received from the command line. Otherwise the child will trigger a lookup
4931 failure and thereby defer the delivery if it tries to use (for example) a
4932 cached ldap connection that the parent has called unbind on.
4934 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4935 of $address_data from the recipient verification was clobbered by the
4936 sender verification.
4938 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4939 was its contents. (It was OK if the option was not defined at all.)
4941 11. A "Completed" log line is now written for messages that are removed from
4942 the spool by the -Mrm option.
4944 12. $host_address is now set to the target address during the checking of
4945 ignore_target_hosts.
4947 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4948 being passed; this would have caused $sender_host_name to have been used if
4949 matching the list had actually called for a host name (not very likely,
4950 since this list is usually IP addresses). A host name is now passed as
4953 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4954 code that specifies a non-restarting handler (typically sigaction() in
4955 modern systems) in an attempt to fix a rare and obscure crash bug.
4957 15. Narrowed the window for a race in the daemon that could cause it to ignore
4958 SIGCHLD signals. This is not a major problem, because they are used only to
4959 wake it up if nothing else does.
4961 16. A malformed maildirsize file could cause Exim to calculate negative values
4962 for the mailbox size or file count. Odd effects could occur as a result.
4963 The maildirsize information is now recalculated if the size or filecount
4966 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4967 support for a long time. Removed HAVE_SYS_VFS_H.
4969 18. Updated exipick to current release from John Jetmore.
4971 19. Allow an empty sender to be matched against a lookup in an address list.
4972 Previously the only cases considered were a regular expression, or an
4975 20. Exim went into a mad DNS lookup loop when doing a callout where the
4976 host was specified on the transport, if the DNS lookup yielded more than
4979 21. The RFC2047 encoding function was originally intended for short strings
4980 such as real names; it was not keeping to the 75-character limit for
4981 encoded words that the RFC imposes. It now respects the limit, and
4982 generates multiple encoded words if necessary. To be on the safe side, I
4983 have increased the buffer size for the ${rfc2047: expansion operator from
4986 22. Failure to deliver a bounce message always caused it to be frozen, even if
4987 there was an errors_to setting on the router. The errors_to setting is now
4990 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4991 canonical form (fully expanded) before being placed in
4992 $sender_host_address.
4994 24. Updated eximstats to version 1.33
4996 25. Include certificate and key file names in error message when GnuTLS fails
4997 to set them up, because the GnuTLS error message doesn't include the name
4998 of the failing file when there is a problem reading it.
5000 26. Expand error message when OpenSSL has problems setting up cert/key files.
5003 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5004 (this can affect the format of dates).
5006 28. exim_tidydb, when checking for the continued existence of a message for
5007 which it has found a message-specific retry record, was not finding
5008 messages that were in split spool directories. Consequently, it was
5009 deleting retry records that should have stayed in existence.
5011 29. eximstats updated to version 1.35
5012 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5013 1.35 - bugfix such that pie charts by volume are generated correctly
5015 30. The SPA authentication driver was not abandoning authentication and moving
5016 on to the next authenticator when an expansion was forced to fail,
5017 contradicting the general specification for all authenticators. Instead it
5018 was generating a temporary error. It now behaves as specified.
5020 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5021 (the order specifies the preference for clients). The order is now AES256,
5022 AES128, 3DES, ARCFOUR128.
5024 31. Small patch to Sieve code - explicitly set From: when generating an
5027 32. Exim crashed if a remote delivery caused a very long error message to be
5028 recorded - for instance if somebody sent an entire SpamAssassin report back
5029 as a large number of 550 error lines. This bug was coincidentally fixed by
5030 increasing the size of one of Exim's internal buffers (big_buffer) that
5031 happened as part of the Exiscan merge. However, to be on the safe side, I
5032 have made the code more robust (and fixed the comments that describe what
5035 33. Some experimental protocols are using DNS PTR records for new purposes. The
5036 keys for these records are domain names, not reversed IP addresses. The
5037 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5038 leaves it alone. Component reversal etc. now happens only for IP addresses.
5041 34. The host_aton() function is supposed to be passed a string that is known
5042 to be a valid IP address. However, in the case of IPv6 addresses, it was
5043 not checking this. This is a hostage to fortune. Exim now panics and dies
5044 if the condition is not met. A case was found where this could be provoked
5045 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5046 components; fortuitously, this particular loophole had already been fixed
5047 by change 4.50/55 or 4.44/33 above.
5049 If there are any other similar loopholes, the new check in host_aton()
5050 itself should stop them being exploited. The report I received stated that
5051 data on the command line could provoke the exploit when Exim was running as
5052 exim, but did not say which command line option was involved. All I could
5053 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5054 running as the user.
5057 35. There was a buffer overflow vulnerability in the SPA authentication code
5058 (which came originally from the Samba project). I have added a test to the
5059 spa_base64_to_bits() function which I hope fixes it.
5062 36. The daemon start-up calls getloadavg() while still root for those OS that
5063 need the first call to be done as root, but it missed one case: when
5064 deliver_queue_load_max is set with deliver_drop_privilege. This is
5065 necessary for the benefit of the queue runner, because there is no re-exec
5066 when deliver_drop_privilege is set.
5068 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5069 was in use, was not putting the data itself into the right store pool;
5070 consequently, it could be overwritten for a subsequent message in the same
5071 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5074 38. Sometimes the final signoff response after QUIT could fail to get
5075 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5076 < 0 before doing a fflush(). This bug looks as though it goes back to the
5077 introduction of TLS in release 3.20, but "sometimes" must have been rare
5078 because the tests only now provoked it.
5084 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5085 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5086 Now it returns an integer. A number of calls were still expecting a T/F
5087 return. Fortuitously, in all cases, the tests worked in OK situations,
5088 which is the norm. However, things would have gone wrong on any write
5089 failures on the smtp file descriptor. This function is used when sending
5090 messages over SMTP and also when doing verify callouts.
5092 2. When Exim is called to do synchronous delivery of a locally submitted
5093 message (the -odf or -odi options), it no longer closes stderr before doing
5096 3. Implemented the mua_wrapper option.
5098 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5100 5. Implemented the functions header_remove(), header_testname(),
5101 header_add_at_position(), and receive_remove_recipient(), and exported them
5104 6. If an ACL "warn" statement specified the addition of headers, Exim already
5105 inserted X-ACL-Warn: at the start if there was no header name. However, it
5106 was not making this test for the second and subsequent header lines if
5107 there were newlines in the string. This meant that an invalid header could
5108 be inserted if Exim was badly configured.
5110 7. Allow an ACL "warn" statement to add header lines at the start or after all
5111 the Received: headers, as well as at the end.
5113 8. Added the rcpt_4xx retry error code.
5115 9. Added postmaster_mailfrom=xxx to callout verification option.
5117 10. Added mailfrom=xxxx to the callout verification option, for verify=
5120 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5121 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5123 12. Inserted some casts to stop certain compilers warning when using pointer
5124 differences as field lengths or precisions in printf-type calls (mostly
5125 affecting debugging statements).
5127 13. Added optional readline() support for -be (dynamically loaded).
5129 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5130 same clock tick as a message's arrival, so that its received time was the
5131 same as the "first fail" time on the retry record, and that message
5132 remained on the queue past the ultimate address timeout, every queue runner
5133 would try a delivery (because it was past the ultimate address timeout) but
5134 after another failure, the ultimate address timeout, which should have then
5135 bounced the address, did not kick in. This was a "< instead of <=" error;
5136 in most cases the first failure would have been in the next clock tick
5137 after the received time, and all would be well.
5139 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5140 being recognized when the domain list was tested by the match_domain
5141 condition in an expansion string.
5143 16. Added the ${str2b64: operator.
5145 17. Exim was always calling setrlimit() to set a large limit for the number of
5146 processes, without checking whether the existing limit was already
5147 adequate. (It did check for the limit on file descriptors.) Furthermore,
5148 errors from getrlimit() and setrlimit() were being ignored. Now they are
5149 logged to the main and panic logs, but Exim does carry on, to try to do its
5150 job under whatever limits there are.
5152 18. Imported PCRE 5.0.
5154 19. Trivial typo in log message " temporarily refused connection" (the leading
5157 20. If the log selector return_path_on_delivery was set and an address was
5158 redirected to /dev/null, the delivery process crashed because it assumed
5159 that a return path would always be set for a "successful" delivery. In this
5160 case, the whole delivery is bypassed as an optimization, and therefore no
5163 21. Internal re-arrangement: the function for sending a challenge and reading
5164 a response while authentication was assuming a zero-terminated challenge
5165 string. It's now changed to take a pointer and a length, to allow for
5166 binary data in such strings.
5168 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5170 23. Exim was not respecting finduser_retries when seeking the login of the
5171 uid under which it was called; it was always trying 10 times. (The default
5172 setting of finduser_retries is zero.) Also, it was sleeping after the final
5173 failure, which is pointless.
5175 24. Implemented tls_on_connect_ports.
5177 25. Implemented acl_smtp_predata.
5179 26. If the domain in control=submission is set empty, Exim assumes that the
5180 authenticated id is a complete email address when it generates From: or
5181 Sender: header lines.
5183 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5184 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5185 chown and chgrp in /bin and hostname in /usr/bin.
5187 28. Exim was keeping the "process log" file open after each use, just as it
5188 does for the main log. This opens the possibility of it remaining open for
5189 long periods when the USR1 signal hits a daemon. Occasional processlog
5190 errors were reported, that could have been caused by this. Anyway, it seems
5191 much more sensible not to leave this file open at all, so that is what now
5194 29. The long-running daemon process does not normally write to the log once it
5195 has entered its main loop, and it closes the log before doing so. This is
5196 so that log files can straightforwardly be renamed and moved. However,
5197 there are a couple of unusual error situations where the daemon does write
5198 log entries, and I had neglected to close the log afterwards.
5200 30. The text of an SMTP error response that was received during a remote
5201 delivery was being truncated at 512 bytes. This is too short for some of
5202 the long messages that one sometimes sees. I've increased the limit to
5205 31. It is now possible to make retry rules that apply only when a message has a
5206 specific sender, in particular, an empty sender.
5208 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5209 it possible to be selective about when SMTP synchronization is enforced.
5211 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5213 32. Implemented hosts_connection_nolog.
5215 33. Added an ACL for QUIT.
5217 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5220 35. Added mailbox_size and mailbox_filecount to appendfile.
5222 36. Added control = no_multiline_responses to ACLs.
5224 37. There was a bug in the logic of the code that waits for the clock to tick
5225 in the case where the clock went backwards by a substantial amount such
5226 that the microsecond fraction of "now" was more than the microsecond
5227 fraction of "then" (but the whole seconds number was less).
5229 38. Added support for the libradius Radius client library this is found on
5230 FreeBSD (previously only the radiusclient library was supported).
5236 1. When certain lookups returned multiple values in the form name=value, the
5237 quoting of the values was not always being done properly. Specifically:
5238 (a) If the value started with a double quote, but contained no whitespace,
5240 (b) If the value contained whitespace other than a space character (i.e.
5241 tabs or newlines or carriage returns) it was not quoted.
5242 This fix has been applied to the mysql and pgsql lookups by writing a
5243 separate quoting function and calling it from the lookup code. The fix
5244 should probably also be applied to nisplus, ibase and oracle lookups, but
5245 since I cannot test any of those, I have not disturbed their existing code.
5247 2. A hit in the callout cache for a specific address caused a log line with no
5248 reason for rejecting RCPT. Now it says "Previous (cached) callout
5249 verification failure".
5251 3. There was an off-by-one bug in the queryprogram router. An over-long
5252 return line was truncated at 256 instead of 255 characters, thereby
5253 overflowing its buffer with the terminating zero. As well as fixing this, I
5254 have increased the buffer size to 1024 (and made a note to document this).
5256 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5257 when Exim is waiting for an SMTP response from a remote server, Exim
5258 restarts its select() call on the socket, thereby resetting its timeout.
5259 This is not a problem when such interrupts are rare. Somebody set up a cron
5260 job to run exiwhat every 2 minutes, which is less than the normal select()
5261 timeout (5 or 10 minutes). This meant that the select() timeout never
5262 kicked in because it was always reset. I have fixed this by comparing the
5263 time when an interrupt arrives with the time at the start of the first call
5264 to select(). If more time than the timeout has elapsed, the interrupt is
5265 treated as a timeout.
5267 5. Some internal re-factoring in preparation for the addition of Sieve
5268 extensions (by MH). In particular, the "personal" test is moved to a
5269 separate function, and given an option for scanning Cc: and Bcc: (which is
5270 not set for Exim filters).
5272 6. When Exim created an email address using the login of the caller as the
5273 local part (e.g. when creating a From: or Sender: header line), it was not
5274 quoting the local part when it contained special characters such as @.
5276 7. Installed new OpenBSD configuration files.
5278 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5279 try to make them clearer.
5281 9. Callout options, other than the timeout value, were being ignored when
5282 verifying sender addresses in header lines. For example, when using
5284 verify = header_sender/callout=no_cache
5286 the cache was (incorrectly) being used.
5288 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5289 only the Cygwin environment.
5291 11. When return_path_on_delivery was set as a log selector, if different remote
5292 addresses in the same message used different return paths and parallel
5293 remote delivery occurred, the wrong values would sometimes be logged.
5294 (Whenever a remote delivery process finished, the return path value from
5295 the most recently started remote delivery process was logged.)
5297 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5298 lines when AUTH and/or TLS are in use. This is the "received protocol"
5299 field. Exim used to use "asmtp" for authenticated SMTP, without any
5300 indication (in the protocol name) for TLS use. Now it follows the RFC and
5301 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5302 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5303 names appear in log lines as well as in Received: header lines.
5305 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5306 extensions, and comparison tests, and to fix some bugs.
5308 14. Changes to the "personal" filter test:
5310 (1) The test was buggy in that it was just doing the equivalent of
5311 "contains" tests on header lines. For example, if a user's address was
5312 anne@some.where, the "personal" test would incorrectly be true for
5314 To: susanne@some.where
5316 This test is now done by extracting each address from the header in turn,
5317 and checking the entire address. Other tests that are part of "personal"
5318 are now done using regular expressions (for example, to check local parts
5319 of addresses in From: header lines).
5321 (2) The list of non-personal local parts in From: addresses has been
5322 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5323 taken from the Sieve specification recommendations.
5325 (3) If the message contains any header line starting with "List-" it is
5326 treated as non-personal.
5328 (4) The test for "circular" in the Subject: header line has been removed
5329 because it now seems ill-conceived.
5331 15. Minor typos in src/EDITME comments corrected.
5333 16. Installed latest exipick from John Jetmore.
5335 17. If headers_add on a router specified a text string that was too long for
5336 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5337 of string_sprintf() is now avoided.
5339 18. $message_body_size was not set (it was always zero) when running the DATA
5340 ACL and the local_scan() function.
5342 19. For the "mail" command in an Exim filter, no default was being set for
5343 the once_repeat time, causing a random time value to be used if "once" was
5344 specified. (If the value happened to be <= 0, no repeat happened.) The
5345 default is now 0s, meaning "never repeat". The "vacation" command was OK
5346 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5347 (I found it when inspecting the code).
5349 20. There is now an overall timeout for performing a callout verification. It
5350 defaults to 4 times the callout timeout, which applies to individual SMTP
5351 commands during the callout. The overall timeout applies when there is more
5352 than one host that can be tried. The timeout is checked before trying the
5353 next host. This prevents very long delays if there are a large number of
5354 hosts and all are timing out (e.g. when the network connections are timing
5355 out). The value of the overall timeout can be changed by specifying an
5356 additional sub-option for "callout", called "maxwait". For example:
5358 verify = sender/callout=5s,maxwait=20s
5360 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5361 to the end before writing, but this should make it even safer).
5363 22. Exim was forgetting that it had advertised PIPELINING for the second and
5364 subsequent messages on an SMTP connection. It was also not resetting its
5365 memory on STARTTLS and an internal HELO.
5367 23. When Exim logs an SMTP synchronization error within a session, it now
5368 records whether PIPELINING has been advertised or not.
5370 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5371 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5372 rather than long int.
5374 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5376 26. Added the never_mail option to autoreply.
5382 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5383 crash if the getsockname() call failed; this can happen if a connection is
5384 closed very soon after it is established. The problem was simply in the
5385 order in which certain operations were done, causing Exim to try to write
5386 to the SMTP stream before it had set up the file descriptor. The bug has
5387 been fixed by making things happen in the correct order.
5393 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5394 before the connection was closed, thus losing the rejection response.
5396 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5397 some early Solaris releases, but causes trouble in current releases where
5398 socklen_t is defined.
5400 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5403 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5406 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5407 that releasing the top of it at the end releases what was used for sub-
5408 expansions (unless the block got too big). However, discard this block if
5409 the first thing is a variable or header, so that we can use its block when
5410 it is dynamic (useful for very large $message_headers, for example).
5412 6. Lookups now cache *every* query, not just the most recent. A new, separate
5413 store pool is used for this. It can be recovered when all lookup caches are
5414 flushed. Lookups now release memory at the end of their result strings.
5415 This has involved some general refactoring of the lookup sources.
5417 7. Some code has been added to the store_xxx() functions to reduce the amount
5418 of flapping under certain conditions.
5420 8. log_incoming_interface used to affect only the <= reception log lines. Now
5421 it causes the local interface and port to be added to several more SMTP log
5422 lines, for example "SMTP connection from", and rejection lines.
5424 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5426 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5428 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5429 could be overwritten at the end of the current message (or the start of a
5430 new message if it was set in a HELO ACL). The value is now preserved for
5431 the duration of the SMTP connection.
5433 12. If a transport had a headers_rewrite setting, and a matching header line
5434 contained an unqualified address, that address was qualified, even if it
5435 did not match any rewriting rules. The underlying bug was that the values
5436 of the flags that permit the existence of unqualified sender and recipient
5437 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5438 non-local messages, and by -bnq for local messages) were not being
5439 preserved with the message after it was received.
5441 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5442 "next input=" as part of the text comprising the host identity instead of
5443 the correct text. The code was using the same buffer for two different
5444 strings. However, depending on which order the printing function evaluated
5445 its arguments, the bug did not always show up. Under Linux, for example, my
5446 test suite worked just fine.
5448 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5449 doesn't work with some older versions of Perl. It has been changed to "my",
5450 which in any case is probably the better facility to use.
5452 15. A really picky compiler found some instances of statements for creating
5453 error messages that either had too many or two few arguments for the format
5456 16. The size of the buffer for calls to the DNS resolver has been increased
5457 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5458 for addresses that have a lot of PTR records. This alleviates a problem; it
5459 does not fully solve it.
5461 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5462 buffer now truncates the list and logs the incident, which is the same
5463 action as happens when Exim is looking up a host name and its aliases.
5464 Previously in this situation something unpredictable would happen;
5465 sometimes it was "internal error: store_reset failed".
5467 18. If a server dropped the connection unexpectedly when an Exim client was
5468 using GnuTLS and trying to read a response, the client delivery process
5469 crashed while trying to generate an error log message.
5471 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5472 string, for example:
5474 warn message = H1: something\nH2: something
5476 the text was added as a single header line from Exim's point of view
5477 though it ended up OK in the delivered message. However, searching for the
5478 second and subsequent header lines using $h_h2: did not work. This has been
5479 fixed. Similarly, if a system filter added multiple headers in this way,
5480 the routers could not see them.
5482 20. Expanded the error message when iplsearch is called with an invalid key to
5483 suggest using net-iplsearch in a host list.
5485 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5486 ACLs are no longer actually imposed (and a message to that effect is
5489 22. If a "gecos" field in a passwd entry contained escaped characters, in
5490 particular, if it contained a \" sequence, Exim got it wrong when building
5491 a From: or a Sender: header from that name. A second bug also caused
5492 incorrect handling when an unquoted " was present following a character
5493 that needed quoting.
5495 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5496 was not being matched caselessly.
5498 24. Arranged for all hyphens in the exim.8 source to be escaped with
5501 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5502 properly. Recipient callout cache records were still being keyed to include
5503 the sender, even when use_sender was set false. This led to far more
5504 callouts that were necessary. The sender is no longer included in the key
5505 when use_sender is false.
5507 26. Added "control = submission" modifier to ACLs.
5509 27. Added the ${base62d: operator to decode base 62 numbers.
5511 28. dnsdb lookups can now access SRV records.
5513 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5514 the configuration file.
5516 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5517 (-v) mode. This makes the output for a verbose queue run more intelligible.
5519 31. Added a use_postmaster feature to recipient callouts.
5521 32. Added the $body_zerocount variable, containing the number of binary zero
5522 bytes in the message body.
5524 33. The time of last modification of the "new" subdirectory is now used as the
5525 "mailbox time last read" when there is a quota error for a maildir
5528 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5530 35. Added +ignore_unknown as a special item in host lists.
5532 36. Code for decoding IPv6 addresses in host lists is now included, even if
5533 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5534 address was recognized as an IP address, but was then not correctly decoded
5535 into binary, causing unexpected and incorrect effects when compared with
5542 1. Very minor rewording of debugging text in manualroute to say "list of
5543 hosts" instead of "hostlist".
5545 2. If verify=header_syntax was set, and a header line with an unqualified
5546 address (no domain) and a large number of spaces between the end of the
5547 name and the colon was received, the reception process suffered a buffer
5548 overflow, and (when I tested it) crashed. This was caused by some obsolete
5549 code that should have been removed. The fix is to remove it!
5551 3. When running in the test harness, delay a bit after writing a bounce
5552 message to get a bit more predictability in the log output.
5554 4. Added a call to search_tidyup() just before forking a reception process. In
5555 theory, someone could use a lookup in the expansion of smtp_accept_max_
5556 per_host which, without the tidyup, could leave open a database connection.
5558 5. Added the variables $recipient_data and $sender_data which get set from a
5559 lookup success in an ACL "recipients" or "senders" condition, or a router
5560 "senders" option, similar to $domain_data and $local_part_data.
5562 6. Moved the writing of debug_print from before to after the "senders" test
5565 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5566 problems for message scanning, either using a data ACL, or using
5567 local_scan() because the Received: header was not generated till after they
5568 were called (in order to set the time as the time of reception completion).
5569 I have revised the way this works. The header is now generated after the
5570 body is received, but before the ACL or local_scan() are called. After they
5571 are run, the timestamp in the header is updated.
5577 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5578 before starting a queue runner without re-exec. This happened only when
5579 deliver_drop_privilege was set or when the Exim user was set to root. The
5580 effect of the bug was that timeouts during subsequent deliveries caused
5581 crashes instead of being properly handled. The handler is now left at its
5582 default (and expected) setting.
5584 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5585 message, again when deliver_drop_privilege is set or Exim is run as root.
5586 The bug described in (1) was not present in this case, but the tidying up
5587 of the other signals was missing. I have made the two cases consistent.
5589 3. The ignore_target_hosts setting on a manualroute router was being ignored
5590 for hosts that were looked up using the /MX notation.
5592 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5595 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5596 operated on the sender address. After changing the $sender_address to <>
5597 for the sender address verify, Exim was re-instated it as the original
5598 (before rewriting) address, but remembering that it had rewritten it, so it
5599 wasn't rewriting it again. This bug also had the effect of breaking the
5600 sender address verification caching when the sender address was rewritten.
5602 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5603 This has been changed so that if the ip literal address matches
5604 ignore_target_hosts, the router declines.
5606 7. Added expansion conditions match_domain, match_address, and match_local_
5607 part (NOT match_host).
5609 8. The placeholder for the Received: header didn't have a length field set.
5611 9. Added code to Exim itself and to exim_lock to test for a specific race
5612 condition that could lead to file corruption when using MBX delivery. The
5613 issue is with the lockfile that is created in /tmp. If this file is removed
5614 after a process has opened it but before that process has acquired a lock,
5615 there is the potential for a second process to recreate the file and also
5616 acquire a lock. This could lead to two Exim processes writing to the file
5617 at the same time. The added code performs the same test as UW imapd; it
5618 checks after acquiring the lock that its file descriptor still refers to
5619 the same named file.
5621 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5622 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5625 11. Added the smtp_active_hostname option. If used, this will typically be made
5626 to depend on the incoming interface address. Because $interface_address is
5627 not set up until the daemon has forked a reception process, error responses
5628 that can happen earlier (such as "too many connections") no longer contain
5631 12. If an expansion in a condition on a "warn" statement fails because a lookup
5632 defers, the "warn" statement is abandoned, and the next ACL statement is
5633 processed. Previously this caused the whole ACL to be aborted.
5635 13. Added the iplsearch lookup type.
5637 14. Added ident_timeout as a log selector.
5639 15. Added tls_certificate_verified as a log selector.
5641 16. Added a global option tls_require_ciphers (compare the smtp transport
5642 option of the same name). This controls incoming TLS connections.
5644 17. I finally figured out how to make tls_require_ciphers do a similar thing
5645 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5646 before starting the TLS session.
5648 18. Tabs are now shown as \t in -bP output.
5650 19. If the log selector return_path_on_delivery was set, Exim crashed when
5651 bouncing a message because it had too many Received: header lines.
5653 20. If two routers both had headers_remove settings, and the first one included
5654 a superfluous trailing colon, the final name in the first list and the
5655 first name in the second list were incorrectly joined into one item (with a
5656 colon in the middle).
5662 1. Added -C and -D options to the exinext utility, mainly to make it easier
5663 to include in the automated testing, but these could be helpful when
5664 multiple configurations are in use.
5666 2. The exinext utility was not formatting the output nicely when there was
5667 an alternate port involved in the retry record key, nor when there was a
5668 message id as well (for retries that were specific to a specific message
5669 and a specific host). It was also confused by IPv6 addresses, because of
5670 the additional colons they contain. I have fixed the IPv4 problem, and
5671 patched it up to do a reasonable job for IPv6.
5673 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5674 delivery, the log line now contains "pipelined" if PIPELINING was used.
5676 4. An SMTP transport process used to panic and die if the bind() call to set
5677 an explicit outgoing interface failed. This has been changed; it is now
5678 treated in the same way as a connect() failure.
5680 5. A reference to $sender_host_name in the part of a conditional expansion
5681 that was being skipped was still causing a DNS lookup. This no longer
5684 6. The def: expansion condition was not recognizing references to header lines
5685 that used bh_ and bheader_.
5687 7. Added the _cache feature to named lists.
5689 8. The code for checking quota_filecount in the appendfile transport was
5690 allowing one more file than it should have been.
5692 9. For compatibility with Sendmail, the command line option
5700 and sets the incoming protocol and host name (for trusted callers). The
5701 host name and its colon can be omitted when only the protocol is to be set.
5702 Note the Exim already has two private options, -pd and -ps, that refer to
5703 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5704 "s", but I don't think that's a major issue.
5706 10. A number of refactoring changes to the code, none of which should affect
5709 (a) The number of logging options was getting close to filling up the
5710 32-bit word that was used as a bit map. I have split them into two classes:
5711 those that are passed in the argument to log_write(), and those that are
5712 only ever tested independently outside of that function. These are now in
5713 separate 32-bit words, so there is plenty of room for expansion again.
5714 There is no change in the user interface or the logging behaviour.
5716 (b) When building, for example, log lines, the code previously used a
5717 macro that called string_cat() twice, in order to add two strings. This is
5718 not really sufficiently general. Furthermore, there was one instance where
5719 it was actually wrong because one of the argument was used twice, and in
5720 one call a function was used. (As it happened, calling the function twice
5721 did not affect the overall behaviour.) The macro has been replaced by a
5722 function that can join an arbitrary number of extra strings onto a growing
5725 (c) The code for expansion conditions now uses a table and a binary chop
5726 instead of a serial search (which was left over from when there were very
5727 few conditions). Also, it now recognizes conditions like "pam" even when
5728 the relevant support is not compiled in: a suitably worded error message is
5729 given if an attempt is made to use such a condition.
5731 11. Added ${time_interval:xxxxx}.
5733 12. A bug was causing one of the ddress fields not to be passed back correctly
5734 from remote delivery subprocesses. The field in question was not being
5735 subsequently used, so this caused to problems in practice.
5737 13. Added new log selectors queue_time and deliver_time.
5739 14. Might have fixed a bug in maildirsizefile handling that threw up
5740 "unexpected character" debug warnings, and recalculated the data
5741 unnecessarily. In any case, I expanded the warning message to give more
5744 15. Added the message "Restricted characters in address" to the statements in
5745 the default ACL that block characters like @ and % in local parts.
5747 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5748 Three changes have been made:
5750 (a) There was a serious bug; a negative response to MAIL caused the whole
5751 recipient domain to be cached as invalid, thereby blocking all messages
5752 to all local parts at the same domain, from all senders. This bug has
5753 been fixed. The domain is no longer cached after a negative response to
5754 MAIL if the sender used is not empty.
5756 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5759 (c) A new callout option, "use_sender" has been added for people who want
5760 the modified behaviour.
5766 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5769 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5770 indeed breaks things for older releases.
5772 3. Added additional logging to the case where there is a problem reading data
5773 from a filter that is running in a subprocess using a pipe, in order to
5774 try to track down a specific problem.
5776 4. Testing facility fudge: when running in the test harness and attempting
5777 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5778 getting "No route to host". Convert this to a timeout.
5780 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5783 6. Some OS don't have socklen_t but use size_t instead. This affects the
5784 fifth argument of getsockopt() amongst other things. This is now
5785 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5786 can be set for individual OS. I have set it for SunOS5, OSF1, and
5787 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5788 some earlier ones do not.
5790 7. Change 4.30/15 was not doing the test caselessly.
5792 8. The standard form for an IPv6 address literal was being rejected by address
5793 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5794 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5795 this, as well as the form without the "IPv6" on the front (but only when
5796 address literals are enabled, of course).
5798 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5800 10. Exim crashed if a message with an empty sender address specified by -f
5801 encountered a router with an errors_to setting. This could be provoked only
5802 by a command such as
5806 where an empty string was supplied; "<>" did not hit this bug.
5808 11. Installed PCRE release 4.5.
5810 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5811 remained set. It is now erased.
5813 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5814 times from message ids (which are base 36 rather than the normal 62).
5816 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5817 were being counted as actual protocol errors, and logged if the log
5818 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5819 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5820 and DATA following a set of rejected RCPTs do not count as protocol errors.
5821 In other words, Exim assumes they were pipelined, though this may not
5822 actually be the case. Of course, in all cases the client gets an
5823 appropriate error code.
5825 15. If a lookup fails in an ACL condition, a message about the failure may
5826 be available; it is used if testing the ACL cannot continue, because most
5827 such messages specify what the cause of the deferral is. However, some
5828 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5829 that caused an old message to be retained and used if a later statement
5830 caused a defer, replacing the real cause of the deferral.
5832 16. If an IP address had so many PTR records that the DNS lookup buffer
5833 was not large enough to hold them, Exim could crash while trying to process
5834 the truncated data. It now detects and logs this case.
5836 17. Further to 4.21/58, another change has been made: if (and only if) the
5837 first line of a message (the first header line) ends with CRLF, a bare LF
5838 in a subsequent header line has a space inserted after it, so as not to
5839 terminate the header.
5841 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5842 unnecessarily, used atoi() instead of strtol(), and didn't check the
5843 termination when getting file sizes from file names by regex.
5845 19. Completely re-implemented the support for maildirsize files, in the light
5846 of a number of problems with the previous contributed implementation
5847 (4.30/29). In particular:
5849 . If the quota is zero, the maildirsize file is maintained, but no quota is
5852 . If the maildir directory does not exist, it is created before any attempt
5853 to write a maildirsize file.
5855 . The quota value in the file is just a cache; if the quota is changed in
5856 the transport, the new value overrides.
5858 . A regular expression is available for excluding directories from the
5861 20. The autoreply transport checks the characters in options that define the
5862 message's headers; it allows continued headers, but it was checking with
5863 isspace() after an embedded newline instead of explicitly looking for a
5866 21. If all the "regular" hosts to which an address was routed had passed their
5867 expiry times, and had not reached their retry times, the address was
5868 bounced, even if fallback hosts were defined. Now Exim should go on to try
5871 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5872 equivalent code in the SMTP transport. Some hosts send humungous responses
5873 to HELO/EHLO, more than 1024 it seems.
5875 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5876 gives compiler warnings in some environments. I've now done it "properly",
5879 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5880 (because of problems with the built-in one) was declared to return uschar *
5881 instead of char *, causing compiler failure.
5883 25. Fixed a file descriptor leak when processing alias/forward files.
5885 26. Fixed a minor format string issue in dbfn.c.
5887 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5889 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5890 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5893 29. When a sender address is verified, it is cached, to save repeating the test
5894 when there is more than one recipient in a message. However, when the
5895 verification involves a callout, it is possible for different callout
5896 options to be set for different recipients. It is too complicated to keep
5897 track of this in the cache, so now Exim always runs a verification when a
5898 callout is required, relying on the callout cache for the optimization.
5899 The overhead is duplication of the address routing, but this should not be
5902 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5903 to be verified with callout=postmaster, and the main callout worked but the
5904 postmaster check failed, the verification correctly failed. However, if a
5905 subsequent RCPT command asked for sender verification *without* the
5906 postmaster check, incorrect caching caused this verification also to fail,
5909 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5910 it was not caching the DNS options (qualify_single, search_parents) that
5911 were used when the lookup failed. A subsequent lookup with different
5912 options therefore always gave the same answer, though there were cases
5913 where it should not have. (Example: a "domains = !$mx_any" option on a
5914 dnslookup router: the "domains" option is always processed without any
5915 widening, but the router might have qualify_single set.) Now Exim uses the
5916 cached value only when the same options are set.
5918 32. Added John Jetmore's "exipick" utility to the distribution.
5920 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5921 than a timeout (e.g. a certificate is required, and is not provided), an
5922 Exim server now closes the connection immediately. Previously it waited for
5923 the client to close - but if the client is SSL, it seems that they each
5924 wait for each other, leading to a delay before one of them times out.
5926 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5927 maintained 0.8.x compatibility because I don't think many are using it, and
5928 it is clearly obsolete.
5930 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5933 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5934 client certificate was expired. A simple patch fixes this, though I don't
5935 understand the full logic of why the verify callback is called multiple
5938 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5939 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5940 which causes problems with some clients (such as the Certicom SSL Plus
5941 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5942 disables the coutermeasure allowing Eudora to connect."
5944 38. Exim was not checking that a write() to a log file succeeded. This could
5945 lead to Bad Things if a log got too big, in particular if it hit a file
5946 size limit. Exim now panics and dies if it cannot write to a log file, just
5947 as it does if it cannot open a log file.
5949 39. Modified OS/Makefile-Linux so that it now contains
5951 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5953 The two -D definitions ensure that Exim is compiled with large file
5954 support, which makes it possible to handle log files that are bigger than
5957 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5958 instance) a domain was checked against a named list that involved a lookup,
5959 causing $domain_data to be set, then another domain was checked against the
5960 same list, then the first domain was re-checked, the value of $domain_data
5961 after the final check could be wrong. In particular, if the second check
5962 failed, it could be set empty. This bug probably also applied to
5965 41. The strip_trailing_dot option was not being applied to the address given
5966 with the -f command-line option.
5968 42. The code for reading a message's header from the spool was incrementing
5969 $received_count, but never initializing it. This meant that the value was
5970 incorrect (doubled) while delivering a message in the same process in which
5971 it was received. In the most common configuration of Exim, this never
5972 happens - a fresh exec is done - but it can happen when
5973 deliver_drop_privilege is set.
5975 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5976 it now includes up to 150 characters of the unexpected data in the log
5979 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5980 and building data strings. The size of both of these buffers was 10 000
5981 bytes - far larger than anybody would *ever* want, thought I. Needless to
5982 say, somebody hit the limit. I have increased the maximum line length to
5983 20 000 and the maximum data length of concatenated lines to 100 000. I have
5984 also fixed two bugs, because there was no checking on these buffers. Tsk,
5985 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5986 buffer is too small.
5988 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5989 lsearch lookups. Now it does.
5991 46. When parsing a route_list item in a manualroute router, a fixed-length
5992 buffer was used for the list of hosts. I made this 1024 bytes long,
5993 thinking that nobody would ever have a list of hosts that long. Wrong.
5994 Somebody had a whole pile of complicated expansion conditions, and the
5995 string was silently truncated, leading to an expansion error. It turns out
5996 that it is easier to change to an unlimited length (owing to other changes
5997 that have happened since this code was originally written) than to build
5998 structure for giving a limitation error. The length of the item that
5999 expands into the list of hosts is now unlimited.
6001 47. The lsearch lookup could not handle data where the length of text line was
6002 more than 4095 characters. Such lines were truncated, leading to shortened
6003 data being returned. It should now handle lines of any length.
6005 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6006 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6009 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6010 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6011 no longer happens in comment lines. A list of replacements is now placed
6012 at the head of all of the source files, except those whose only change is
6013 to replace PERL_COMMAND in the very first #! line.
6015 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6016 messages on the queue, with a bottom-up merge sort, using code contributed
6017 by Michael Haardt. This should make operations like -bp somewhat faster on
6018 large queues. It won't affect queue runners, except when queue_run_in_order
6021 51. Installed eximstats 1.31 in the distribution.
6023 52. Added support for SRV lookups to the dnslookup router.
6025 53. If an ACL referred to $message_body or $message_body_end, the value was not
6026 reset for any messages that followed in the same SMTP session.
6028 54. The store-handling optimization for building very long strings was not
6029 differentiating between the different store pools. I don't think this
6030 actually made any difference in practice, but I've tidied it.
6032 55. While running the routers to verify a sender address, $sender_address
6033 was still set to the sender address. This is wrong, because when routing to
6034 send a bounce to the sender, it would be empty. Therefore, I have changed
6035 it so that, while verifying a sender address, $sender_address is set to <>.
6036 (There is no change to what happens when verifying a recipient address.)
6038 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6039 target A or AAAA records (if not already returned) without resetting the
6040 qualify_single or search_parents options of the DNS resolver. These are
6041 inappropriate in this case because the targets of MX and SRV records must
6042 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6043 target that, when qualified, matched something in the local domain. These
6044 two options are now turned off when doing these lookups.
6046 57. It seems that at least some releases of Reiserfs (which does not have the
6047 concept of a fixed number of inodes) returns zero and not -1 for the
6048 number of available inodes. This interacted badly with check_spool_inodes,
6049 which assumed that -1 was the "no such thing" setting. What I have done is
6050 to check that the total number of inodes is greater than zero before doing
6051 the test of how many are available.
6053 58. When a "warn" ACL statement has a log_message modifier, the message is
6054 remembered, and not repeated. This is to avoid a lot of repetition when a
6055 message has many recipients that cause the same warning to be written.
6056 However, Exim was preserving the list of already written lines for an
6057 entire SMTP session, which doesn't seem right. The memory is now reset if a
6058 new message is started.
6060 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6061 debugging output unless log_rewrite was also set.
6063 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6064 of (int)(handle) when we know that handle contains (void *)(-1).
6066 61. The Exim daemon panic-logs an error return when it closes the incoming
6067 connection. However "connection reset by peer" seems to be common, and
6068 isn't really an error worthy of noting specially, so that particular error
6071 62. When Exim is trying to find all the local interfaces, it used to panic and
6072 die if the ioctl to get the interface flags failed. However, it seems that
6073 on at least one OS (Solaris 9) it is possible to have an interface that is
6074 included in the list of interfaces, but for which you get a failure error
6075 for this call. This happens when the interface is not "plumbed" into a
6076 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6077 failure of the "get flags" call assumes that the interface is down.
6079 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6080 makes life easier for people who are doing arithmetic on fields extracted
6081 from dates, where you often get leading zeros that should not be
6082 interpreted as octal.
6084 64. Added qualify_domain to the redirect router, to override the global
6087 65. If a pathologically long header line contained very many addresses (the
6088 report of this problem mentioned 10 000) and each of them was rewritten,
6089 Exim could use up a very large amount of memory. (It kept on making new
6090 copies of the header line as it rewrote, and never released the old ones.)
6091 At the expense of a bit more processing, the header rewriting function has
6092 been changed so that it no longer eats memory in this way.
6094 66. The generation of the Received: header has been moved from the time that a
6095 message starts to be received, to the time that it finishes. The timestamp
6096 in the Received: header should now be very close to that of the <= log
6097 line. There are two side-effects of this change:
6099 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6100 logged header lines no longer include the local Received: line, because
6101 it has not yet been created. The same applies to a copy of the message
6102 that is returned to a non-SMTP sender when a message is rejected.
6104 (b) When a filter file is tested using -bf, no additional Received: header
6105 is added to the test message. After some thought, I decided that this
6108 This change does not affect the value of $received_for. It is still set
6109 after address rewriting, but before local_scan() is called.
6111 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6113 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6114 gave an unhelpful panic error message, and a defer error. I have managed to
6115 change this behaviour so that it now rejects any supplied certificate,
6116 which seems right, as the list of acceptable certificates is empty.
6118 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6119 gave an unhelpful defer error. I have not managed to make this reject any
6120 supplied certificates, but the error message it gives is "no certificate
6121 supplied", which is not helpful.
6123 70. exigrep's output now also includes lines that are not associated with any
6124 message, but which match the given pattern. Implemented by a patch from
6125 Martin Sluka, which also tidied up the Perl a bit.
6127 71. Recipient callout verification, like sender verification, was using <> in
6128 the MAIL FROM command. This isn't really the right thing, since the actual
6129 sender may affect whether the remote host accepts the recipient or not. I
6130 have changed it to use the actual sender in the callout; this means that
6131 the cache record is now keyed on a recipient/sender pair, not just the
6132 recipient address. There doesn't seem to be a real danger of callout loops,
6133 since a callout by the remote host to check the sender would use <>.
6134 [SEE ABOVE: changed after hitting problems.]
6136 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6137 temporary errors. However, in the case of such a code being given after
6138 the end of a data transmission (i.e. after ".") Exim was failing to write
6139 a retry record for the message. (Yes, there was some broken host that was
6140 actually sending 8xx at this point.)
6142 73. An unknown lookup type in a host list could cause Exim to panic-die when
6143 the list was checked. (An example that provoked this was putting <; in the
6144 middle of a list instead of at the start.) If this happened during a DATA
6145 ACL check, a -D file could be left lying around. This kind of configuration
6146 error no longer causes Exim to die; instead it causes a defer error. The
6147 incident is still logged to the main and panic logs.
6149 74. Buglet left over from Exim 3 conversion. The message "too many messages
6150 in one connection" was written to the rejectlog but not the mainlog, except
6151 when address rewriting (yes!) was being logged.
6153 75. Added write_rejectlog option.
6155 76. When a system filter was run not as root (that is, when system_filter_user
6156 was set), the values of the $n variables were not being returned to the
6157 main process; thus, they were not subsequently available in the $sn
6160 77. Added +return_path_on_delivery log selector.
6162 78. A connection timeout was being treated differently from recipients deferred
6163 when testing hosts_max_try with a message that was older than the host's
6164 retry timeout. (The host should not be counted, thus allowing all hosts to
6165 be tried at least once before bouncing.) This may have been the cause of an
6166 occasionally reported bug whereby a message would remain on the queue
6167 longer than the retry timeout, but would be bounced if a delivery was
6168 forced. I say "may" because I never totally pinned down the problem;
6169 setting up timeout/retry tests is difficult. See also the next item.
6171 79. The ultimate address timeout was not being applied to errors that involved
6172 a combination of host plus message (for example, a timeout on a MAIL
6173 command). When an address resolved to a number of possible hosts, and they
6174 were not all tried for each delivery (e.g. because of hosts_max_try), a
6175 message could remain on the queue longer than the retry timeout.
6177 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6180 81. Fixed an obscure SMTP outgoing bug which required at least the following
6181 conditions: (a) there was another message waiting for the same server;
6182 (b) the server returned 5xx to all RCPT commands in the first message so
6183 that the message was not completed; (c) the server dropped the connection
6184 or gave a negative response to the RSET that Exim sends to abort the
6185 transaction. The observed case was a dropped connection after DATA that had
6186 been sent in pipelining mode. That is, the server had advertised PIPELINING
6187 but was not implementing it correctly. The effect of the bug was incorrect
6188 behaviour, such as trying another host, and this could lead to a crash.
6194 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6195 and daemon.c were passed as pointers to ints; they should have been
6196 pointers to socklen_t variables (which are typically unsigned ints).
6198 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6201 3. Fixed a really odd bug that affected only the testing scheme; patching a
6202 certain fixed string in the binary changed the value of another string that
6203 happened to be identical to the end of the original first string.
6205 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6206 name", it returns that address as the IP address. On some operating
6207 systems (e.g. Solaris), it also passes back the IP address string as the
6208 "host name". However, on others (e.g. Linux), it passes back an empty
6209 string. Exim wasn't checking for this, and was changing the host name to an
6210 empty string, assuming it had been canonicalized.
6212 5. Although rare, it is permitted to have more than one PTR record for a given
6213 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6214 all the names associated with an address, because they do in Solaris.
6215 However, it seems that they do not in Linux for data that comes from the
6216 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6217 I found this out when I moved to a new Linux workstation and tried to run
6218 the Exim test suite.
6220 To get round this problem I have changed the code so that it now does its
6221 own call to the DNS to look up PTR records when searching for a host name.
6222 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6223 addresses that are only in /etc/hosts are still found.
6225 This behaviour is, however, controlled by an option called host_lookup_
6226 order, which defaults to "bydns:byaddr". If people want to use the other
6227 order, or indeed, just use one or the other means of lookup, they can
6228 specify it in this variable.
6230 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6231 some operating systems, this comes back from gethostbyaddr() as an empty
6232 string, and this is what Exim used to test for. However, it seems that in
6233 other systems, "." is yielded. Exim now tests for this case too.
6235 7. The values of check_spool_space and check_log_space are now held internally
6236 as a number of kilobytes instead of an absolute number of bytes. If a
6237 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6238 kilobyte. This means that much larger values can be stored.
6240 8. Exim monitor: an attempt to get the action menu when not actually pointing
6241 at a message produces an empty menu entitled "No message selected". This
6242 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6243 no entries in it ("Shell widget menu has zero width and/or height"). So I
6244 have added a single, blank menu entry in this case.
6246 9. Added ${quote_local_part.
6248 10. MIME decoding is now applied to the contents of Subject: header lines when
6251 11. Now that a reference to $sender_host_address automatically causes a reverse
6252 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6253 host lookup before query-style lookups in lists that might use this
6254 variable. This has therefore been abolished, and the "net-" prefix is no
6255 longer necessary for query-style lookups.
6257 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6258 have been a typo for LFLAGS, so it has been changed.
6260 13. The install script calls Exim with "-C /dev/null" in order to find the
6261 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6262 to be output. However, since Exim outputs its version number before the
6263 error, it didn't break the script. It just looked ugly. I fixed this by
6264 always allowing "-C /dev/null" if the caller is root.
6266 14. Ignore overlarge ACL variable number when reading spool file - insurance
6267 against a later release with more variables having written the file.
6269 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6270 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6271 this, as well as the form without the "IPv6" on the front.
6273 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6274 OS/Makefile-Darwin file.
6276 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6277 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6279 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6280 variable. However, some OS use ints for inodes. I've added cast to long int
6281 to get rid of the compiler warning.
6283 19. I had forgotten to lock out "/../" in configuration file names when
6284 ALT_CONFIG_PREFIX was set.
6286 20. Routers used for verification do not need to specify transports. However,
6287 if such a router generated a host list, and callout was configured, Exim
6288 crashed, because it could not find a port number from the (non-existent)
6289 transport. It now assumes port 25 in this circumstance.
6291 21. Added the -t option to exigrep.
6293 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6294 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6295 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6296 them is set, LOOKUP_LSEARCH is forced.
6298 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6300 24. Added sender and host information to the "rejected by local_scan()" log
6301 line; previously there was no indication of these.
6303 25. Added .include_if_exists.
6305 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6306 Linux. It turns out that not all file systems support this. Apparently some
6307 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6308 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6309 not supported on the file descriptor, is now ignored when Exim is trying to
6310 sync a directory. This applies only to Linux.
6312 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6314 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6315 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6318 29. Added support for maildirsize files from supplied patch (modified a bit).
6320 30. The use of :fail: followed by an empty string could lead Exim to respond to
6321 sender verification failures with (e.g.):
6323 550 Verification failed for <xxx>
6324 550 Sender verify failed
6326 where the first response line was missing the '-' that indicates it is not
6327 the final line of the response.
6329 31. The loop for finding the name of the user that called Exim had a hardwired
6330 limit of 10; it now uses the value of finduser_retries, which is used for
6331 all other user lookups.
6333 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6336 33. Exim was neglecting to zero errno before one call of strtol() when
6337 expanding a string and expecting an integer value. On some systems this
6338 resulted in spurious "integer overflow" errors. Also, it was casting the
6339 result into an int without checking.
6341 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6342 did not work. The code looks as if it has *never* worked, though it appears
6343 to have been documented since at least release 1.62. I have made it work.
6345 35. The "timeout_DNS" error in retry rules, also documented since at least
6346 1.62, also never worked. As it isn't clear exactly what this means, and
6347 clearly it isn't a major issue, I have abolished the feature by treating it
6348 as "timeout", and writing a warning to the main and panic logs.
6350 36. The display of retry rules for -brt wasn't always showing the error code
6353 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6354 timeout_connect_A, timeout_connect_MX.
6356 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6357 to the empty sender.
6359 39. The daemon was not analysing the content of -oX till after it had closed
6360 stderr and disconnected from the controlling terminal. This meant that any
6361 syntax errors were only noted on the panic log, and the return code from
6362 the command was 0. By re-arranging the code a little, I've made the
6363 decoding happen first, so such errors now appear on stderr, and the return
6364 code is 1. However, the actual setting up of the sockets still happens in
6365 the disconnected process, so errors there are still only recorded on the
6368 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6369 connections (as happens on some IP stacks) was logged at start up time as
6370 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6371 it from "IPv6 and IPv4", which means that two separate sockets are being
6374 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6375 says whether AF_INET or AF_INET6 was passed as an argument.
6377 42. Exiwhat output was messed up when time zones were included in log
6380 43. Exiwhat now gives more information about the daemon's listening ports,
6381 and whether -tls-on-connect was used.
6383 44. The "port" option of the smtp transport is now expanded.
6385 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6386 silently ignored. Now an error message is written to the main and panic
6389 46. There's a new ACL modifier called "logwrite" which writes to a log file
6390 as soon as it is encountered.
6392 47. Added $local_user_uid and $local_user_gid at routing time.
6394 48. Exim crashed when trying to verify a sender address that was being
6397 49. Exim was recognizing only a space character after ".include". It now also
6398 recognizes a tab character.
6400 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6401 extracting the relevant information from the specification. The man page no
6402 longer contains scrambled data for the -d option, and I've added a section
6403 at the front about calling Exim under different names.
6405 51. Added "extra_headers" argument to the "mail" command in filter files.
6407 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6410 53. Installed eximstats 1.29.
6412 54. Added transport_filter_timeout as a generic transport option.
6414 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6415 Cc: header lines. This was required by RFC 822, but it not required by RFC
6418 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6419 incoming messages that did not have them. Now it does so only if the
6420 message originates locally, that is, if there is no associated remote host
6421 address. When Resent- header lines are present, this applies to the Resent-
6422 lines rather than the non-Resent- lines.
6424 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6425 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6427 58. Messages for configuration errors now include the name of the main
6428 configuration file - useful now that there may be more than one file in a
6429 list (.included file names were always shown).
6431 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6432 for those rare installations that do not start the daemon as root or run it
6433 setuid root. I've cut out the call to initgroups() if the daemon is not
6436 60. The Exim user and group can now be bound into the binary as text strings
6437 that are looked up at the start of Exim's processing.
6439 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6441 62. Added $mailstore_basename variable.
6443 63. Installed patch to sieve.c from Michael Haardt.
6445 64. When Exim failed to open the panic log after failing to open the main log,
6446 the original message it was trying to log was written to stderr and debug
6447 output, but if they were not available (the usual case in production), it
6448 was lost. Now it is written to syslog before the two lines that record the
6449 failures to open the logs.
6451 65. Users' Exim filters run in subprocesses under the user's uid. It is
6452 possible for a "deliver" command or an alias in a "personal" command to
6453 provoke an address rewrite. If logging of address rewriting is configured,
6454 this fails because the process is not running as root or exim. There may be
6455 a better way of dealing with this, but for the moment (because 4.30 needs
6456 to be released), I have disabled address rewrite logging when running a
6457 filter in a non-root, non-exim process.
6463 1. The buildconfig auxiliary program wasn't quoting the value set for
6464 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6465 not defined. This bug was masked in 4.22 by the effect that was fixed in
6468 2. Some messages that were rejected after a message id was allocated were
6469 shown as "incomplete" by exigrep. It no longer does this for messages that
6470 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6472 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6473 have allow_domain_literals set, the ID did not get logged in the <= line.
6474 Domain literals are now always recognized in Message-ID: header lines.
6476 4. The first argument for a ${extract expansion item is the key name or field
6477 number. Leading and trailing spaces in this item were not being ignored,
6478 causing some misleading effects.
6480 5. When deliver_drop_privilege was set, single queue runner processes started
6481 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6482 same command in the process it spins off) were not dropping privilege.
6484 6. When the daemon running as "exim" started a queue runner, it always
6485 re-executed Exim in the spun-off process. This is a waste of effort when
6486 deliver_drop_privilege is set. The new process now just calls the
6487 queue-runner function directly.
6493 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6496 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6497 ignored. Though the use of -oP was forcing the writing of a pid file, it
6498 was always written to the default place.
6500 3. If the message "no IP address found for host xxxx" is generated during
6501 incoming verification, it is now followed by identification of the incoming
6502 connection (so you can more easily find what provoked it).
6504 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6506 5. Added some features to "harden" Exim a bit more against certain attacks:
6508 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6509 be put in Local/Makefile. This is like the never_users runtime option,
6510 but it cannot be overridden. The default setting is "root".
6512 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6513 prefix string with which any file named in a -C command line option
6516 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6517 is retained for -C and -D only if the caller of Exim is root. Without
6518 it, the exim user may also use -C and -D and retain privilege.
6520 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6521 command line option is disabled.
6523 6. Macro names set by the -D option must start with an upper case letter, just
6524 like macro names defined in the configuration file.
6526 7. Added "dereference=" facility to LDAP.
6528 8. Two instances of the typo "uknown" in the source files are fixed.
6530 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6531 the Configure-Makefile script screwed up while processing it.
6533 10. Incorporated PCRE 4.4.
6535 11. The SMTP synchronization check was not operating right at the start of an
6536 SMTP session. For example, it could not catch a HELO sent before the client
6537 waited for the greeting. There is now a check for outstanding input at the
6538 point when the greeting is written. Because of the duplex, asynchronous
6539 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6540 way, but not yet received, when the check is performed.
6542 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6543 on TCP/IP sockets, because this apparently causes some broken clients to
6546 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6547 unchanged) from the Cygwin maintainer.
6549 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6550 was supported instead of testing for mbx. Effectively a typo.
6552 15. The spa authenticator server code was not checking that the input it
6553 received was valid base64.
6555 16. The debug output line for the "set" modifier in ACLs was not showing the
6556 name of the variable that was being set.
6558 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6560 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6561 lookup on its own. Something else was needed to trigger the lookup. For
6562 example, a match in host_lookup or the need for a host name in a host list.
6563 Now, if $sender_host_name is referenced and the host name has not yet been
6564 looked up, a lookup is performed. If the lookup fails, the variable remains
6565 empty, and $host_lookup_failed is set to "1".
6567 19. Added "eqi" as a case-independent comparison operator.
6569 20. The saslauthd authentication condition could segfault if neither service
6570 nor realm was specified.
6572 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6573 error message that was logged was misleading, and incoming SMTP
6574 connections were dropped. The message is now more accurate, and temporary
6575 errors are given to SMTP connections.
6577 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6578 (except RSET) with a 421 error, until QUIT is received. However, it was
6579 failing to send a response to QUIT.
6581 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6582 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6583 ACL is now run after the helo_try_verify_hosts code.
6585 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6586 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6587 case-independent, so other case variants are also recognized). Apparently
6588 some systems use these upper case variants.
6590 25. If more than two messages were waiting for the same host, and a transport
6591 filter was specified for the transport, Exim sent two messages over the
6592 same TCP/IP connection, and then failed with "socket operation on non-
6593 socket" when it tried to send the third.
6595 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6597 27. The extern definition of crypt16() in expand.c was not being excluded when
6598 the OS had its own crypt16() function.
6600 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6601 as a preferred synonym for return_size_limit, both as an option and as an
6604 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6606 30. Changed the default configuration ACL to relax the local part checking rule
6607 for addresses that are not in any local domains. For these addresses,
6608 slashes and pipe symbols are allowed within local parts, but the sequence
6609 /../ is explicitly forbidden.
6611 31. SPA server authentication was not clearing the challenge buffer before
6614 32. log_message in a "warn" ACL statement was writing to the reject log as
6615 well as to the main log, which contradicts the documentation and doesn't
6616 seem right (because no rejection is happening). So I have stopped it.
6618 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6619 However, I am unable to do any testing of this.
6621 34. Fixed an infelicity in the appendfile transport. When checking directories
6622 for a mailbox, to see if any needed to be created, it was accidentally
6623 using path names with one or more superfluous leading slashes; tracing
6624 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6626 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6627 contents are added to the log line that is written for every discarded
6628 recipient. (Previously a log_message setting was ignored.)
6630 36. The ${quote: operator now quotes the string if it is empty.
6632 37. The install script runs exim in order to find its version number. If for
6633 some reason other than non-existence or emptiness, which it checks, it
6634 could not run './exim', it was installing it with an empty version number,
6635 i.e. as "exim-". This error state is now caught, and the installation is
6638 38. An argument was missing from the function that creates an error message
6639 when Exim fails to connect to the socket for saslauthd authentication.
6640 This could cause Exim to crash, or give a corrupted message.
6642 39. Added isip, isip4, and isip6 to ${if conditions.
6644 40. The ACL variables $acl_xx are now saved with the message, and can be
6645 accessed later in routers, transports, and filters.
6647 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6648 strings in the file are not string-expanded.
6650 42. If a MAIL command specified a SIZE value that was too large to fit into an
6651 int variable, the check against message_size_limit failed. Such values are
6652 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6653 day this will have to be increased, but I don't think I want to be around
6654 when emails are that large.
6661 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6662 iconv() is not standard in FreeBSD.
6664 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6665 IPv6 enabled. The observed symptom was a segmentation fault on return from
6666 the function os_common_find_running_interfaces() in src/os.c.
6668 3. In the check_special_case() function in daemon.c I had used "errno" as an
6669 argument name, which causes warnings on some systems. This was basically a
6670 typo, since it was named "eno" in the comments!
6672 4. The code that waits for the clock to tick (at a resolution of some fraction
6673 of a second) so as to ensure message-id uniqueness was always waiting for
6674 at least one whole tick, when it could have waited for less. [This is
6675 almost certainly not relevant at current processor speeds, where it is
6676 unlikely to ever wait at all. But we try to future-proof.]
6678 5. The function that sleeps for a time interval that includes fractions of a
6679 second contained a race. It did not block SIGALRM between setting the
6680 timer, and suspending (a couple of lines later). If the interval was short
6681 and the sigsuspend() was delayed until after it had expired, the suspension
6682 never ended. On busy systems this could lead to processes getting stuck for
6685 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6686 process, before it forks any delivery processes. The open lookup caching
6687 mechanism meant that the open file or database connection was passed into
6688 the delivery process. The problem was that delivery processes always tidy
6689 up cached lookup data. This could cause a problem for the next delivery
6690 process started by the queue runner, because the external queue runner
6691 process does not know about the closure. So the next delivery process
6692 still has data in the lookup cache. In the case of a file lookup, there was
6693 no problem because closing a file descriptor in a subprocess doesn't affect
6694 the parent. However, if the lookup was caching a connection to a database,
6695 the connection was closed, and the second delivery process was likely to
6696 see errors such as "PGSQL: query failed: server closed the connection
6697 unexpectedly". The problem has been fixed by closing all cached lookups
6698 in a queue runner before running a delivery process.
6700 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6701 seem to have the "const" qualifier which it has on other OS. I've
6704 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6705 *of the same type* (client or server) with the same public name that an
6706 error should be diagnosed.
6708 9. When Exim looked up a host name for an IP address, but failed to find the
6709 original IP address when looking up the host name (a safety check), it
6710 output the message "<ip address> does not match any IP for NULL", which was
6711 confusing, to say the least. The bug was that the host name should have
6712 appeared instead of "NULL".
6714 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6715 user that is built into the binary, and the -C or -D options is used, root
6716 privilege is dropped before the configuration file is read. In addition,
6717 logging is switched to stderr instead of the normal log files. If the
6718 configuration then re-defines the Exim user, the unprivileged environment
6719 is probably not what is expected, so Exim logs a panic warning message (but
6722 However, if deliver_drop_privilege is set, the unprivileged state may well
6723 be exactly what is intended, so the warning has been cut out in that case,
6724 and Exim is allowed to try to write to its normal log files.
6730 1. smtp_return_error_details was not giving details for temporary sender
6731 or receiver verification errors.
6733 2. Diagnose a configuration error if two authenticators have the same public
6736 3. Exim used not to create the message log file for a message until the first
6737 delivery attempt. This could be confusing when incoming messages were held
6738 for policy or load reasons. The message log file is now created at the time
6739 the message is received, and an initial "Received" line is written to it.
6741 4. The automatically generated man page for command line options had a minor
6742 bug that caused no ill effects; however, a more serious problem was that
6743 the procedure for building the man page automatically didn't always
6744 operate. Consequently, release 4.20 contains an out-of-date version. This
6745 shouldn't happen again.
6747 5. When building Exim with embedded Perl support, the script that builds the
6748 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6749 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6751 6. The freeze_tell option was not being used for messages that were frozen on
6752 arrival, either by an ACL or by local_scan().
6754 7. Added the smtp_incomplete_transaction log selector.
6756 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6757 was accepting AUTH without a new EHLO.
6759 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6760 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6763 10. Logging of TCP/IP connections (when configured) now happens in the main
6764 daemon process instead of the child process, so that the TCP/IP connection
6765 count is more accurate (but it can never be perfect).
6767 11. The use of "drop" in a nested ACL was not being handled correctly in the
6768 outer ACL. Now, if condition failure induced by the nested "drop" causes
6769 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6770 or "require"), the connection is dropped.
6772 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6773 that yield "discard" can now be used with an "accept" or a "discard" verb,
6774 but an error is generated for any others (because I can't see a useful way
6775 to define what should happen).
6777 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6778 are now processed in the same way as lines in the Exim configuration file.
6779 In particular, continuation lines are supported.
6781 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6783 15. Added -ti meaning -t -i.
6785 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6786 domains, and warn by logging if others are found.
6788 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6789 returned from GIFCONF when Exim is trying to find the list of interfaces on
6790 a host. The code in os.c has been modified to copy each ifreq to an aligned
6791 structure in all cases.
6793 Also, in some cases, the returned ifreq's were being copied to a 'struct
6794 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6795 means the last couple of bytes of an IPv6 address could be chopped if the
6796 ifreq contained only a normal sockaddr (14 bytes storage).
6798 18. Named domain lists were not supported in the hosts_treat_as_local option.
6799 An entry such as +xxxx was not recognized, and was treated as a literal
6802 19. Ensure that header lines added by a DATA ACL are included in the reject log
6803 if the ACL subsequently rejects the message.
6805 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6806 MD5 (which is deprecated).
6808 21. When testing a filter file using -bf, Exim was writing a message when it
6809 took the sender from a "From " line in the message, but it was not doing so
6810 when it took $return_path from a Return-Path: header line. It now does.
6812 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6813 with a valid header line field name (a series of printing characters
6814 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6816 23. Changed "disc" in the source to "disk" to conform to the documentation and
6817 the book and for uniformity.
6819 24. Ignore Sendmail's -Ooption=value command line item.
6821 25. When execve() failed while trying to run a command in a pipe transport,
6822 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6823 could be confused with a return value of 69 from the command itself. This
6824 has been changed to 127, the value the shell returns if it is asked to run
6825 a non-existent command. The wording for the related log line suggests a
6826 non-existent command as the problem.
6828 26. If received_header_text expands to an empty string, do not add a Received:
6829 header line to the message. (Well, it adds a token one on the spool, but
6830 marks it "old" so that it doesn't get used or transmitted.)
6832 27. Installed eximstats 1.28 (addition of -nt option).
6834 28. There was no check for failure on the call to getsockname() in the daemon
6835 code. This can fail if there is a shortage of resources on the system, with
6836 ENOMEM, for example. A temporary error is now given on failure.
6838 29. Contrary to the C standard, it seems that in some environments, the
6839 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6840 program. Exim now does this explicitly; it affects the formatting of
6841 timestamps using strftime().
6843 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6844 complaints. I've now initialized all the variables, to avoid this.
6846 32. Header lines added by a system filter were not being "seen" during
6847 transport-time rewrites.
6849 33. The info_callback() function passed to OpenSSL is set up with type void
6850 (*)(SSL *, int, int), as described somewhere. However, when calling the
6851 function (actually a macro) that sets it up, the type void(*)() is
6852 expected. I've put in a cast to prevent warnings from picky compilers.
6854 34. If a DNS black list lookup found a CNAME record, but there were no A
6855 records associated with the domain it pointed at, Exim crashed.
6857 35. If a DNS black list lookup returned more than one A record, Exim ignored
6858 all but the first. It now scans all returned addresses if a particular IP
6859 value is being sought. In this situation, the contents of the
6860 $dnslist_value variable are a list of all the addresses, separated by a
6863 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6864 to accept a host name and all its aliases if the forward lookup for any of
6865 them yielded the IP address of the incoming connection. Now it accepts only
6866 those names whose forward lookup yields the correct IP address. Any other
6867 names are discarded. This closes a loophole whereby a rogue DNS
6868 administrator could create reverse DNS records to break through a
6869 wildcarded host restriction in an ACL.
6871 37. If a user filter or a system filter that ran in a subprocess used any of
6872 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6873 the wrong values were passed to the pipe command ($thisaddress had the
6874 value of $0, $0 had the value of $1, etc). This bug was introduced by
6875 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6877 38. Improved the line breaking for long SMTP error messages from ACLs.
6878 Previously, if there was no break point between 40 and 75 characters, Exim
6879 left the rest of the message alone. Two changes have been made: (a) I've
6880 reduced the minimum length to 35 characters; (b) if it can't find a break
6881 point between 35 and 75 characters, it looks ahead and uses the first one
6882 that it finds. This may give the occasional overlong line, but at least the
6883 remaining text gets split now.
6885 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6886 file descriptors might be low, and that setting 1000 would always raise it.
6887 It turns out that in some environments, the limit is already over 1000 and
6888 that lowering it causes trouble. So now Exim takes care not to decrease it.
6890 40. When delivering a message, the value of $return_path is set to $sender_
6891 address at the start of routing (routers may change the value). By an
6892 oversight, this default was not being set up when an address was tested by
6893 -bt or -bv, which affected the outcome if any router or filter referred to
6896 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6897 log only when "message" or "log_message" are set. However, if one of the
6898 conditions was an address verification, or a call to a nested ACL, the
6899 messages generated by the underlying test were being passed through. This
6900 no longer happens. The underlying message is available in $acl_verify_
6901 message for both "message" and "log_message" expansions, so it can be
6902 passed through if needed.
6904 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6905 new expansion $bh_ to give the encoded byte string without charset
6906 translation. Translation happens only if iconv() is available; HAVE_ICONV
6907 indicates this at build time. HEADERS_CHARSET gives the charset to
6908 translate to; headers_charset can change it in the configuration, and
6909 "headers charset" can change it in an individual filter file.
6911 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6912 that creates RFC 2047 encoded "words" labels them as that charset instead
6913 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6914 expansion operator; (ii) when Exim creates a From: line for a local
6915 message; (iii) when a header line is rewritten to include a "phrase" part.
6917 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6918 buggy, causing it to skip the first lines of messages whose message ID
6919 ended in 'D'. This would not have bitten before Exim release 4.14, because
6920 message IDs were unlikely to end in 'D' before then. The effect was to have
6921 incorrect size information for certain domains.
6923 45. #include "config.h" was missing at the start of the crypt16.c module. This
6924 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6927 46. If there was a timeout during a "random" callout check, Exim treated it as
6928 a failure of the random address, and carried on sending RSET and the real
6929 address. If the delay was just some slowness somewhere, the response to the
6930 original RCPT would be taken as a response to RSET and so on, causing
6931 mayhem of various kinds.
6933 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6934 when I implemented it. It didn't allow for the fact that some option values
6935 may legitimately be negative (e.g. size_addition), and it didn't even do
6936 the right test for positive values.
6938 48. Domain names in DNS records are case-independent. Exim always looks them up
6939 in lower case. Some resolvers return domain names in exactly the case they
6940 appear in the zone file, that is, they may contain uppercase letters. Not
6941 all resolvers do this - some return always lower case. Exim was treating a
6942 change of case by a resolver as a change of domain, similar to a widening
6943 of a domain abbreviation. This triggered its re-routing code and so it was
6944 trying to route what was effectively the same domain again. This normally
6945 caused routing to fail (because the router wouldn't handle the domain
6946 twice). Now Exim checks for this case specially, and just changes the
6947 casing of the domain that it ultimately uses when it transmits the message
6950 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6953 50. If a filter generated a file delivery with a non-absolute name (possible if
6954 no home directory exists for the router), the forbid_file option was not
6957 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6958 the existing equality matching.
6960 52. Exim was using ints instead of ino_t variables in some places where it was
6961 dealing with inode numbers.
6963 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6964 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6965 called TMPDIR, and if it finds it is different, it changes its value.
6967 54. The smtp_printf() function is now made available to local_scan() so
6968 additional output lines can be written before returning. There is also an
6969 smtp_fflush() function to enable the detection of a dropped connection.
6970 The variables smtp_input and smtp_batched_input are exported to
6973 55. Changed the default runtime configuration: the message "Unknown user"
6974 has been removed from the ACL, and instead placed on the localuser router,
6975 using the cannot_route_message feature. This means that any verification
6976 failures that generate their own messages won't get overridden. Similarly,
6977 the "Unrouteable address" message that was in the ACL for unverifiable
6978 relay addresses has also been removed.
6980 56. Added hosts_avoid_esmtp to the smtp transport.
6982 57. The exicyclog script was not checking for the esoteric option
6983 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6984 will work only if exicyclog is run under the appropriate euid.
6986 58. Following a discussion on the list, the rules by which Exim recognises line
6987 endings on incoming messages have been changed. The -dropcr and drop_cr
6988 options are now no-ops, retained only for backwards compatibility. The
6989 following line terminators are recognized: LF CRLF CR. However, special
6990 processing applies to CR:
6992 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6993 nor a local message in the state where . is a terminator.
6995 (ii) If a bare CR is encountered in a header line, an extra space is added
6996 after the line terminator so as not to end the header. The reasoning
6997 behind this is that bare CRs in header lines are most likely either
6998 to be mistakes, or people trying to play silly games.
7000 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7001 was being incorrectly given as 18 bytes larger than it should have been.
7002 This is a VOB (very old bug).
7004 60. This may never have affected anything current, but just in case it has:
7005 When the local host is found other than at the start of a list of hosts,
7006 the local host, those with the same MX, and any that follow, are discarded.
7007 When the list in question was part of a longer list of hosts, the following
7008 hosts (not currently being processed) were also being discarded. This no
7009 longer happens. I'm not sure if this situation could ever has previously
7012 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7015 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7016 header. This is something that is recommended in a new Internet Draft, and
7017 is something that is documented as being done by Sendmail. There are two
7018 possible values. For messages generated by the autoreply transport, Exim
7021 Auto-Submitted: auto-replied
7023 whereas for all other generated messages (e.g. bounces) it adds
7025 Auto-Submitted: auto-generated
7027 63. The "personal" condition in filters now includes a test for the
7028 Auto-Submitted: header. If it contains the string "auto-" the message it
7029 not considered personal.
7031 64. Added rcpt_include_affixes as a generic transport option.
7033 65. Added queue_only_override (default true).
7035 66. Added the syslog_duplication option.
7037 67. If what should have been the first header line of a message consisted of
7038 a space followed by a colon, Exim was mis-interpreting it as a header line.
7039 It isn't of course - it is syntactically invalid and should therefore be
7040 treated as the start of the message body. The misbehaviour could have
7041 caused a number of strange effects, including loss of data in subsequent
7042 header lines, and spool format errors.
7044 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7045 client host had authenticated. This control can now be exercised by an ACL
7046 for more flexibility.
7048 69. By default, callouts do not happen when testing with -bh. There is now a
7049 variant, -bhc, which does actually run the callout code, including
7050 consulting and updating the callout cache.
7052 70. Added support for saslauthd authentication, courtesy of Alexander
7055 71. If statvfs() failed on the spool or log directories while checking their
7056 size for availability, Exim confusingly gave the error "space shortage".
7057 Furthermore, in debugging mode it crashed with a floating point exception.
7058 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7059 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7060 serious problem, Exim now writes to the main and panic logs when this
7061 happens, with details of the failure. It then refuses to accept the
7062 incoming message, giving the message "spool directory problem" or "log
7063 directory problem" with a 421 code for SMTP messages.
7065 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7066 0, 1, and 2 exist, because some OS complain for execs without them (see
7067 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7068 descriptors. However, the code omitted to check that the open succeeded,
7069 causing mysterious errors if for some reason the permissions on /dev/null
7070 got screwed. Now Exim writes a message to the main and panic logs, and
7071 bombs out if it can't open /dev/null.
7073 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7074 interact so that it is all more flexible. It is supposed to remain
7075 backwards compatible. Also added extra_local_interfaces.
7077 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7078 to bomb out with an assertion failure - to the client this appears as a
7079 connection drop. This problem occurs in the part of the code that was taken
7080 from the Samba project. Fortunately, the assertion is in a very simple
7081 function, so I have fixed this by reproducing the function inline in the
7082 one place where it is called, and arranging for authentication to fail
7083 instead of killing the process with assert().
7085 75. The SPA client code was not working when the server requested OEM rather
7086 than Unicode encoding.
7088 76. Added code to make require_files with a specific uid setting more usable in
7089 the case where statting the file as root fails - usually a non-root-mounted
7090 NFS file system. When this happens and the failure is EACCES, Exim now
7091 forks a subprocess and does the per-uid checking as the relevant uid.
7093 77. Added process_log_path.
7095 78. If log_file_path was not explicitly set, a setting of check_log_space or
7096 check_log_inodes was ignored.
7098 79. If a space check for the spool or log partitions fails, the incident is now
7099 logged. Of course, in the latter case the data may get lost...
7101 80. Added the %p formatting code to string_format() so that it can be used to
7102 print addresses in debug_print(). Adjusted all the address printing in the
7103 debugging in store.c to use %p rather than %d.
7105 81. There was a concern that a line of code in smtp_in.c could overflow a
7106 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7107 initially expressed, the concern was not well-founded, because trailing
7108 spaces are removed early. However, if the trailing spaces were followed by
7109 a NULL, they did not get removed, so the overflow was possible. Two fixes
7112 (a) I re-wrote the offending code in a cleaner fashion.
7113 (b) If an incoming SMTP command contains a NULL character, it is rejected
7116 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7117 runs initgroups(), so that if the Exim user is in any additional groups,
7118 they will be used during message reception.
7124 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob