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 GF/02 Catch and remove uninitialized value warning in exiqsumm
19 Check for existence of @ARGV before looking at $ARGV[0]
21 JH/01 Replace the store_release() internal interface with store_newblock(),
22 which internalises the check required to safely use the old one, plus
23 the allocate and data copy operations duplicated in both (!) of the
26 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
27 modifier. This matches the restriction on the commandline.
29 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
30 Previously only the last row was returned.
32 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
33 we assumed that tags in the header were well-formed, and parsed the
34 element content after inspecting only the first char of the tag.
35 Assumptions at that stage could crash the receive process on malformed
38 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
39 While running the DKIM ACL we operate on the Permanent memory pool so that
40 variables created with "set" persist to the DATA ACL. Also (at any time)
41 DNS lookups that fail create cache records using the Permanent pool. But
42 expansions release any allocations made on the current pool - so a dnsdb
43 lookup expansion done in the DKIM ACL releases the memory used for the
44 DNS negative-cache, and bad things result. Solution is to switch to the
45 Main pool for expansions.
46 While we're in that code, add checks on the DNS cache during store_reset,
47 active in the testsuite.
48 Problem spotted, and debugging aided, by Wolfgang Breyha.
50 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
51 When none of the hosts presented to a transport match an already-open
52 connection, close it and proceed with the list. Previously we would
53 queue the message. Spotted by Lena with Yahoo, probably involving
56 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
57 Previously a spurious "250 OK id=" response was appended to the proper
60 JH/08 The "support for" informational output now, which built with Content
61 Scanning support, has a line for the malware scanner interfaces compiled
62 in. Interface can be individually included or not at build time.
64 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
65 by the template makefile "src/EDITME". The "STREAM" support for an older
66 ClamAV interface method is removed.
68 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
69 rows affected is given instead).
71 JH/11 The runtime Berkeley DB library version is now additionally output by
72 "exim -d -bV". Previously only the compile-time version was shown.
74 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
75 SMTP connection. Previously, when one had more receipients than the
76 first, an abortive onward connection was made. Move to full support for
77 multiple onward connections in sequence, handling cutthrough connection
78 for all multi-message initiating connections.
80 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
81 routers. Previously, a multi-recipient message would fail to match the
82 onward-connection opened for the first recipient, and cause its closure.
84 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
85 a timeout on read on a GnuTLS initiating connection, resulting in the
86 initiating connection being dropped. This mattered most when the callout
87 was marked defer_ok. Fix to keep the two timeout-detection methods
90 JH/15 Relax results from ACL control request to enable cutthrough, in
91 unsupported situations, from error to silently (except under debug)
92 ignoring. This covers use with PRDR, frozen messages, queue-only and
95 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
97 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
98 metadata, resulting in a crash in free().
100 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
101 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
102 Broken also in d185889f4, with init system revamp.
104 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
105 Previously we abruptly closed the connection after reading a malware-
106 found indication; now we go on to read the "scan ok" response line,
109 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
110 ACL. Previously, a crash would result.
112 JH/19 Speed up macro lookups during configuration file read, by skipping non-
113 macro text after a replacement (previously it was only once per line) and
114 by skipping builtin macros when searching for an uppercase lead character.
116 JH/20 DANE support moved from Experimental to mainline. The Makefile control
117 for the build is renamed.
119 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
120 was allocated for every new TLS startup, meaning one per message. Fix
121 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
123 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
124 reported the original. Fix to report (as far as possible) the ACL
125 result replacing the original.
127 JH/23 Fix memory leak during multi-message connections using STARTTLS under
128 OpenSSL. Certificate information is loaded for every new TLS startup,
129 and the resources needed to be freed.
131 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
133 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
136 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
137 DATA response info to the (existing) per-recipient response info for
138 the "C=" log element. It can have useful tracking info from the
139 destination system. Patch from Simon Arlott.
141 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
142 length value. Previously this would segfault.
144 HS/02 Support Avast multiline protoocol, this allows passing flags to
145 newer versions of the scanner.
147 JH/28 Ensure that variables possibly set during message acceptance are marked
148 dead before release of memory in the daemon loop. This stops complaints
149 about them when the debug_store option is enabled. Discovered specifically
150 for sender_rate_period, but applies to a whole set of variables.
151 Do the same for the queue-runner and queue-list loops, for variables set
152 from spool message files. Do the same for the SMTP per-message loop, for
153 certain variables indirectly set in ACL operations.
155 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
156 as a multi-recipient message from a mailinglist manager). The coding had
157 an arbitrary cutoff number of characters while checking for more input;
158 enforced by writing a NUL into the buffer. This corrupted long / fast
159 input. The problem was exposed more widely when more pipelineing of SMTP
160 responses was introduced, and one Exim system was feeding another.
161 The symptom is log complaints of SMTP syntax error (NUL chars) on the
162 receiving system, and refused recipients seen by the sending system
163 (propating to people being dropped from mailing lists).
164 Discovered and pinpointed by David Carter.
166 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
167 replaced by the ${authresults } expansion.
169 JH/31 Bug 2257: Fix pipe transport to not use a socket-only syscall.
171 HS/03 Set a handler for SIGTERM and call exit(3) if running as PID 1. This
172 allows proper process termination in container environments.
174 JH/32 Bug 2258: Fix spool_wireformat in combination with LMTP transport.
175 Previously the "final dot" had a newline after it; ensure it is CR,LF.
177 JH/33 SPF: remove support for the "spf" ACL condition outcome values "err_temp"
178 and "err_perm", deprecated since 4.83 when the RFC-defined words
179 "temperror" and "permerror" were introduced.
181 JH/34 Re-introduce enforcement of no cutthrough delivery on transports having
182 transport-filters or DKIM-signing. The restriction was lost in the
183 consolidation of verify-callout and delivery SMTP handling.
184 Extend the restriction to also cover ARC-signing.
186 JH/35 Cutthrough: for a final-dot response timeout (and nonunderstood responses)
187 in defer=pass mode supply a 450 to the initiator. Previously the message
190 PP/02 DANE: add dane_require_tls_ciphers SMTP Transport option; if unset,
191 tls_require_ciphers is used as before.
197 JH/01 Rework error string handling in TLS interface so that the caller in
198 more cases is responsible for logging. This permits library-sourced
199 string to be attached to addresses during delivery, and collapses
200 pairs of long lines into single ones.
202 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
203 during configuration. Wildcards are allowed and expanded.
205 JH/02 Rework error string handling in DKIM to pass more info back to callers.
206 This permits better logging.
208 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
209 do not close it down and have the child transport start it up again on
210 the passed-on TCP connection. Instead, proxy the child (and any
211 subsequent ones) for TLS via a unix-domain socket channel. Logging is
212 affected: the continued delivery log lines do not have any DNSSEC, TLS
213 Certificate or OCSP information. TLS cipher information is still logged.
215 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
216 identical IP addresses on different listening ports. Will also affect
219 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
220 add noisy ifdef guards to special-case this sillyness.
221 Patch from Bernd Kuhls.
223 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
224 than 255 are no longer allowed.
226 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
227 Disable the session-cache too, which might reduce our load. Since we
228 currrectly use a new context for every connection, both as server and
229 client, there is no benefit for these.
230 GnuTLS appears to not support tickets server-side by default (we don't
231 call gnutls_session_ticket_enable_server()) but client side is enabled
232 by default on recent versions (3.1.3 +) unless the PFS priority string
235 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
236 <https://reproducible-builds.org/specs/source-date-epoch/>.
238 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
239 the check for any unsuccessful recipients did not notice the limit, and
240 erroneously found still-pending ones.
242 JH/08 Pipeline CHUNKING command and data together, on kernels that support
243 MSG_MORE. Only in-clear (not on TLS connections).
245 JH/09 Avoid using a temporary file during transport using dkim. Unless a
246 transport-filter is involved we can buffer the headers in memory for
247 creating the signature, and read the spool data file once for the
248 signature and again for transmission.
250 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
251 4.77 as the kernel support then wasn't solid, having issues in 64bit
252 mode. Now, it's been long enough. Add support for FreeBSD also.
254 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
255 case where the routing stage had gathered several addresses to send to
256 a host before calling the transport for the first, we previously failed
257 to close down TLS in the old transport process before passing the TCP
258 connection to the new process. The new one sent a STARTTLS command
259 which naturally failed, giving a failed delivery and bloating the retry
260 database. Investigation and fix prototype from Wolfgang Breyha.
262 JH/12 Fix check on SMTP command input synchronisation. Previously there were
263 false-negatives in the check that the sender had not preempted a response
264 or prompt from Exim (running as a server), due to that code's lack of
265 awareness of the SMTP input buffering.
267 PP/04 Add commandline_checks_require_admin option.
268 Exim drops privileges sanely, various checks such as -be aren't a
269 security problem, as long as you trust local users with access to their
270 own account. When invoked by services which pass untrusted data to
271 Exim, this might be an issue. Set this option in main configuration
272 AND make fixes to the calling application, such as using `--` to stop
275 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
276 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
277 responses to those, into a single TLS record each way (this usually means
278 a single packet). As a side issue, smtp_enforce_sync now works on TLS
281 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
282 affects you only if you're dancing at the edge of the param size limits.
283 If you are, and this message makes sense to you, then: raise the
284 configured limit or use OpenSSL 1.1. Nothing we can do for older
287 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
288 cmdline element to get the documented default one. Previously it was
291 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
292 are unsuable for non-message-specific reasons - eg. connection timeout,
293 banner-time rejection.
295 JH/16 Fix logging of delivery remote port, when specified by router, under
298 PP/06 Repair manualroute's ability to take options in any order, even if one
299 is the name of a transport.
302 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
304 JH/17 Change the list-building routines interface to use the expanding-string
305 triplet model, for better allocation and copying behaviour.
307 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
308 Previously it was constructed the first time a possibly-matching string
309 was met in the configuration file input during startup; now it is done
312 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
313 compatible one, to avoid the (poorly documented) possibility of a config
314 file in the working directory redirecting the DB files, possibly correpting
315 some existing file. CVE-2017-10140 assigned for BDB.
317 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
318 cache-hot. Previously, although the result was properly cached, the
319 initial verify call returned a defer.
321 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
322 the main verify for receipient in uncached-mode.
324 JH/22 Retire historical build files to an "unsupported" subdir. These are
325 defined as "ones for which we have no current evidence of testing".
327 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
328 if present. Previously it was ignored.
330 JH/24 Start using specified-initialisers in C structure init coding. This is
331 a C99 feature (it's 2017, so now considered safe).
333 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
334 if was a fixed-sized field and bitmask ops via macros; it is now more
337 PP/07 GitHub PR 56: Apply MariaDB build fix.
338 Patch provided by Jaroslav Škarvada.
340 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
341 during Coverity cleanups [4.87 JH/47]
342 Diagnosis and fix provided by Michael Fischer v. Mollard.
344 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
345 the right size to place the terminating semicolon on its own folded
346 line, the header hash was calculated to an incorrect value thanks to
347 the (relaxed) space the fold became.
349 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
350 and confused the parent.
352 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
353 which could crash as a result. This could lead to undeliverable messages.
355 JH/28 Logging: "next input sent too soon" now shows where input was truncated
358 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
359 matters on fast-turnover and PID-randomising systems, which were getting
360 out-of-order delivery.
362 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
363 a possibly-overlapping copy. The symptom was that "Remote host closed
364 connection in response to HELO" was logged instead of the actual 4xx
367 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
368 Previously only that bufferd was discarded, resulting in SYMTP command
371 JH/32 DKIM: when a message has multiple signatures matching an identity given
372 in dkim_verify_signers, run the dkim acl once for each. Previously only
373 one run was done. Bug 2189.
375 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
376 "panic the current process" to "deliberately defer". The panic log is
377 still written with the problem list name; the mail and reject logs now
378 get a temp-reject line for the message that was being handled, saying
379 something like "domains check lookup or other defer". The SMTP 451
380 message is still "Temporary local problem".
382 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
383 A crafted sequence of BDAT commands could result in in-use memory beeing
384 freed. CVE-2017-16943.
386 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
387 from SMTP input. Previously it was always done; now only done for DATA
388 and not BDAT commands. CVE-2017-16944.
390 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
391 to the message (such as an overlong header line). Previously this was
392 not done and we did not exit BDAT mode. Followon from the previous item
393 though a different problem.
399 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
400 than -2003 did; needs libidn2 in addition to libidn.
402 JH/02 The path option on a pipe transport is now expanded before use.
404 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
405 Patch provided by "Björn", documentation fix added too.
407 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
408 missing a wire-to-host endian conversion.
410 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
411 close after a BDAT command line could be taken as a following command,
412 giving a synch failure. Fix by only checking for synch immediately
413 before acknowledging the chunk.
415 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
416 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
417 macro. Patches provided by Josh Soref.
419 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
420 Previously we did not; the RFC seems ambiguous and VRFY is not listed
421 by IANA as a service extension. However, John Klensin suggests that we
424 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
425 the dkim code may be unix-mode line endings rather than smtp wire-format
426 CRLF, so prepend a CR to any bare LF.
428 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
429 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
431 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
432 the first were themselves being wrongly included in the feed into dkim
433 processing; with most chunk sizes in use this resulted in an incorrect
434 body hash calculated value.
436 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
437 DKIM signature block, for verification. Although advised against by
438 standards it is specifically not ruled illegal.
440 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
442 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
443 missing a body hash (the bh= tag).
445 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
446 It seems that HAProxy sends the Proxy Protocol information in clear and
447 only then does a TLS startup, so do the same.
449 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
450 TCP connections (such as for Spamd) unless the daemon successfully set
451 Fast Open mode on its listening sockets. This fixes breakage seen on
452 too-old kernels or those not configured for Fast Open, at the cost of
453 requiring both directions being enabled for TFO, and TFO never being used
454 by non-daemon-related Exim processes.
456 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
457 endings, at least on the first header line. Try to canonify any that get
458 past that check, despite the cost.
460 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
461 now limited to an arbitrary five deep, while parsing addresses with the
462 strip_excess_angle_brackets option enabled.
464 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
465 instead leave the unprompted TLS handshake in socket buffer for the
466 TLS library to consume.
468 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
470 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
472 JH/16 Drop variables when they go out of scope. Memory management drops a whole
473 region in one operation, for speed, and this leaves assigned pointers
474 dangling. Add checks run only under the testsuite which checks all
475 variables at a store-reset and panics on a dangling pointer; add code
476 explicitly nulling out all the variables discovered. Fixes one known
477 bug: a transport crash, where a dangling pointer for $sending_ip_address
478 originally assigned in a verify callout, is re-used.
480 PP/06 Drop '.' from @INC in various Perl scripts.
482 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
484 PP/08 Reduce a number of compilation warnings under clang; building with
485 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
486 should be warning-free.
488 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
490 HS/01 Fix portability problems introduced by PP/08 for platforms where
491 realloc(NULL) is not equivalent to malloc() [SunOS et al].
493 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
494 chunk. This allows us to accept broken chunked messages. We need a more
495 general solution here.
497 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
498 already-broken messages in the queue.
500 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
502 JH/19 Fix reference counting bug in routing-generated-address tracking.
508 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
509 supports it and a size is available (ie. the sending peer gave us one).
511 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
512 years of being deprecated). The replacements are the ACLs
513 acl_smtp_mime and acl_not_smtp_mime.
515 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
516 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
517 or even an in-clear connection were permitted. Now, if the host lookup
518 was dnssec and dane was requested then the host is only used if the
519 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
520 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
521 if one fails this test.
522 This means that a poorly-configured remote DNS will make it incommunicado;
523 but it protects against a DNS-interception attack on it.
525 JH/04 Bug 1810: make continued-use of an open smtp transport connection
526 non-noisy when a race steals the message being considered.
528 JH/05 If main configuration option tls_certificate is unset, generate a
529 self-signed certificate for inbound TLS connections.
531 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
532 in rewrites and routers.
534 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
535 and logged a warning sing 4.83; now they are a configuration file error.
537 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
538 (lacking @domain). Apply the same qualification processing as RCPT.
540 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
542 JH/10 Support ${sha256:} applied to a string (as well as the previous
545 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
546 a cutthrough deliver is pending, as we always want to make a connection.
547 This also avoids re-routing the message when later placing the cutthrough
548 connection after a verify cache hit.
549 Do not update it with the verify result either.
551 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
552 when routing results in more than one destination address.
554 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
555 signing (which inhibits the cutthrough capability). Previously only
556 the presence of an option was tested; now an expansion evaluating as
557 empty is permissible (obviously it should depend only on data available
558 when the cutthrough connection is made).
560 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
561 the relevant preceding SMTP command did not note the pipelining mode.
563 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
564 Previously they were not counted.
566 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
567 as one having no matching records. Previously we deferred the message
568 that needed the lookup.
570 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
571 distinguished as "(=".
573 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
574 for missing MX records. Previously it only worked for missing A records.
576 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
578 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
579 after the data-go-ahead and data-ack. Patch from Jason Betts.
581 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
582 even for a "none" policy. Patch from Tony Meyer.
584 JH/22 Fix continued use of a connection for further deliveries. If a port was
585 specified by a router, it must also match for the delivery to be
588 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
589 When one of the recipients of a message was unsuitable for the connection
590 (has no matching addresses), we lost track of needing to mark it
591 deferred. As a result mail would be lost.
593 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
595 JH/25 Decoding ACL controls is now done using a binary search; the source code
596 takes up less space and should be simpler to maintain. Merge the ACL
597 condition decode tables also, with similar effect.
599 JH/26 Fix problem with one_time used on a redirect router which returned the
600 parent address unchanged. A retry would see the parent address marked as
601 delivered, so not attempt the (identical) child. As a result mail would
604 JH/27 Fix a possible security hole, wherein a process operating with the Exim
605 UID can gain a root shell. Credit to http://www.halfdog.net/ for
606 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
609 JH/28 Enable {spool,log} filesystem space and inode checks as default.
610 Main config options check_{log,spool}_{inodes,space} are now
611 100 inodes, 10MB unless set otherwise in the configuration.
613 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
614 Previously it only applied to the main-section connection policy
617 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
619 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
620 by me. Added RFC7919 DH primes as an alternative.
622 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
623 are not in the system include path.
625 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
626 GnuTLS, when a session startup failed (eg because the client disconnected)
627 Exim did stdio operations after fclose. This was exposed by a recent
628 change which nulled out the file handle after the fclose.
630 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
631 signed directly by the cert-signing cert, rather than an intermediate
632 OCSP-signing cert. This is the model used by LetsEncrypt.
634 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
636 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
637 an incoming connection.
639 HS/02 Bug 1802: Do not half-close the connection after sending a request
642 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
643 fallback to "prime256v1".
645 JH/34 SECURITY: Use proper copy of DATA command in error message.
646 Could leak key material. Remotely exploitable. CVE-2016-9963.
652 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
653 and 3.4.4 - once the server is enabled to respond to an OCSP request
654 it does even when not requested, resulting in a stapling non-aware
655 client dropping the TLS connection.
657 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
658 support variable-length bit vectors. No functional change.
660 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
661 The I= interface field on outgoing lines is now after the H= remote
662 host field, same as incoming lines. There is a separate
663 outgoing_interface log selector which allows you to disable the
666 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
667 If not running log_selector +smtp_connection the mainlog would be held
668 open indefinitely after a "too many connections" event, including to a
669 deleted file after a log rotate. Leave the per net connection logging
670 leaving it open for efficiency as that will be quickly detected by the
671 check on the next write.
673 HS/01 Bug 1671: Fix post transport crash.
674 Processing the wait-<transport> messages could crash the delivery
675 process if the message IDs didn't exist for some reason. When
676 using 'split_spool_directory=yes' the construction of the spool
677 file name failed already, exposing the same netto behaviour.
679 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
680 mime_regex ACL conditions.
682 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
683 to DSN fail messages (bounces): remote IP, remote greeting, remote response
684 to HELO, local diagnostic string.
686 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
687 log line to debug. Even when configured with a tls authenticator many
688 client connections are expected to not authenticate in this way, so
689 an authenticate fail is not an error.
691 HS/02 Add the Exim version string to the process info. This way exiwhat
692 gives some more detail about the running daemon.
694 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
695 matter for fast-change records such as DNSBLs.
697 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
698 retry record, even if constant. There may be multiple transports with
699 different interface settings and the retry behaviour needs to be kept
702 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
704 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
706 JH/10 Bug 840: fix log_defer_output option of pipe transport
708 JH/11 Bug 830: use same host for all RCPTS of a message, even under
709 hosts_randomize. This matters a lot when combined with mua_wrapper.
711 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
712 ${quote_pgsql:<string>} operator.
714 JH/13 Bug 1708: avoid misaligned access in cached lookup.
716 JH/14 Change header file name for freeradius-client. Relevant if compiling
717 with Radius support; from the Gentoo tree and checked under Fedora.
719 JH/15 Bug 1712: Introduce $prdr_requested flag variable
721 JH/16 Bug 1714: Permit an empty string as expansion result for transport
722 option transport_filter, meaning no filtering.
724 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
726 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
727 defaults to "*" (all hosts). The variable is now available when not built
728 with TLS, default unset, mainly to enable keeping the testsuite sane.
729 If a server certificate is not supplied (via tls_certificate) an error is
730 logged, and clients will find TLS connections fail on startup. Presumably
731 they will retry in-clear.
732 Packagers of Exim are strongly encouraged to create a server certificate
733 at installation time.
735 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
736 with the $config_file variable.
738 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
739 in transport context, after the attempt, and per-recipient. The latter type
740 is per host attempted. The event data is the error message, and the errno
741 information encodes the lookup type (A vs. MX) used for the (first) host,
742 and the trailing two digits of the smtp 4xx response.
744 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
745 to write to mainlog (or rejectlog, paniclog) in the window between file
746 creation and permissions/ownership being changed. Particularly affects
747 installations where exicyclog is run as root, rather than exim user;
748 result is that the running daemon panics and dies.
750 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
752 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
753 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
754 "pri" and "weight". Note that the previous implicit priority given by the
755 list order is no longer honoured.
757 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
760 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
761 by defining SUPPORT_SOCKS.
763 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
764 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
765 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
766 variables are renamed to proxy_{local,external}_{address,port}.
768 JH/27 Move Internationalisation support from Experimental to mainline, enabled
769 for a build by defining SUPPORT_I18N
771 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
772 of the query string, and make ${quote_redis:} do that quoting.
774 JH/29 Move Events support from Experimental to mainline, enabled by default
775 and removable for a build by defining DISABLE_EVENT.
777 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
779 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
780 cached by the daemon.
782 JH/32 Move Redis support from Experimental to mainline, enabled for a build
783 by defining LOOKUP_REDIS. The libhiredis library is required.
785 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
786 keys are given for lookup.
788 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
789 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
790 only supported when built with TLS support. The PolarSSL SHA routines
791 are still used when the TLS library is too old for convenient support.
793 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
794 openssl_options), for security. OpenSSL forces this from version 1.1.0
795 server-side so match that on older versions.
797 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
798 allocation for $value could be released as the expansion processing
799 concluded, but leaving the global pointer active for it.
801 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
802 and to use the domains and local_parts ACL conditions.
804 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
805 incorrectly not doubled on cutthrough transmission, hence seen as a
806 body-termination at the receiving system - resulting in truncated mails.
807 Commonly the sender saw a TCP-level error, and retransmitted the message
808 via the normal store-and-forward channel. This could result in duplicates
809 received - but deduplicating mailstores were liable to retain only the
810 initial truncated version.
812 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
814 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
816 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
817 we're in there, support oversigning also; bug 1309.
819 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
821 HS/04 Add support for keep_environment and add_environment options.
823 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
824 either intentional arithmetic overflow during PRNG, or testing config-
827 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
828 delivery resulted in actual delivery. Cancel cutthrough before DATA
831 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
832 rejects a recipient: pass the reject to the originator.
834 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
835 Many were false-positives and ignorable, but it's worth fixing the
838 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
839 for the new environment-manipulation done at startup. Move the routines
840 from being local to tls.c to being global via the os.c file.
842 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
843 an extract embedded as result-arg for a map, the first arg for extract
844 is unavailable so we cannot tell if this is a numbered or keyed
845 extraction. Accept either.
851 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
854 JH/02 The smtp transport option "multi_domain" is now expanded.
856 JH/03 The smtp transport now requests PRDR by default, if the server offers
859 JH/04 Certificate name checking on server certificates, when exim is a client,
860 is now done by default. The transport option tls_verify_cert_hostnames
861 can be used to disable this per-host. The build option
862 EXPERIMENTAL_CERTNAMES is withdrawn.
864 JH/05 The value of the tls_verify_certificates smtp transport and main options
865 default to the word "system" to access the system default CA bundle.
866 For GnuTLS, only version 3.0.20 or later.
868 JH/06 Verification of the server certificate for a TLS connection is now tried
869 (but not required) by default. The verification status is now logged by
870 default, for both outbound TLS and client-certificate supplying inbound
873 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
876 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
877 Status Notification (bounce) messages are now MIME format per RFC 3464.
878 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
879 under the control of the dsn_advertise_hosts option, and routers may
880 have a dsn_lasthop option.
882 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
883 default, modifiable by a malware= option. The list separator for
884 the options can now be changed in the usual way. Bug 68.
886 JH/10 The smtp_receive_timeout main option is now expanded before use.
888 JH/11 The incoming_interface log option now also enables logging of the
889 local interface on delivery outgoing connections.
891 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
892 if the interface and destination host and port all match.
894 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
897 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
898 Patch from Andrew Lewis.
900 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
901 now supports optional time-restrictions, weighting, and priority
902 modifiers per server. Patch originally by <rommer@active.by>.
904 JH/16 The spamd_address main option now supports a mixed list of local
905 and remote servers. Remote servers can be IPv6 addresses, and
906 specify a port-range.
908 JH/17 Bug 68: The spamd_address main option now supports an optional
909 timeout value per server.
911 JH/18 Bug 1581: Router and transport options headers_add/remove can
912 now have the list separator specified.
914 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
917 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
920 JH/21 Support for the A6 type of dns record is withdrawn.
922 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
923 rather than the verbs used.
925 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
926 from 255 to 1024 chars.
928 JH/24 Verification callouts now attempt to use TLS by default.
930 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
931 are generic router options now. The defaults didn't change.
933 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
934 Original patch from Alexander Shikoff, worked over by JH.
936 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
937 files if ZCAT_COMMAND is not executable.
939 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
941 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
943 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
944 Normally benign, it bites when the pair was led to by a CNAME;
945 modern usage is to not canonicalize the domain to a CNAME target
946 (and we were inconsistent anyway for A-only vs AAAA+A).
948 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
950 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
951 when evaluating $sender_host_dnssec.
953 JH/31 Check the HELO verification lookup for DNSSEC, adding new
954 $sender_helo_dnssec variable.
956 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
958 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
960 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
962 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
963 documented as working, but never had. Support all but $spam_report.
965 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
966 added for tls authenticator.
968 HS/03 Add perl_taintmode main config option
974 TL/01 When running the test suite, the README says that variables such as
975 no_msglog_check are global and can be placed anywhere in a specific
976 test's script, however it was observed that placement needed to be near
977 the beginning for it to behave that way. Changed the runtest perl
978 script to read through the entire script once to detect and set these
979 variables, reset to the beginning of the script, and then run through
980 the script parsing/test process like normal.
982 TL/02 The BSD's have an arc4random API. One of the functions to induce
983 adding randomness was arc4random_stir(), but it has been removed in
984 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
985 function when detected.
987 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
988 cause callback expansion.
990 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
991 syntax errors in an expansion can be treated as a string instead of
992 logging or causing an error, due to the internal use of bool_lax
993 instead of bool when processing it.
995 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
996 server certificates when making smtp deliveries.
998 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
1000 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
1002 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
1004 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
1005 Merged patch from Sebastian Wiedenroth.
1007 JH/05 Fix results-pipe from transport process. Several recipients, combined
1008 with certificate use, exposed issues where response data items split
1009 over buffer boundaries were not parsed properly. This eventually
1010 resulted in duplicates being sent. This issue only became common enough
1011 to notice due to the introduction of connection certificate information,
1012 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1014 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1015 size buffer was used, resulting in syntax errors when an expansion
1018 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1019 version 3.3.6 or later.
1021 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1022 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1023 both become "event_action", the variables become $event_name, $event_data
1024 and $event_defer_errno. There is a new variable $verify_mode, usable in
1025 routers, transports and related events. The tls:cert event is now also
1026 raised for inbound connections, if the main configuration event_action
1029 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1030 early OCSP support, but no stapling (appears to be less than 1.0.0).
1032 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1033 server certificate names available under the smtp transport option
1034 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1037 JH/10 Time-related extraction expansions from certificates now use the main
1038 option "timezone" setting for output formatting, and are consistent
1039 between OpenSSL and GnuTLS compilations. Bug 1541.
1041 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1042 encoded parameter in the incoming message. Bug 1558.
1044 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1045 include certificate info, eximon was claiming there were spoolfile
1048 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1050 JH/14 Log delivery-related information more consistently, using the sequence
1051 "H=<name> [<ip>]" wherever possible.
1053 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1054 are problematic for Debian distribution, omit them from the release
1057 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1059 JH/16 Fix string representation of time values on 64bit time_t architectures.
1062 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1063 output list separator was used.
1068 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1069 checkers that were complaining about end of non-void function with no
1072 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1073 This was a regression introduced in 4.83 by another bugfix.
1075 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1077 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1078 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1084 TF/01 Correctly close the server side of TLS when forking for delivery.
1086 When a message was received over SMTP with TLS, Exim failed to clear up
1087 the incoming connection properly after forking off the child process to
1088 deliver the message. In some situations the subsequent outgoing
1089 delivery connection happened to have the same fd number as the incoming
1090 connection previously had. Exim would try to use TLS and fail, logging
1091 a "Bad file descriptor" error.
1093 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1094 utilities have not been installed.
1096 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1097 temporary space as the ACL may create new global variables.
1099 TL/01 LDAP support uses per connection or global context settings, depending
1100 upon the detected version of the libraries at build time.
1102 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1103 to extract and use the src ip:port in logging and expansions as if it
1104 were a direct connection from the outside internet. PPv2 support was
1105 updated based on HAProxy spec change in May 2014.
1107 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1109 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1110 Properly escape header and check for NULL return.
1112 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1115 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1117 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1118 characters in header names, implemented as a verify condition.
1119 Contributed by Michael Fischer v. Mollard.
1121 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1122 results permerror and temperror. Previous values are deprecated but
1123 still accepted. In a future release, err_perm and err_temp will be
1124 completely removed, which will be a backward incompatibility if the
1125 ACL tests for either of these two old results. Patch contributed by
1126 user bes-internal on the mailing list.
1128 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1130 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1131 selectors, in both main and reject logs.
1133 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1136 JH/07 Add malware type "sock" for talking to simple daemon.
1138 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1140 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1141 routers/transports under cutthrough routing.
1143 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1144 numbers. Touch up "bool" conditional to keep the same definition.
1146 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1148 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1150 JH/12 Expand items in router/transport headers_add or headers_remove lists
1151 individually rather than the list as a whole. Bug 1452.
1153 Required for reasonable handling of multiple headers_ options when
1154 they may be empty; requires that headers_remove items with embedded
1155 colons must have them doubled (or the list-separator changed).
1157 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1158 view the policy declared in the DMARC record. Currently, $dmarc_status
1159 is a combined value of both the record presence and the result of the
1162 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1164 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1165 dnslookup router and the smtp transport (applying to the forward
1168 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1169 of ldap servers used for a specific lookup. Patch provided by Heiko
1172 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1173 New variable $lookup_dnssec_authenticated for observability.
1175 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1176 Patch submitted by Lars Timman.
1178 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1180 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1181 Requires trusted mode and valid format message id, aborts otherwise.
1182 Patch contributed by Heiko Schlichting.
1184 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1185 certextract with support for various fields. Bug 1358.
1187 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1188 is requested by default, modifiable by smtp transport option
1191 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1192 operate on certificate variables to give certificate fingerprints
1193 Also new ${sha256:cert_variable}.
1195 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1197 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1198 Christian Aistleitner.
1200 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1202 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1203 file. Patch from Wolfgang Breyha.
1205 JH/25 Expand the coverage of the delivery $host and $host_address to
1206 client authenticators run in verify callout. Bug 1476.
1208 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1209 align with daemon_smtp_ports. Bug 72.
1211 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1212 support and error reporting did not work properly.
1214 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1215 and is readable. Patch from Andrew Colin Kissa.
1217 TL/14 Enhance documentation of ${run expansion and how it parses the
1218 commandline after expansion, particularly in the case when an
1219 unquoted variable expansion results in an empty value.
1221 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1223 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1224 by renaming away from it. Addresses GH issue 15, reported by
1227 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1229 TL/15 SECURITY: prevent double expansion in math comparison functions
1230 (can expand unsanitized data). Not remotely exploitable.
1237 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1239 PP/02 Make -n do something, by making it not do something.
1240 When combined with -bP, the name of an option is not output.
1242 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1245 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1246 $sender_host_name and config options to manage this, and basic check
1249 PP/05 DSCP support for outbound connections and control modifier for inbound.
1251 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1252 (Only plugin which currently uses this is kerberos4, which nobody should
1253 be using, but we should make it available and other future plugins might
1254 conceivably use it, even though it would break NAT; stuff *should* be
1255 using channel bindings instead).
1257 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1258 name; added for Sendmail compatibility; requires admin caller.
1259 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1260 just ignore it); requires trusted caller.
1261 Also parse but ignore: -Ac -Am -X<logfile>
1264 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1266 TL/02 Add +smtp_confirmation as a default logging option.
1268 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1269 Patch by Magnus Holmgren from 2007-02-20.
1271 TL/04 Bugzilla 1281 - Spec typo.
1272 Bugzilla 1283 - Spec typo.
1273 Bugzilla 1290 - Spec grammar fixes.
1275 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1277 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1279 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1280 mailing list by by Dmitry Isaikin.
1282 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1284 JH/02 Support "G" suffix to numbers in ${if comparisons.
1286 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1288 NM/01 Bugzilla 1197 - Spec typo
1289 Bugzilla 1196 - Spec examples corrections
1291 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1293 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1294 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1297 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1298 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1300 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1301 "acl {{name}{arg}...}", and optional args on acl condition
1304 JH/05 Permit multiple router/transport headers_add/remove lines.
1306 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1308 JH/07 Avoid using a waiting database for a single-message-only transport.
1309 Performance patch from Paul Fisher. Bugzilla 1262.
1311 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1314 JH/09 Add $headers_added variable, with content from use of ACL modifier
1315 add_header (but not yet added to the message). Bugzilla 199.
1317 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1318 Pulled from Bugzilla 817 by Wolfgang Breyha.
1320 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1322 (nb: this is the same fix as in Exim 4.80.1)
1324 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1327 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1330 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1332 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1333 advertise SMTP AUTH mechanism to us, instead of a generic
1334 protocol violation error. Also, make Exim more robust to bad
1335 data from the Dovecot auth socket.
1337 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1339 When a queue runner is handling a message, Exim first routes the
1340 recipient addresses, during which it prunes them based on the retry
1341 hints database. After that it attempts to deliver the message to
1342 any remaining recipients. It then updates the hints database using
1345 So if a recipient address works intermittently, it can get repeatedly
1346 deferred at routing time. The retry hints record remains fresh so the
1347 address never reaches the final cutoff time.
1349 This is a fairly common occurrence when a user is bumping up against
1350 their storage quota. Exim had some logic in its local delivery code
1351 to deal with this. However it did not apply to per-recipient defers
1352 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1354 This change adds a proper retry rule check during routing so that the
1355 final cutoff time is checked against the message's age. We only do
1356 this check if there is an address retry record and there is not a
1357 domain retry record; this implies that previous attempts to handle
1358 the address had the retry_use_local_parts option turned on. We use
1359 this as an approximation for the destination being like a local
1360 delivery, as in LMTP.
1362 I suspect this new check makes the old local delivery cutoff check
1363 redundant, but I have not verified this so I left the code in place.
1365 TF/02 Correct gecos expansion when From: is a prefix of the username.
1367 Test 0254 submits a message to Exim with the header
1371 When I ran the test suite under the user fanf2, Exim expanded
1372 the header to contain my full name, whereas it should have added
1373 a Resent-Sender: header. It erroneously treats any prefix of the
1374 username as equal to the username.
1376 This change corrects that bug.
1378 GF/01 DCC debug and logging tidyup
1379 Error conditions log to paniclog rather than rejectlog.
1380 Debug lines prefixed by "DCC: " to remove any ambiguity.
1382 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1384 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1385 Bug spotted by Jeremy Harris; was flawed since initial commit.
1386 Would have resulted in OCSP responses post-SNI triggering an Exim
1387 NULL dereference and crash.
1389 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1391 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1392 Bug detection, analysis and fix by Samuel Thibault.
1393 Bugzilla 1331, Debian bug #698092.
1395 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1397 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1398 Server implementation by Todd Lyons, client by JH.
1399 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1400 config variable "prdr_enable" controls whether the server
1401 advertises the facility. If the client requests PRDR a new
1402 acl_data_smtp_prdr ACL is called once for each recipient, after
1403 the body content is received and before the acl_smtp_data ACL.
1404 The client is controlled by both of: a hosts_try_prdr option
1405 on the smtp transport, and the server advertisement.
1406 Default client logging of deliveries and rejections involving
1407 PRDR are flagged with the string "PRDR".
1409 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1410 fclose(). Diagnosis by Todd Lyons.
1412 PP/17 Update configure.default to handle IPv6 localhost better.
1413 Patch by Alain Williams (plus minor tweaks).
1416 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1417 This is now consistent with GnuTLS, and is now documented: the
1418 previous undocumented portable approach to treating the option as
1419 unset was to force an expansion failure. That still works, and
1420 an empty string is now equivalent.
1422 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1423 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1424 not performing validation itself.
1426 PP/20 Added force_command boolean option to pipe transport.
1427 Patch from Nick Koston, of cPanel Inc.
1429 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1432 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1434 PP/21 Fix eximon continuous updating with timestamped log-files.
1435 Broken in a format-string cleanup in 4.80, missed when I repaired the
1436 other false fix of the same issue.
1437 Report and fix from Heiko Schlichting.
1440 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1441 Report from Prashanth Katuri.
1443 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1444 It's SecureTransport, so affects any MacOS clients which use the
1445 system-integrated TLS libraries, including email clients.
1447 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1448 using a MIME ACL for non-SMTP local injection.
1449 Report and assistance in diagnosis by Warren Baker.
1451 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1453 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1455 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1456 last id that failed so it may be referenced in subsequent ACL's.
1458 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1461 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1462 ldap library initialization, allowing self-signed CA's to be
1463 used. Also properly sets require_cert option later in code by
1464 using NULL (global ldap config) instead of ldap handle (per
1465 session). Bug diagnosis and testing by alxgomz.
1467 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1468 the src/util/ subdirectory.
1470 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1471 renamed to Transport Post Delivery Action by Jeremy Harris, as
1474 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1475 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1476 redis_servers = needs to be configured which will be used by the redis
1477 lookup. Patch from Warren Baker, of The Packet Hub.
1479 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1481 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1482 hostname or reverse DNS when processing a host list. Used suggestions
1483 from multiple comments on this bug.
1485 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1487 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1488 Missed a few lines, added it to make the runtest require no keyboard
1491 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1492 contains upper case chars. Make router use caseful_local_part.
1494 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1495 support when GnuTLS has been built with p11-kit.
1501 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1503 This, or similar/improved, will also be change PP/11 of 4.82.
1509 PP/01 Handle short writes when writing local log-files.
1510 In practice, only affects FreeBSD (8 onwards).
1511 Bugzilla 1053, with thanks to Dmitry Isaikin.
1513 NM/01 Bugzilla 949 - Documentation tweak
1515 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1518 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1520 PP/02 Implemented gsasl authenticator.
1522 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1524 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1525 `pkg-config foo` for cflags/libs.
1527 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1528 with rest of GSASL and with heimdal_gssapi.
1530 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1531 `pkg-config foo` for cflags/libs for the TLS implementation.
1533 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1534 properties get this fed in as external SSF. A number of robustness
1535 and debugging improvements to the cyrus_sasl authenticator.
1537 PP/08 cyrus_sasl server now expands the server_realm option.
1539 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1540 Patch by Jeremy Harris.
1542 PP/10 Added dbmjz lookup type.
1544 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1546 PP/12 MAIL args handles TAB as well as SP, for better interop with
1547 non-compliant senders.
1548 Analysis and variant patch by Todd Lyons.
1550 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1551 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1552 Patch from Dirk Mueller <dmueller@suse.com>
1554 PP/13 tls_peerdn now print-escaped for spool files.
1555 Observed some $tls_peerdn in wild which contained \n, which resulted
1556 in spool file corruption.
1558 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1559 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1560 or write after TLS renegotiation, which otherwise led to messages
1563 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1564 as a tracking header (ie: a signed header comes before the signature).
1565 Patch from Wolfgang Breyha.
1567 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1568 comma-sep list; embedded commas doubled.
1570 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1572 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1574 Report and patch from Dmitry Banschikov.
1576 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1577 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1578 failures. We appear to now support TLS1.1+ with Exim.
1580 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1581 lets Exim select keys and certificates based upon TLS SNI from client.
1582 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1583 before an outbound SMTP session. New log_selector, +tls_sni.
1585 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1586 NULL dereference. Report and patch from Alun Jones.
1588 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1589 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1590 Not seeing resolver debug output on NetBSD, but suspect this is a
1591 resolver implementation change.
1593 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1594 Left warnings. Added "eximon gdb" invocation mode.
1596 PP/21 Defaulting "accept_8bitmime" to true, not false.
1598 PP/22 Added -bw for inetd wait mode support.
1600 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1601 locate the relevant includes and libraries. Made this the default.
1603 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1604 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1606 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1607 This may cause build issues on older platforms.
1609 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1610 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1611 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1612 Added SNI support via GnuTLS too.
1613 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1615 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1617 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1618 Applied second patch from Janne, implementing suggestion to default
1619 multiple-strings-in-record handling to match SPF spec.
1621 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1623 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1624 read-only, out of scope).
1625 Patch from Wolfgang Breyha, report from Stuart Northfield.
1627 PP/29 Fix three issues highlighted by clang analyser static analysis.
1628 Only crash-plausible issue would require the Cambridge-specific
1629 iplookup router and a misconfiguration.
1630 Report from Marcin Mirosław.
1632 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1634 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1635 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1636 As part of this, removing so much warning spew let me fix some minor
1637 real issues in debug logging.
1639 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1640 assignment on my part. Fixed.
1642 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1643 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1644 Janne Snabb (who went above and beyond: thank you).
1646 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1647 string otherwise requires a connection and a bunch more work and it's
1648 relatively easy to get wrong. Should also expose TLS library linkage
1651 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1652 64-bit ${eval} (JH/03).
1654 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1655 GNU libc to support some of the 64-bit stuff, should not lead to
1656 conflicts. Defined before os.h is pulled in, so if a given platform
1657 needs to override this, it can.
1659 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1660 protection layer was required, which is not implemented.
1661 Bugzilla 1254, patch from Wolfgang Breyha.
1663 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1664 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1665 tls_dhparam take prime identifiers. Also unbreak combination of
1666 OpenSSL+DH_params+TLSSNI.
1668 PP/39 Disable SSLv2 by default in OpenSSL support.
1674 PP/01 Solaris build fix for Oracle's LDAP libraries.
1675 Bugzilla 1109, patch from Stephen Usher.
1677 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1679 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1682 TF/02 Fix a couple more cases where we did not log the error message
1683 when unlink() failed. See also change 4.74-TF/03.
1685 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1686 lock up or crash if it happened to be inside a call to libc when it
1687 got a SIGUSR1 from exiwhat.
1689 The SIGUSR1 handler appends the current process status to the process
1690 log which is later printed by exiwhat. It used to use the general
1691 purpose logging code to do this, but several functions it calls are
1692 not safe for signals.
1694 The new output code in the SIGUSR1 handler is specific to the process
1695 log, and simple enough that it's easy to inspect for signal safety.
1696 Removing some special cases also simplifies the general logging code.
1697 Removing the spurious timestamps from the process log simplifies
1700 TF/04 Improved ratelimit ACL condition.
1702 The /noupdate option has been deprecated in favour of /readonly which
1703 has clearer semantics. The /leaky, /strict, and /readonly update modes
1704 are mutually exclusive. The update mode is no longer included in the
1705 database key; it just determines when the database is updated. (This
1706 means that when you upgrade Exim will forget old rate measurements.)
1708 Exim now checks that the per_* options are used with an update mode that
1709 makes sense for the current ACL. For example, when Exim is processing a
1710 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1711 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1712 must specify per_mail/readonly. If you omit the update mode it defaults to
1713 /leaky where that makes sense (as before) or /readonly where required.
1715 The /noupdate option is now undocumented but still supported for
1716 backwards compatibility. It is equivalent to /readonly except that in
1717 ACLs where /readonly is required you may specify /leaky/noupdate or
1718 /strict/noupdate which are treated the same as /readonly.
1720 A useful new feature is the /count= option. This is a generalization
1721 of the per_byte option, so that you can measure the throughput of other
1722 aggregate values. For example, the per_byte option is now equivalent
1723 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1725 The per_rcpt option has been generalized using the /count= mechanism
1726 (though it's more complicated than the per_byte equivalence). When it is
1727 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1728 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1729 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1730 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1731 Note that using per_rcpt with a non-readonly update mode in more than
1732 one ACL will cause the recipients to be double-counted. (The per_mail
1733 and per_byte options don't have this problem.)
1735 The handling of very low rates has changed slightly. If the computed rate
1736 is less than the event's count (usually one) then this event is the first
1737 after a long gap. In this case the rate is set to the same as this event's
1738 count, so that the first message of a spam run is counted properly.
1740 The major new feature is a mechanism for counting the rate of unique
1741 events. The new per_addr option counts the number of different
1742 recipients that someone has sent messages to in the last time period. It
1743 behaves like per_rcpt if all the recipient addresses are different, but
1744 duplicate recipient addresses do not increase the measured rate. Like
1745 the /count= option this is a general mechanism, so the per_addr option
1746 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1747 example, measure the rate that a client uses different sender addresses
1748 with the options per_mail/unique=$sender_address. There are further
1749 details in the main documentation.
1751 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1753 TF/06 Removed a few PCRE remnants.
1755 TF/07 Automatically extract Exim's version number from tags in the git
1756 repository when doing development or release builds.
1758 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1759 Bugzilla 879. Patch from Paul Fisher.
1761 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1762 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1765 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1767 PP/05 Variable $av_failed, true if the AV scanner deferred.
1768 Bugzilla 1078. Patch from John Horne.
1770 PP/06 Stop make process more reliably on build failure.
1771 Bugzilla 1087. Patch from Heiko Schlittermann.
1773 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1774 Bugzilla 1089. Patch from Heiko Schlittermann.
1776 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1777 Bugzilla 1131. Patch from Holger Weiß.
1779 PP/09 Handle IPv6 addresses with SPF.
1780 Bugzilla 860. Patch from Wolfgang Breyha.
1782 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1784 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1787 PP/11 match_* no longer expand right-hand-side by default.
1788 New compile-time build option, EXPAND_LISTMATCH_RHS.
1789 New expansion conditions, "inlist", "inlisti".
1791 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1793 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1795 PP/14 fix log_write() format string regression from TF/03.
1796 Bugzilla 1152. Patch from Dmitry Isaikin.
1802 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1804 PP/02 Harmonised TLS library version reporting; only show if debugging.
1805 Layout now matches that introduced for other libraries in 4.74 PP/03.
1807 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1809 PP/04 New "dns_use_edns0" global option.
1811 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1814 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1815 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1817 TK/01 Updated PolarSSL code to 0.14.2.
1818 Bugzilla 1097. Patch from Andreas Metzler.
1820 PP/07 Catch divide-by-zero in ${eval:...}.
1821 Fixes bugzilla 1102.
1823 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1826 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1827 format-string attack -- SECURITY: remote arbitrary code execution.
1829 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1830 time unintentionally subject to list matching rules, letting the header
1831 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1832 arbitrary string expansion). This allowed for information disclosure.
1834 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1835 INT_MIN/-1 -- value coerced to INT_MAX.
1841 NM/01 Workaround for PCRE version dependency in version reporting
1844 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1845 This fixes portability to compilers other than gcc, notably
1846 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1848 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1849 makefiles for portability to HP-UX and POSIX correctness.
1851 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1852 Also via indented variable definition in the Makefile.
1853 (Debugging by Oliver Heesakkers).
1855 PP/02 Restore caching of spamd results with expanded spamd_address.
1856 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1858 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1859 Improves build reliability. Fix from: Frank Elsner
1861 NM/02 Fix wide character breakage in the rfc2047 coding
1862 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1864 NM/03 Allow underscore in dnslist lookups
1865 Fixes bug 1026. Patch from Graeme Fowler
1867 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1868 Code patches from Adam Ciarcinski of NetBSD.
1870 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1873 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1874 is logged, to avoid truncation. Patch from John Horne.
1876 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1877 Patch from Jakob Hirsch.
1879 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1880 SQL string expansion failure details.
1881 Patch from Andrey Oktyabrski.
1883 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1884 Patch from Simon Arlott.
1886 PP/09 New lookups functionality failed to compile on old gcc which rejects
1887 extern declarations in function scope.
1888 Patch from Oliver Fleischmann
1890 PP/10 Use sig_atomic_t for flags set from signal handlers.
1891 Check getgroups() return and improve debugging.
1892 Fixed developed for diagnosis in bug 927 (which turned out to be
1895 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1896 Patch from Mark Zealey.
1898 PP/12 Bugzilla 1056: Improved spamd server selection.
1899 Patch from Mark Zealey.
1901 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1902 Based on patch from Heiko Schlittermann.
1904 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1905 Patch from Uwe Doering, sign-off by Michael Haardt.
1907 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1908 variable declaration deep within a block. Bug and patch from
1911 PP/15 lookups-Makefile IRIX compatibility coercion.
1913 PP/16 Make DISABLE_DKIM build knob functional.
1915 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1916 Patch by Simon Arlott
1918 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1919 variable argument macros. Our copy now differs from upstream.
1925 TF/01 Failure to get a lock on a hints database can have serious
1926 consequences so log it to the panic log.
1928 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1929 controlled using the smtp_confirmation log selector.
1931 TF/03 Include the error message when we fail to unlink a spool file.
1933 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1934 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1935 for maintaining out-of-tree patches for some time.
1937 PP/01 Bugzilla 139: Documentation and portability issues.
1938 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1939 Handle per-OS dynamic-module compilation flags.
1941 PP/02 Let /dev/null have normal permissions.
1942 The 4.73 fixes were a little too stringent and complained about the
1943 permissions on /dev/null. Exempt it from some checks.
1944 Reported by Andreas M. Kirchwitz.
1946 PP/03 Report version information for many libraries, including
1947 Exim version information for dynamically loaded libraries. Created
1948 version.h, now support a version extension string for distributors
1949 who patch heavily. Dynamic module ABI change.
1951 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1952 privilege escalation vulnerability whereby the Exim run-time user
1953 can cause root to append content of the attacker's choosing to
1956 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1959 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1960 If dropping privileges for untrusted macros, we disabled normal logging
1961 on the basis that it would fail; for the Exim run-time user, this is not
1962 the case, and it resulted in successful deliveries going unlogged.
1963 Fixed. Reported by Andreas Metzler.
1969 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1970 only prepend for the Resent-* case. Fixes regression introduced in
1971 Exim 4.70 by NM/22 for Bugzilla 607.
1973 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1974 increasing numbers of administrators be bitten by this.
1976 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1978 PP/03 Bugzilla 994: added openssl_options main configuration option.
1980 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1982 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1984 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1986 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1987 without a peer certificate, leading to a segfault because of an
1988 assumption that peers always have certificates. Be a little more
1989 paranoid. Problem reported by Martin Tscholak.
1991 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1992 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1993 NB: ClamAV planning to remove STREAM in "middle of 2010".
1994 CL also introduces -bmalware, various -d+acl logging additions and
1995 more caution in buffer sizes.
1997 PP/09 Implemented reverse_ip expansion operator.
1999 PP/10 Bugzilla 937: provide a "debug" ACL control.
2001 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
2003 PP/12 Bugzilla 973: Implement --version.
2005 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2007 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2009 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2011 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2012 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2013 ignore trailing whitespace.
2015 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2017 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2020 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2021 Notification from Dr Andrew Aitchison.
2023 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2024 ExtendedDetectionInfo response format.
2025 Notification from John Horne.
2027 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2030 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2031 XSL and documented dependency on system catalogs, with examples of how
2034 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2037 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2038 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2039 configuration file which is writeable by the Exim user or group.
2041 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2042 of configuration files to cover files specified with the -C option if
2043 they are going to be used with root privileges, not just the default
2046 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2047 option (effectively making it always true).
2049 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2050 files to be used while preserving root privileges.
2052 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2053 that rogue child processes cannot use them.
2055 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2056 run-time user, instead of root.
2058 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2059 Exim run-time user without dropping privileges.
2061 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2062 result string, instead of calling string_vformat() twice with the same
2065 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2066 for other users. Others should always drop root privileges if they use
2067 -C on the command line, even for a whitelisted configure file.
2069 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2071 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2077 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2078 $data_path, and $header_path variables; fixed documentation bugs and
2081 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2082 exipick to access non-standard spools, including the "frozen" queue
2085 NM/01 Bugzilla 965: Support mysql stored procedures.
2086 Patch from Alain Williams
2088 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2090 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2091 Patch from Andreas Metzler
2093 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2094 Patch from Kirill Miazine
2096 NM/05 Bugzilla 671: Added umask to procmail example.
2098 JJ/03 installed exipick 20100323.0, fixing doc bug
2100 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2101 directory. Notification and patch from Dan Rosenberg.
2103 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2105 TK/02 Improve log output when DKIM signing operation fails.
2107 MH/01 Treat the transport option dkim_domain as a colon separated
2108 list, not as a single string, and sign the message with each element,
2109 omitting multiple occurences of the same signer.
2111 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2112 Bugzilla 985, 986. Patch by Simon Arlott
2114 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2115 Patch by Simon Arlott
2117 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2118 MBX locking. Notification from Dan Rosenberg.
2124 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2126 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2128 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2130 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2132 NM/04 Bugzilla 915: Flag broken perl installation during build.
2138 TK/01 Added patch by Johannes Berg that expands the main option
2139 "spamd_address" if it starts with a dollar sign.
2141 TK/02 Write list of recipients to X-Envelope-Sender header when building
2142 the mbox-format spool file for content scanning (suggested by Jakob
2145 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2146 (http://www.dcc-servers.net/) support via dccifd. Activated by
2147 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2149 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2150 by Mark Daniel Reidel <mr@df.eu>.
2152 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2153 When building exim an external PCRE library is now needed -
2154 PCRE is a system library on the majority of modern systems.
2155 See entry on PCRE_LIBS in EDITME file.
2157 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2158 conversation. Added nologin parameter to request.
2159 Patch contributed by Kirill Miazine.
2161 TF/01 Do not log submission mode rewrites if they do not change the address.
2163 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2165 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2166 log files in place. Contributed by Roberto Lima.
2168 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2170 TF/03 Bugzilla 615: When checking the local_parts router precondition
2171 after a local_part_suffix or local_part_prefix option, Exim now
2172 does not use the address's named list lookup cache, since this
2173 contains cached lookups for the whole local part.
2175 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2176 Robert Millan. Documentation is in experimental-spec.txt.
2178 TF/04 Bugzilla 668: Fix parallel build (make -j).
2180 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2182 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2183 Patch provided by Jan Srzednicki.
2185 TF/05 Leading white space used to be stripped from $spam_report which
2186 wrecked the formatting. Now it is preserved.
2188 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2189 that they are available at delivery time.
2191 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2193 TF/08 TLS error reporting now respects the incoming_interface and
2194 incoming_port log selectors.
2196 TF/09 Produce a more useful error message if an SMTP transport's hosts
2197 setting expands to an empty string.
2199 NM/06 Bugzilla 744: EXPN did not work under TLS.
2200 Patch provided by Phil Pennock.
2202 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2203 Patch provided by Richard Godbee.
2205 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2206 acl_smtp_notquit, added index entry.
2208 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2209 Patch provided by Eugene Bujak.
2211 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2212 accept(). Patch provided by Maxim Dounin.
2214 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2215 Patch provided by Phil Pennock.
2217 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2219 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2220 Patch provided by Brad "anomie" Jorsch.
2222 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2223 Patch provided by Dean Brooks.
2225 TK/05 Add native DKIM support (does not depend on external libraries).
2227 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2228 Patch provided by Graeme Fowler.
2230 NM/16 Bugzilla 851: Documentation example syntax fix.
2232 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2234 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2237 NM/19 Bugzilla 745: TLS version reporting.
2238 Patch provided by Phil Pennock.
2240 NM/20 Bugzilla 167: bool: condition support.
2241 Patch provided by Phil Pennock.
2243 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2244 clients. Patch provided by Phil Pennock.
2246 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2247 Patch provided by Brad "anomie" Jorsch.
2249 NM/23 Bugzilla 687: Fix misparses in eximstats.
2250 Patch provided by Heiko Schlittermann.
2252 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2253 Patch provided by Heiko Schlittermann.
2255 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2256 plus update to original patch.
2258 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2260 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2261 Patch provided by David Brownlee.
2263 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2265 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2267 NM/30 Bugzilla 888: TLS documentation bugfixes.
2269 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2271 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2272 Unlike the original bugzilla I have changed all shell scripts in src tree.
2274 NM/33 Bugzilla 898: Transport filter timeout fix.
2275 Patch by Todd Rinaldo.
2277 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2278 Patch by Serge Demonchaux.
2280 NM/35 Bugzilla 39: Base64 decode bug fixes.
2281 Patch by Jakob Hirsch.
2283 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2285 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2287 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2289 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2295 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2296 ALT-N's libdkim that I have put here:
2297 http://duncanthrax.net/exim-experimental/
2299 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2300 were called 'true' and it seems that C99 defines that as a reserved
2301 keyword to be used with 'bool' variable types. That means you could
2302 not include C99-style headers which use bools without triggering
2303 build errors in sieve.c.
2305 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2306 as mailq or other aliases. Changed the --help handling significantly
2307 to do whats expected. exim_usage() emits usage/help information.
2309 SC/01 Added the -bylocaldomain option to eximstats.
2311 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2313 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2315 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2321 PH/01 Another patch from the Sieve maintainer.
2323 PH/02 When an IPv6 address is converted to a string for single-key lookup
2324 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2325 dots are used instead of colons so that keys in lsearch files need not
2326 contain colons. This was done some time before quoting was made available
2327 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2328 (notated using the quote facility) so as to distinguish them from IPv4
2329 keys. This meant that lookups for IP addresses in host lists did not work
2330 for iplsearch lookups.
2332 This has been fixed by arranging for IPv6 addresses to be expressed with
2333 colons if the lookup type is iplsearch. This is not incompatible, because
2334 previously such lookups could never work.
2336 The situation is now rather anomalous, since one *can* have colons in
2337 ordinary lsearch keys. However, making the change in all cases is
2338 incompatible and would probably break a number of configurations.
2340 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2343 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2344 conversion specification without a maximum field width, thereby enabling
2345 a rogue spamd server to cause a buffer overflow. While nobody in their
2346 right mind would setup Exim to query an untrusted spamd server, an
2347 attacker that gains access to a server running spamd could potentially
2348 exploit this vulnerability to run arbitrary code as the Exim user.
2350 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2351 $primary_hostname instead of what libspf2 thinks the hosts name is.
2353 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2354 a directory entry by the name of the lookup key. Previously, if a
2355 symlink pointed to a non-existing file or a file in a directory that
2356 Exim lacked permissions to read, a lookup for a key matching that
2357 symlink would fail. Now it is enough that a matching directory entry
2358 exists, symlink or not. (Bugzilla 503.)
2360 PH/03 The body_linecount and body_zerocount variables are now exported in the
2363 PH/04 Added the $dnslist_matched variable.
2365 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2366 This means they are set thereafter only if the connection becomes
2369 PH/06 Added the client_condition to authenticators so that some can be skipped
2370 by clients under certain conditions.
2372 PH/07 The error message for a badly-placed control=no_multiline_responses left
2373 "_responses" off the end of the name.
2375 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2377 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2378 (without spaces) instead of just copying the configuration text.
2380 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2382 PH/11 Added $max_received_linelength.
2384 PH/12 Added +ignore_defer and +include_defer to host lists.
2386 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2387 way in which PCRE > 7.0 is built.
2389 PH/14 Implemented queue_only_load_latch.
2391 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2392 MAIL command. The effect was to mangle the value on 64-bit systems.
2394 PH/16 Another patch from the Sieve maintainer.
2396 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2398 PH/18 If a system quota error occurred while trying to create the file for
2399 a maildir delivery, the message "Mailbox is full" was not appended to the
2400 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2401 only to a quota excession during the actual writing of the file.
2403 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2404 characters?) which causes problems in log lines. The DN values are now
2405 passed through string_printing() before being added to log lines.
2407 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2408 and InterBase are left for another time.)
2410 PH/21 Added message_body_newlines option.
2412 PH/22 Guard against possible overflow in moan_check_errorcopy().
2414 PH/23 POSIX allows open() to be a macro; guard against that.
2416 PH/24 If the recipient of an error message contained an @ in the local part
2417 (suitably quoted, of course), incorrect values were put in $domain and
2418 $local_part during the evaluation of errors_copy.
2424 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2425 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2428 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2429 issue a MAIL command.
2431 PH/02 In an ACL statement such as
2433 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2435 if a client was not listed at all, or was listed with a value other than
2436 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2437 the condition was not true (as it should be), so access was not denied.
2438 The bug was that the ! inversion was incorrectly passed on to the second
2439 item. This has been fixed.
2441 PH/03 Added additional dnslists conditions == and =& which are different from
2442 = and & when the dns lookup returns more than one IP address.
2444 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2445 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2447 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2448 FSYNC, which compiles an option called disable_fsync that allows for
2449 bypassing fsync(). The documentation is heavily laced with warnings.
2451 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2453 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2454 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2455 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2456 including adding "make clean"; (3) Added -fPIC when compiling the test
2457 dynamically loaded module, to get rid of a warning.
2459 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2460 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2461 = 0s. The bug is otherwise harmless.
2463 PH/07 There was a bug in the dovecot authenticator such that the value of
2464 $auth1 could be overwritten, and so not correctly preserved, after a
2465 successful authentication. This usually meant that the value preserved by
2466 the server_setid option was incorrect.
2468 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2470 PH/09 Installed PCRE release 7.0.
2472 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2473 run for batched SMTP input. It is now run at the start of every message
2474 in the batch. While fixing this I discovered that the process information
2475 (output by running exiwhat) was not always getting set for -bs and -bS
2476 input. This is fixed, and it now also says "batched" for BSMTP.
2478 PH/11 Added control=no_pipelining.
2480 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2481 patch, slightly modified), and move the expansion of helo_data till after
2482 the connection is made in the smtp transport (so it can use these
2485 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2487 PH/14 Added log_selector = +pid.
2489 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2491 PH/16 Add ${if forany and ${if forall.
2493 PH/17 Added dsn_from option to vary the From: line in DSNs.
2495 PH/18 Flush SMTP output before performing a callout, unless control =
2496 no_callout_flush is set.
2498 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2499 was true (the default) a successful delivery failed to delete the retry
2500 item, thus causing premature timeout of the address. The bug is now
2503 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2505 PH/21 Long custom messages for fakedefer and fakereject are now split up
2506 into multiline responses in the same way that messages for "deny" and
2507 other ACL rejections are.
2509 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2510 with slight modification.
2512 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2513 draft, changing the syntax and factoring some duplicate code".
2515 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2516 for deliveries of the second and subsequent messages over the same SMTP
2519 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2520 ${reduce, with only minor "tidies".
2522 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2524 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2525 expansion side effects.
2527 PH/27 When a message times out after an over-quota error from an Exim-imposed
2528 quota, the bounce message says "mailbox is full". This message was not
2529 being given when it was a system quota that was exceeded. It now should
2532 MH/03 Made $recipients available in local_scan(). local_scan() already has
2533 better access to the recipient list through recipients_list[], but
2534 $recipients can be useful in postmaster-provided expansion strings.
2536 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2537 in the case of a MAIL command with additional options following the
2538 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2539 were accidentally chopped off.
2541 PH/29 SMTP synchronization checks are implemented when a command is read -
2542 there is a check that no more input is waiting when there shouldn't be
2543 any. However, for some commands, a delay in an ACL can mean that it is
2544 some time before the response is written. In this time, more input might
2545 arrive, invalidly. So now there are extra checks after an ACL has run for
2546 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2547 pipelining has not been advertised.
2549 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2551 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2552 returned in the SMTP response when smtp_return_error_details was false.
2553 This has been fixed.
2555 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2556 instead of the C I/O that was originally supplied, because problems were
2557 reported on Solaris.
2559 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2560 Exim which did not show up earlier: it was assuming that a call to
2561 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2562 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2563 was a macro that became an assignment, so it seemed to work. This has
2564 changed to a proper function call with a void return, hence the compile
2565 error. Exim's code has been fixed.
2567 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2570 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2572 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2574 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2575 because it turns out to be too big or there is a timeout) there is no
2576 "Completed" line in the log. When some messages of this type were
2577 selected by exigrep, they were listed as "not completed". Others were
2578 picked up by some special patterns. I have improved the selection
2579 criteria to be more general.
2581 PH/38 The host_find_failed option in the manualroute router can now be set
2582 to "ignore", to completely ignore a host whose IP address cannot be
2583 found. If all hosts are ignored, the behaviour is controlled by the new
2584 host_all_ignored option.
2586 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2587 homing or because of multiple MX records with /mx) generated more than
2588 one IP address, and the following item turned out to be the local host,
2589 all the secondary addresses of the first item were incorrectly removed
2590 from the list, along with the local host and any following hosts (which
2591 is what is supposed to happen).
2593 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2594 whoever called Exim into the -H file. In the case of the daemon it was
2595 behaving confusingly. When first started, it used values for whoever
2596 started the daemon, but after a SIGHUP it used the Exim user (because it
2597 calls itself on a restart). I have changed the code so that it now always
2600 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2601 message are rejected with the same error (e.g. no authentication or bad
2602 sender address), and a DATA command is nevertheless sent (as can happen
2603 with PIPELINING or a stupid MUA), the error message that was given to the
2604 RCPT commands is included in the rejection of the DATA command. This is
2605 intended to be helpful for MUAs that show only the final error to their
2608 PH/42 Another patch from the Sieve maintainer.
2610 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2611 Eximstats - Fixed some broken HTML links and added missing column headers
2613 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2614 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2616 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2618 PH/43 Yet another patch from the Sieve maintainer.
2620 PH/44 I found a way to check for a TCP/IP connection going away before sending
2621 the response to the final '.' that terminates a message, but only in the
2622 case where the client has not sent further data following the '.'
2623 (unfortunately, this is allowed). However, in many cases there won't be
2624 any further data because there won't be any more messages to send. A call
2625 to select() can be used: if it shows that the input is "ready", there is
2626 either input waiting, or the socket has been closed. An attempt to read
2627 the next input character can distinguish the two cases. Previously, Exim
2628 would have sent an OK response which the client would never have see.
2629 This could lead to message repetition. This fix should cure that, at
2630 least in a lot of common cases.
2632 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2633 advertised in response to EHLO.
2639 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2640 fixed by 4.65/MH/01 (is this a record?) are fixed:
2642 (i) An empty string was always treated as zero by the numeric comparison
2643 operators. This behaviour has been restored.
2645 (ii) It is documented that the numeric comparison operators always treat
2646 their arguments as decimal numbers. This was broken in that numbers
2647 starting with 0 were being interpreted as octal.
2649 While fixing these problems I realized that there was another issue that
2650 hadn't been noticed. Values of message_size_limit (both the global option
2651 and the transport option) were treated as octal if they started with 0.
2652 The documentation was vague. These values are now always treated as
2653 decimal, and I will make that clear in the documentation.
2659 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2660 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2663 MH/01 Don't check that the operands of numeric comparison operators are
2664 integers when their expansion is in "skipping" mode (fixes bug
2665 introduced by 4.64-PH/07).
2667 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2668 child addresses, Exim now panics and dies. Previously, because the count
2669 is held in a short int, deliveries were likely to be lost. As such a
2670 large number of recipients for a single message is ridiculous
2671 (performance will be very, very poor), I have chosen to impose a limit
2672 rather than extend the field.
2678 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2679 leftover -K file (the existence of which was triggered by #402).
2680 While we were at it, introduced process PID as part of the -K
2681 filename. This should rule out race conditions when creating
2684 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2685 processing considerably. Previous code took too long for large mails,
2686 triggering a timeout which in turn triggers #401.
2688 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2689 in the DK code in transports.c. sendfile() is not really portable,
2690 hence the _LINUX specificness.
2692 TF/01 In the add_headers option to the mail command in an Exim filter,
2693 there was a bug that Exim would claim a syntax error in any
2694 header after the first one which had an odd number of characters
2697 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2698 callout verification, Exim cached a "reject" for the entire domain. This
2699 is correct for most verifications, but it is not correct for a recipient
2700 verification with use_sender or use_postmaster set, because in that case
2701 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2702 case of MAIL FROM:<> rejection from other early rejections (e.g.
2703 rejection of HELO). When verifying a recipient using a non-null MAIL
2704 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2705 Whatever the result of the callout, the value of the domain cache is
2706 left unchanged (for any other kind of callout, getting as far as trying
2707 RCPT means that the domain itself is ok).
2709 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2712 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2713 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2714 session, the daemon ignores ECONNECT errors and logs others; it now
2715 ignores EPIPE as well.
2717 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2718 (quoted-printable decoding).
2720 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2721 later the small subsequent patch to fix an introduced bug.
2723 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2725 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2727 PH/08 An error is now given if message_size_limit is specified negative.
2729 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2730 to be given (somewhat) arbitrary names.
2732 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2735 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2736 miscellaneous code fixes
2738 PH/10 Added the log_reject_target ACL modifier to specify where to log
2741 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2742 hostname. This is wrong, because it relates to the incoming message (and
2743 probably the interface on which it is arriving) and not to the outgoing
2744 callout (which could be using a different interface). This has been
2745 changed to use the value of the helo_data option from the smtp transport
2746 instead - this is what is used when a message is actually being sent. If
2747 there is no remote transport (possible with a router that sets up host
2748 addresses), $smtp_active_hostname is used.
2750 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2751 tweaks were necessary in order to get it to work (see also 21 below):
2752 (a) The code assumed that strncpy() returns a negative number on buffer
2753 overflow, which isn't the case. Replaced with Exim's string_format()
2755 (b) There were several signed/unsigned issues. I just did the minimum
2756 hacking in of casts. There is scope for a larger refactoring.
2757 (c) The code used strcasecmp() which is not a standard C function.
2758 Replaced with Exim's strcmpic() function.
2759 (d) The code set only $1; it now sets $auth1 as well.
2760 (e) A simple test gave the error "authentication client didn't specify
2761 service in request". It would seem that Dovecot has changed its
2762 interface. Fortunately there's a specification; I followed it and
2763 changed what the client sends and it appears to be working now.
2765 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2768 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2769 address A is aliased to B and C, where B exists and C does not. Without
2770 -v the output is "A verified" because verification stops after a
2771 successful redirection if more than one address is generated. However,
2772 with -v the child addresses are also verified. Exim was outputting "A
2773 failed to verify" and then showing the successful verification for C,
2774 with its parentage. It now outputs "B failed to verify", showing B's
2775 parentage before showing the successful verification of C.
2777 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2778 look up a TXT record in a specific list after matching in a combined
2781 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2782 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2783 they consult the DNS. I had assumed they would set it the way they
2784 wanted; and indeed my experiments on Linux seem to show that in some
2785 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2786 To be on the safe side, however, I have now made the interface to
2787 host_find_byname() similar to host_find_bydns(), with an argument
2788 containing the DNS resolver options. The host_find_byname() function now
2789 sets these options at its start, just as host_find_bydns() does. The smtp
2790 transport options dns_qualify_single and dns_search_parents are passed to
2791 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2792 of host_find_byname() use the default settings of RES_DEFNAMES
2793 (qualify_single) but not RES_DNSRCH (search_parents).
2795 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2796 spool_read_header() do less string testing, by means of a preliminary
2797 switch on the second character of optional "-foo" lines. (This is
2798 overdue, caused by the large number of possibilities that now exist.
2799 Originally there were few.) While I was there, I also converted the
2800 str(n)cmp tests so they don't re-test the leading "-" and the first
2801 character, in the hope this might squeeze out yet more improvement.
2803 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2804 flag allowing group syntax was set by the header_syntax check but not
2805 turned off, possible causing trouble later; (2) The flag was not being
2806 set at all for the header_verify test, causing "group"-style headers to
2807 be rejected. I have now set it in this case, and also caused header_
2808 verify to ignore an empty address taken from a group. While doing this, I
2809 came across some other cases where the code for allowing group syntax
2810 while scanning a header line wasn't quite right (mostly, not resetting
2811 the flag correctly in the right place). These bugs could have caused
2812 trouble for malformed header lines. I hope it is now all correct.
2814 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2815 with the "reply" argument non-NULL. The code, however (which originally
2816 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2817 but it didn't always do it. This confused somebody who was copying the
2818 code for some other use. I have removed all the tests.
2820 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2821 feature that was used to support insecure browsers during the U.S. crypto
2822 embargo. It requires special client support, and Exim is probably the
2823 only MTA that supported it -- and would never use it because real RSA is
2824 always available. This code has been removed, because it had the bad
2825 effect of slowing Exim down by computing (never used) parameters for the
2826 RSA_EXPORT functionality.
2828 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2829 authenticator to fail if there's a tab character in the incoming data
2830 (there should never be unless someone is messing about, as it's supposed
2831 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2832 if the connection is using TLS or if the remote IP is the same as the
2833 local IP, and the "valid-client-cert option" if a client certificate has
2836 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2837 authenticators. This can be used for authorization after authentication
2838 succeeds. (In the case of plaintext, it servers for both authentication
2841 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2842 if any retry times were supplied.
2844 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2845 connection, where there is no client IP address to check. In this
2846 situation, the verify now always succeeds.
2848 PH/25 Applied John Jetmore's -Mset patch.
2850 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2852 PH/27 In a string expansion for a processed (not raw) header when multiple
2853 headers of the same name were present, leading whitespace was being
2854 removed from all of them, but trailing whitespace was being removed only
2855 from the last one. Now trailing whitespace is removed from each header
2856 before concatenation. Completely empty headers in a concatenation (as
2857 before) are ignored.
2859 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2860 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2862 PH/29 [Removed. This was a change that I later backed out, and forgot to
2863 correct the ChangeLog entry (that I had efficiently created) before
2864 committing the later change.]
2866 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2867 address errors (4xx response to RCPT) over the same connection as other
2868 messages routed to the same hosts. Such deliveries are always "forced",
2869 so retry times are not inspected. This resulted in far too many retries
2870 for the affected addresses. The effect occurred only when there were more
2871 hosts than the hosts_max_try setting in the smtp transport when it had
2872 the 4xx errors. Those hosts that it had tried were not added to the list
2873 of hosts for which the message was waiting, so if all were tried, there
2874 was no problem. Two fixes have been applied:
2876 (i) If there are any address or message errors in an SMTP delivery, none
2877 of the hosts (tried or untried) are now added to the list of hosts
2878 for which the message is waiting, so the message should not be a
2879 candidate for sending over the same connection that was used for a
2880 successful delivery of some other message. This seems entirely
2881 reasonable: after all the message is NOT "waiting for some host".
2882 This is so "obvious" that I'm not sure why it wasn't done
2883 previously. Hope I haven't missed anything, but it can't do any
2884 harm, as the worst effect is to miss an optimization.
2886 (ii) If, despite (i), such a delivery is accidentally attempted, the
2887 routing retry time is respected, so at least it doesn't keep
2888 hammering the server.
2890 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2891 in ${readsocket because some servers need this prod.
2893 PH/32 Added some extra debug output when updating a wait-xxx database.
2895 PH/33 The hint "could be header name not terminated by colon", which has been
2896 given for certain expansion errors for a long time, was not being given
2897 for the ${if def:h_colon_omitted{... case.
2899 PH/34 The spec says: "With one important exception, whenever a domain list is
2900 being scanned, $domain contains the subject domain." There was at least
2901 one case where this was not true.
2903 PH/35 The error "getsockname() failed: connection reset by peer" was being
2904 written to the panic log as well as the main log, but it isn't really
2905 panic-worthy as it just means the connection died rather early on. I have
2906 removed the panic log writing for the ECONNRESET error when getsockname()
2909 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2910 runs only) independently of the message's sender address. This meant
2911 that, if the 4xx error was in fact related to the sender, a different
2912 message to the same recipient with a different sender could confuse
2913 things. In particular, this can happen when sending to a greylisting
2914 server, but other circumstances could also provoke similar problems.
2915 I have changed the default so that the retry time for these errors is now
2916 based a combination of the sender and recipient addresses. This change
2917 can be overridden by setting address_retry_include_sender=false in the
2920 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2921 remote server are returned as part of bounce messages. This was not
2922 happening for LMTP over a pipe (the lmtp transport), but now it is the
2923 same for both kinds of LMTP.
2925 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2926 in header lines that were in fact CNAMEs. This is no longer the case.
2928 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2929 and queue runs started by the daemon processed all messages. This has
2930 been fixed so that -R and -S can now usefully be given with -q<time>.
2932 PH/40 Import PCRE release 6.7 (fixes some bugs).
2934 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2936 PH/42 Give an error if -q is specified more than once.
2938 PH/43 Renamed the variables $interface_address and $interface_port as
2939 $received_ip_address and $received_port, to make it clear that these
2940 values apply to message reception, and not to the outgoing interface when
2941 a message is delivered. (The old names remain recognized, of course.)
2943 PH/44 There was no timeout on the connect() call when using a Unix domain
2944 socket in the ${readsocket expansion. There now is.
2946 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2947 be meaningful with "accept".
2949 SC/01 Eximstats V1.43
2950 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2952 SC/02 Eximstats V1.44
2953 Use a glob alias rather than an array ref in the generated
2954 parser. This improves both readability and performance.
2956 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2957 Collect SpamAssassin and rejection statistics.
2958 Don't display local sender or destination tables unless
2959 there is data to show.
2960 Added average volumes into the top table text output.
2962 SC/04 Eximstats V1.46
2963 Collect data on the number of addresses (recipients)
2964 as well as the number of messages.
2966 SC/05 Eximstats V1.47
2967 Added 'Message too big' to the list of mail rejection
2968 reasons (thanks to Marco Gaiarin).
2970 SC/06 Eximstats V1.48
2971 Mainlog lines which have GMT offsets and are too short to
2972 have a flag are now skipped.
2974 SC/07 Eximstats V1.49 (Alain Williams)
2975 Added the -emptyok flag.
2977 SC/08 Eximstats V1.50
2978 Fixes for obtaining the IP address from reject messages.
2980 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2981 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2982 whitespace changes from 4.64-PH/27
2984 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2987 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2988 are found, allow negative numbers in numeric criteria)
2990 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2992 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2993 to match changes made in 4.64-PH/43
2995 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2997 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2998 "domain" is actually an IP address, and force a failure. This locks out
2999 those revolvers/nameservers that support "A-for-A" lookups, in
3000 contravention of the specifications.
3002 PH/48 When a host name was looked up from an IP address, and the subsequent
3003 forward lookup of the name timed out, the host name was left in
3004 $sender_host_name, contrary to the specification.
3006 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3007 restricted to single-key lookups, Exim was not diagnosing an error if
3008 * or *@ was used with a query-style lookup.
3010 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3012 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3013 long ago, but noone interested enough thought of it. Let's just say that
3014 the "1.1" means that there are some new functions that weren't there at
3015 some point in the past.
3017 PH/51 Error processing for expansion failure of helo_data from an smtp
3018 transport during callout processing was broken.
3020 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3021 tested/used via the -bh/-bhc/-bs options.
3023 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3024 bug, fixed in subsequent PCRE releases).
3026 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3027 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3029 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3035 SC/01 Use a glob alias rather than an array ref in eximstats generated
3036 parser. This improves both readability and performance.
3038 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3039 Don't display local sender or destination tables in eximstats unless
3040 there is data to show.
3041 Added average volumes into the eximstats top table text output.
3043 SC/03 Collect data on the number of addresses (recipients) as well
3044 as the number of messages in eximstats.
3046 TF/01 Correct an error in the documentation for the redirect router. Exim
3047 does (usually) call initgroups() when daemonizing.
3049 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3050 with consistent privilege compared to when running as a daemon.
3052 TF/03 Note in the spec that $authenticated_id is not set for local
3053 submissions from trusted users.
3055 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3056 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3058 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3059 by adding some example configuration directives to the default
3060 configuration file. A little bit of work is required to uncomment the
3061 directives and define how usernames and passwords are checked, but
3062 there is now a framework to start from.
3064 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3065 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3066 without this. I don't know how relevant this is to other LDAP libraries.
3068 PH/02 Add the verb name to the "unknown ACL verb" error.
3070 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3072 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3074 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3075 directory not expanded when it should be if an expanded home directory
3076 was set for the address (which is overridden by the transport).
3078 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3081 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3082 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3083 because it is too late at that time, and has no effect.
3085 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3086 security issue with \' (bugzilla #107). I could not use the
3087 PQescapeStringConn() function, because it needs a PGconn value as one of
3090 PH/08 When testing addresses using -bt, indicate those final addresses that
3091 are duplicates that would not cause an additional delivery. At least one
3092 person was confused, thinking that -bt output corresponded to deliveries.
3093 (Suppressing duplicates isn't a good idea as you lose the information
3094 about possibly different redirections that led to the duplicates.)
3096 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3097 systems where poll() doesn't work, in particular OS X.
3099 PH/10 Added more information to debugging output for retry time not reached.
3101 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3102 operations in malware.c.
3104 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3107 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3108 syslog_duplication set false, log lines that would normally be written
3109 both the the main log and to the reject log were not written to syslog at
3112 PH/14 In the default configuration, change the use of "message" in ACL warn
3113 statements to "add_header".
3115 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3116 not followed by a command (e.g. "seen endif").
3118 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3119 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3122 PH/17 Added extra conditions to the default value of delay_warning_condition
3126 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3127 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3128 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3131 The Auto-Submitted: and various List- headers are standardised, whereas I
3132 don't think Precedence: ever was.
3134 PH/18 Refactored debugging code in route_finduser() to show more information,
3135 in particular, the error code if getpwnam() issues one.
3137 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3138 This is apparently needed in addition to the PH/07 change above to avoid
3139 any possible encoding problems.
3141 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3142 but not after initializing Perl.
3144 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3145 output them only if debugging. By default they are written stderr,
3146 apparently, which is not desirable.
3148 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3151 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3154 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3156 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3157 authenticated or an ident call has been made. Suppress the default
3158 values for $authenticated_id and $authenticated_sender (but permit -oMai
3159 and -oMas) when testing with -bh.
3161 PH/24 Re-jigged the order of the tests in the default configuration so that the
3162 tests for valid domains and recipients precede the DNS black list and CSA
3163 tests, on the grounds that those ones are more expensive.
3165 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3166 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3167 This bug exists in every version of Exim that I still have, right back to
3170 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3171 However, an attempt to turn on case-sensitivity in a regex key by
3172 including (?-i) didn't work because the subject string was already
3173 lowercased, and the effects were non-intuitive. It turns out that a
3174 one-line patch can be used to allow (?-i) to work as expected.
3180 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3181 other effects) broke the use of negated acl sub-conditions.
3183 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3186 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3187 "Deny" causes Exim to reject the incoming connection with a 554 error.
3188 Unfortunately, if there is a major crisis, such as a disk failure,
3189 tcp-wrappers gives "deny", whereas what one would like would be some
3190 kind of temporary error. A kludge has been added to help with this.
3191 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3192 554 error is used if errno is still zero or contains ENOENT (which occurs
3193 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3196 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3198 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3199 errors. Otherwise a message that provokes a temporary error (when other
3200 messages do not) can cause a whole host to time out.
3202 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3203 addresses were routed directly to files or pipes from a redirect router.
3204 File deliveries just didn't batch; pipe deliveries might have suffered
3207 PH/06 A failure to get a lock for a hints database would erroneously always say
3208 "Failed to get write lock", even when it was really a read lock.
3210 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3211 of 0600. This has been changed to use the value of the lockfile_mode
3212 option (which defaults to 0600).
3214 PH/08 Applied small patch from the Sieve maintainer.
3216 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3217 folder from quota calculations, a direct delivery into this folder messed
3218 up the contents of the maildirsize file. This was because the regex was
3219 used only to exclude .Trash (or whatever) when the size of the mailbox
3220 was calculated. There was no check that a delivery was happening into an
3221 excluded directory. This bug has been fixed by ignoring all quota
3222 processing for deliveries into excluded directories.
3224 PH/10 Added the maildirfolder_create_regex option to appendfile.
3230 PH/01 The code for finding all the local interface addresses on a FreeBSD
3231 system running IPv6 was broken. This may well have applied to all BSD
3232 systems, as well as to others that have similar system calls. The broken
3233 code found IPv4 interfaces correctly, but gave incorrect values for the
3234 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3235 that it would not match correctly against @[] and not recognize the IPv6
3238 PH/02 The ipliteral router was not recognizing addresses of the form user@
3239 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3241 PH/03 Added disable_ipv6.
3243 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3244 decoded form, because it is most often used to construct To: headers
3245 lines in autoreplies, and the decoded form may well be syntactically
3246 invalid. However, $reply_address has leading white space removed, and all
3247 newlines turned into spaces so that the autoreply transport does not
3250 PH/05 If group was specified without a user on a router, and no group or user
3251 was specified on a transport, the group from the router was ignored.
3253 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3254 for visible compile-time settings that can be used to change these
3255 numbers, for those that want even more. Backwards compatibility with old
3256 spool files has been maintained. However, going back to a previous Exim
3257 release will lost any variables that are in spool files.
3259 PH/07 Two small changes when running in the test harness: increase delay when
3260 passing a TCP/IP connection to a new process, in case the original
3261 process has to generate a bounce, and remove special handling of
3262 127.0.0.2 (sic), which is no longer necessary.
3264 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3265 be the same on different OS.
3267 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3270 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3271 whether --show-vars was specified or not
3273 JJ/02 exipick: Added support for new ACL variable spool format introduced
3276 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3277 syntactically invalid From: or Reply-to: line, and a filter used this to
3278 generate an autoreply, and therefore failed to obtain an address for the
3279 autoreply, Exim could try to deliver to a non-existent relative file
3280 name, causing unrelated and misleading errors. What now happens is that
3281 it logs this as a hard delivery error, but does not attempt to create a
3284 PH/11 The exinext utility has a -C option for testing purposes, but although
3285 the given file was scanned by exinext itself; it wasn't being passed on
3286 when Exim was called.
3288 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3289 an end-of-file indication when reading a command response.
3291 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3292 compiled. In many other places in Exim, IPv6 addresses are always
3293 recognized, so I have changed this. It also means that IPv4 domain
3294 literals of the form [IPV4:n.n.n.n] are now always recognized.
3296 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3297 used if the router is not running as root, for example, when verifying at
3298 ACL time, or when using -bh. The debugging output from this situation was
3299 non-existent - all you got was a failure to exec. I have made two
3302 (a) Failures to set uid/gid, the current directory, or a process leader
3303 in a subprocess such as that created by queryprogram now generate
3304 suitable debugging output when -d is set.
3306 (b) The queryprogram router detects when it is not running as root,
3307 outputs suitable debugging information if -d is set, and then runs
3308 the subprocess without attempting to change uid/gid.
3310 PH/15 Minor change to Makefile for building test_host (undocumented testing
3313 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3314 additional section of a DNS packet that returns MX or SRV records.
3315 Instead, it always explicitly searches for A/AAAA records. This avoids
3316 major problems that occur when a DNS server includes only records of one
3317 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3318 fixed another bug: if SRV records were looked up and the corresponding
3319 address records were *not* found in the additional section, the port
3320 values from the SRV records were lost.
3322 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3323 using the correct key (the original address) when searching the retry
3324 rules in order to find which one to use for generating the retry hint.
3326 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3327 adding the default one. Similarly, if it contains a Reply-To: header, the
3328 errors_reply_to option, if set, is not used.
3330 PH/19 When calculating a retry time, Exim used to measure the "time since
3331 failure" by looking at the "first failed" field in the retry record. Now
3332 it does not use this if it is later than than the arrival time of the
3333 message. Instead it uses the arrival time. This makes for better
3334 behaviour in cases where some deliveries succeed, thus re-setting the
3335 "first failed" field. An example is a quota failure for a huge message
3336 when small messages continue to be delivered. Without this change, the
3337 "time since failure" will always be short, possible causing more frequent
3338 delivery attempts for the huge message than are intended.
3339 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3341 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3342 $1, $2, $3) because the numerical variables can be reset during some
3343 expansion items (e.g. "match"), thereby losing the authentication data.
3345 PH/21 Make -bV show the size of off_t variables so that the test suite can
3346 decide whether to run tests for quotas > 2G.
3348 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3349 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3350 filecount value is greater than 2G or if a quota value is greater than 2G
3351 on a system where the size of off_t is not greater than 4, a panic error
3354 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3355 never match. The debug and -bh output now contains an explicit error
3356 message indicating a malformed IPv4 address or mask.
3358 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3359 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3360 PH/23 above applies.
3362 PH/25 Do not write to syslog when running in the test harness. The only
3363 occasion when this arises is a failure to open the main or panic logs
3364 (for which there is an explicit test).
3366 PH/26 Added the /no_tell option to "control=freeze".
3368 PH/27 If a host name lookup failed very early in a connection, for example, if
3369 the IP address matched host_lookup and the reverse lookup yielded a name
3370 that did not have a forward lookup, an error message of the form "no IP
3371 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3372 could be logged. Now it outputs the IP address instead of "NULL".
3374 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3375 allows the sender and the authenticated sender to be set when
3376 submitting a message from within Exim. Since child_open_exim() is
3377 documented for local_scan(), the new function should be too.
3379 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3380 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3381 results in an empty string is now treated as unset.
3383 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3385 PH/31 Added sender_verify_fail logging option.
3387 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3388 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3389 tidied the source and removed it altogether.
3391 PH/33 When a queue run was abandoned because the load average was too high, a
3392 log line was always written; now it is written only if the queue_run log
3393 selector is set. In addition, the log line for abandonment now contains
3394 information about the queue run such as the pid. This is always present
3395 in "start" and "stop" lines but was omitted from the "abandon" line.
3397 PH/34 Omit spaces between a header name and the colon in the error message that
3398 is given when verify = headers_syntax fails (if there are lots of them,
3399 the message gets confusing).
3401 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3402 names, as there are now some PTR records that contain slashes. This check
3403 is only to protect against broken name servers that fall over on strange
3404 characters, so the fact that it applies to all lookups doesn't matter.
3406 PH/36 Now that the new test suite is complete, we can remove some of the
3407 special code in Exim that was needed for the old test suite. For example,
3408 sorting DNS records because real resolvers return them in an arbitrary
3409 order. The new test suite's fake resolver always returns records in the
3412 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3413 bounces) except when queue_only is set, to avoid logging races between
3414 the different processes.
3416 PH/38 Panic-die if .include specifies a non-absolute path.
3418 PH/39 A tweak to the "H" retry rule from its user.
3420 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3421 a label. They prevented compilation on older perls.
3423 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3424 a warning to be raised on newish perls.
3426 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3427 on queue. Changes to match documented behaviour of showing count of
3428 messages matching specified criteria.
3430 PH/40 Changed the default ident timeout from 30s to 5s.
3432 PH/41 Added support for the use of login_cap features, on those BSD systems
3433 that have them, for controlling the resources used by pipe deliveries.
3435 PH/42 The content-scanning code uses fopen() to create files in which to put
3436 message data. Previously it was not paying any attention to the mode of
3437 the files. Exim runs with umask(0) because the rest of the code creates
3438 files with open(), and sets the required mode explicitly. Thus, these
3439 files were ending up world-writeable. This was not a big issue, because,
3440 being within the spool directory, they were not world-accessible. I have
3441 created a function called modefopen, which takes an additional mode
3442 argument. It sets umask(777), creates the file, chmods it to the required
3443 mode, then resets the umask. All the relevant calls to fopen() in the
3444 content scanning code have been changed to use this function.
3446 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3447 to 24 hours. This avoids potential overflow problems when processing G
3448 and H retry rules. I suspect nobody ever tinkers with this value.
3450 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3452 PH/45 When the plaintext authenticator is running as a client, the server's
3453 challenges are checked to ensure they are valid base64 strings. By
3454 default, the authentication attempt is cancelled if an invalid string is
3455 received. Setting client_ignore_invalid_base64 true ignores these errors.
3456 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3457 they are received. Thus, the responses can be made to depend on the
3458 challenges. If an invalid string is ignored, an empty string is placed in
3461 PH/46 Messages that are created by the autoreply transport now contains a
3462 References: header, in accordance with RFCs 2822 and 3834.
3464 PH/47 Added authenticated_sender_force to the smtp transport.
3466 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3468 PH/49 Installed latest patch from the Sieve maintainer.
3470 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3471 was also set, the appendfile transport was unnecessarily scanning a
3472 directory of message files (e.g. for maildir delivery) to find the count
3473 of files (along with the size), even though it did not need this
3474 information. It now does the scan only if it needs to find either the
3475 size of the count of files.
3477 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3479 PH/52 Two bugs concerned with error handling when the smtp transport is
3482 (i) Exim was not creating retry information for temporary errors given
3483 for individual recipients after the DATA command when the smtp transport
3484 was used in LMTP mode. This meant that they could be retried too
3485 frequently, and not timed out correctly.
3487 (ii) Exim was setting the flag that allows error details to be returned
3488 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3489 recipients that were returned after the DATA command.
3491 PH/53 This is related to PH/52, but is more general: for any failing address,
3492 when detailed error information was permitted to be returned to the
3493 sender, but the error was temporary, then after the final timeout, only
3494 "retry timeout exceeded" was returned. Now it returns the full error as
3495 well as "retry timeout exceeded".
3497 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3498 do this, and (what is worse) MTAs that accept it.
3500 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3501 will now be deprecated.
3503 PH/56 New os.c-cygwin from the Cygwin maintainer.
3505 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3506 formats (previously only available in exim formats via -bpr, -bpru,
3507 and -bpra. Now also available in native and exiqgrep formats)
3509 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3510 with very large, slow to parse queues
3512 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3514 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3516 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3517 responses to authentication challenges, though it was showing the
3518 challenges; (ii) I've removed the CR characters from the debug output for
3521 PH/58 Allow for the insertion of a newline as well as a space when a string
3522 is turned into more than one encoded-word during RFC 2047 encoding. The
3523 Sieve code now uses this.
3525 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3526 data_4xx, lost_connection, tls_required.
3528 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3529 message was being sent as an SMTP response.
3531 PH/61 Add -l and -k options to exicyclog.
3533 PH/62 When verifying, if an address was redirected to one new address, so that
3534 verification continued, and the new address failed or deferred after
3535 having set something in $address_data, the value of $address_data was not
3536 passed back to the ACL. This was different to the case when no
3537 redirection occurred. The value is now passed back in both cases.
3539 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3540 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3541 use login_cap.h, so on its own it isn't the distinguishing feature. The
3542 new name refers directly to the setclassresources() function.
3544 PH/65 Added configuration files for NetBSD3.
3546 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3548 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3549 is preferred over IPv4.
3551 PH/68 The bounce_return_message and bounce_return_body options were not being
3552 honoured for bounces generated during the reception of non-SMTP messages.
3553 In particular, this applied to messages rejected by the ACL. This bug has
3554 been fixed. However, if bounce_return_message is true and bounce_return_
3555 body is false, the headers that are returned for a non-SMTP message
3556 include only those that have been read before the error was detected.
3557 (In the case of an ACL rejection, they have all been read.)
3559 PH/69 The HTML version of the specification is now built in a directory called
3560 spec_html instead of spec.html, because the latter looks like a path with
3561 a MIME-type, and this confuses some software.
3563 PH/70 Catch two compiler warnings in sieve.c.
3565 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3566 function verify_get_ident() calls ip_connect() to connect a socket, but
3567 if the "connect()" function timed out, ip_connect() used to close the
3568 socket. However, verify_get_ident() also closes the socket later, and in
3569 between Exim writes to the log, which may get opened at this point. When
3570 the socket was closed in ip_connect(), the log could get the same file
3571 descriptor number as the socket. This naturally causes chaos. The fix is
3572 not to close the socket in ip_connect(); the socket should be closed by
3573 the function that creates it. There was only one place in the code where
3574 this was missing, in the iplookup router, which I don't think anybody now
3575 uses, but I've fixed it anyway.
3577 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3578 well as to direct DNS lookups. Otherwise the handling of names in host
3579 lists is inconsistent and therefore confusing.
3585 PH/01 Two changes to the default runtime configuration:
3587 (1) Move the checks for relay_from_hosts and authenticated clients from
3588 after to before the (commented out) DNS black list checks.
3590 (2) Add control=submission to the relay_from_hosts and authenticated
3591 clients checks, on the grounds that messages accepted by these
3592 statements are most likely to be submissions.
3594 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3596 (1) Generate an error if the third argument for the ${prvs expansion is
3599 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3602 (3) Reset the variables that are obtained from the first argument of
3603 ${prvscheck and used in the second argument before leaving the code,
3604 because their memory is reclaimed, so using them afterwards may do
3607 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3608 one (it's much easier than Tom thought :-).
3610 (5) Because of (4), we can now allow for the use of $prvscheck_result
3611 inside the third argument.
3613 PH/03 For some reason, the default setting of PATH when running a command from
3614 a pipe transport was just "/usr/bin". I have changed it to
3617 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3618 anything to be listed in the output from -bV.
3620 PH/05 When a filter generated an autoreply, the entire To: header line was
3621 quoted in the delivery log line, like this:
3623 => >A.N.Other <ano@some.domain> <original@ddress> ...
3625 This has been changed so that it extracts the operative address. There
3626 may be more than one such address. If so, they are comma-separated, like
3629 => >ano@some.domain,ona@other.domain <original@ddress> ...
3631 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3632 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3633 not being looked up in the rDNS to get a host name, Exim was showing the
3634 IP address twice in Received: lines, even though the IP addresses were
3635 identical. For example:
3637 Received: from [1.2.3.4] (helo=[1.2.3.4])
3639 However, if the real host name was known, it was omitting the HELO data
3640 if it matched the actual IP address. This has been tidied up so that it
3641 doesn't show the same IP address twice.
3643 PH/07 When both +timestamp and +memory debugging was on, the value given by
3644 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3645 called by the debug printing, thereby overwriting the timestamp buffer.
3646 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3648 PH/08 When the original message was included in an autoreply transport, it
3649 always said "this is a copy of the message, including all the headers",
3650 even if body_only or headers_only was set. It now gives an appropriate
3653 PH/09 Applied a patch from the Sieve maintainer which:
3655 o fixes some comments
3656 o adds the (disabled) notify extension core
3657 o adds some debug output for the result of if/elsif tests
3658 o points to the current vacation draft in the documentation
3659 and documents the missing references header update
3663 o fixes a bug in processing the envelope test (when testing
3664 multiple envelope elements, the last element determined the
3667 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3668 Electronic Mail") by including:
3670 Auto-submitted: auto-generated
3672 in the messages that it generates (bounce messages and others, such as
3673 warnings). In the case of bounce messages for non-SMTP messages, there was
3674 also a typo: it was using "Auto_submitted" (underscore instead of
3675 hyphen). Since every message generated by Exim is necessarily in response
3676 to another message, thes have all been changed to:
3678 Auto-Submitted: auto-replied
3680 in accordance with these statements in the RFC:
3682 The auto-replied keyword:
3684 - SHOULD be used on messages sent in direct response to another
3685 message by an automatic process,
3687 - MUST NOT be used on manually-generated messages,
3689 - MAY be used on Delivery Status Notifications (DSNs) and Message
3690 Disposition Notifications (MDNs),
3692 - MUST NOT be used on messages generated by automatic or periodic
3693 processes, except for messages which are automatic responses to
3696 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3697 to the default Received: header definition.
3699 PH/12 Added log selector acl_warn_skipped (default on).
3701 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3702 variables because (a) they are in the wrong storage pool and (b) even if
3703 they were copied, it wouldn't work properly because of the caching.
3705 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3706 checking when decoding. Apparently there are clients that generate
3707 overlong encoded strings. Why am I not surprised?
3709 PH/15 If the first argument of "${if match_address" was not empty, but did not
3710 contain an "@" character, Exim crashed. Now it writes a panic log message
3711 and treats the condition as false.
3713 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3715 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3716 extension "envelope-auth". The code is finished and in agreement with
3717 other implementations, but there is no documentation so far and in fact,
3718 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3719 not changing the active code.
3721 Print executed "if" and "elsif" statements when debugging is used. This
3722 helps a great deal to understand what a filter does.
3724 Document more things not specified clearly in RFC3028. I had all this
3725 sorted out, when out of a sudden new issues came to my mind. Oops."
3727 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3730 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3731 canonical form (as documented). However, after a host name lookup from
3732 the IP address, check_host() was doing a simple string comparison with
3733 addresses acquired from the DNS when checking that the found name did
3734 have the original IP as one of its addresses. Since any found IPv6
3735 addresses are likely to be in abbreviated form, the comparison could
3736 fail. Luckily, there already exists a function for doing the comparison
3737 by converting both addresses to binary, so now that is used instead of
3738 the text comparison.
3740 PH/20 There was another similar case to PH/19, when a complete host name was
3741 given in a host list; looking up its IP address could give an abbreviated
3742 form, whereas the current host's name might or might not be abbreviated.
3743 The same fix has been applied.
3749 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3750 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3753 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3754 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3756 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3758 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3759 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3760 Also, for "Auto-Submitted", treat anything other than "no" as
3761 non-personal, in accordance with RFC 3834. (Previously it treated
3762 anything starting "auto-" as non-personal.)
3764 TF/01 The control=submission/name=... option had a problem with syntax
3765 errors if the name included a slash character. The /name= option
3766 now slurps the rest of the string, so it can include any characters
3767 but it must come last in the list of options (after /sender_retain
3770 PH/05 Some modifications to the interface to the fake nameserver for the new
3778 TK/01 Added the "success_on_redirect" address verification option. See
3779 NewStuff for rationale and an example.
3781 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3783 PH/02 Patch to exigrep to allow it to work on syslog lines.
3785 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3786 fread() to skip over the body file's header line, because in Cygwin the
3787 header line is locked and is inaccessible.
3789 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3790 co-exist for some time) to make it clear that it is the Exim ID that is
3791 referenced, not the Message-ID: header line.
3793 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3794 string_format() function, because snprintf() does not exist on all
3797 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3798 ${stat: expansion item.
3800 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3801 protocol synchronization error", to keep the pedants happy.
3803 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3804 well as for IRIX systems, when gcc is being used. See the host.c source
3807 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3809 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3812 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3813 transport and to the smtp transport in LMTP mode.
3815 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3817 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3818 run a filter in a subprocess. This could lead to confusion in subsequent
3819 lookups in the parent process. There should also be a search_tidyup() at
3820 the end of the subprocess.
3822 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3823 only if the host matched helo_try_verify_hosts, which caused the
3824 verification to occur when the EHLO/HELO command was issued. The ACL just
3825 tested the remembered result. Now, if a previous verification attempt has
3826 not happened, "verify = helo" does it there and then.
3828 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3830 TK/03 Fix log output including CR from clamd.
3832 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3833 exist provoked a memory error which could cause a segfault.
3835 PH/15 Installed PCRE 6.2
3837 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3839 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3840 of the problem. Specifically, suggested +O2 rather than +O1 for the
3843 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3845 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3846 record, the port was not correctly logged when the outgoing_port log
3847 selector was set (it logged the transort's default port).
3849 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3850 fallback_hosts, and "hosts" in the smtp transport.
3852 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3853 host errors such as "Connection refused".
3855 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3856 authentication with radiusclient 0.4.9:
3858 - Error returned from rc_read_config was caught wrongly
3859 - Username/password not passed on to radius server due to wrong length.
3861 The presumption is that some radiusclient API changes for 4.51/PH/17
3862 were not taken care of correctly. The code is still untested by me (my
3863 Linux distribution still has 0.3.2 of radiusclient), but it was
3864 contributed by a Radius user.
3866 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3867 expanding the "port" option of the smtp transport.
3869 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3870 while reading a MIME header. Thanks to Tom Hughes for a patch.
3872 PH/24 Include config.h inside local_scan.h so that configuration settings are
3875 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3876 that in an ACL for RCPT (for example), you can examine exactly what was
3879 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3880 commands, but it was not correctly comparing the address with the actual
3881 client host address. Thus, it would show the EHLO address in Received:
3882 header lines when this was not necessary.
3884 PH/27 Added the % operator to ${eval:}.
3886 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3887 it should be ignoring failures (because with -C, for example, it has lost
3888 privilege). It wasn't ignoring creation failures other than "already
3891 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3894 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3895 because an input line was too long, either on its own, or by virtue of
3896 too many continuations, the temporary file was not being removed, and the
3897 return code was incorrect.
3899 PH/31 Missing "BOOL" in function definition in filtertest.c.
3901 PH/32 Applied Sieve patches from the maintainer.
3903 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3905 PH/33 Added "verify = not_blind".
3907 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3908 Local/Makefile (with some defaults set). These are used in built scripts
3909 such as exicyclog, but they have never been used in the exim_install
3910 script (though there are many overriding facilities there). I have
3911 arranged that the exim_install script now takes note of these two
3914 PH/35 Installed configuration files for Dragonfly.
3916 PH/36 When a locally submitted message by a trusted user did not contain a
3917 From: header, and the sender address was obtained from -f or from an SMTP
3918 MAIL command, and the trusted user did not use -F to supply a sender
3919 name, $originator_name was incorrectly used when constructing a From:
3920 header. Furthermore, $originator_name was used for submission mode
3921 messages from external hosts without From: headers in a similar way,
3922 which is clearly wrong.
3924 PH/37 Added control=suppress_local_fixups.
3926 PH/38 When log_selector = +received_sender was set, and the addition of the
3927 sender made the log line's construction buffer exactly full, or one byte
3928 less than full, an overflow happened when the terminating "\n" was
3931 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3932 when the result of a list match is failure because a DNS lookup failed.
3934 PH/40 RM_COMMAND is now used in the building process.
3936 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3937 the "build-* directories that it finds.
3939 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3940 address in a domain literal was a prefix of an interface address.
3942 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3943 when verifying a sender address, unless rewrite_headers is false.
3945 PH/44 Wrote a long comment about why errors_to addresses are verified as
3946 recipients, not senders.
3948 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3949 the ratelimit ACL was added.
3951 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3953 PH/46 Added extra information about PostgreSQL errors to the error string.
3955 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3956 avoiding the need to install special zones in a real server. This is
3957 backwards compatible; if it can't find the fake resolver, it drops back.
3958 Thus, both old and new test suites can be run.
3960 TF/02 Added util/ratelimit.pl
3962 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3963 clock is set back in time.
3965 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3966 Candler <B.Candler@pobox.com>.
3968 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3969 OK for addresses that are the result of redirections.
3971 PH/48 A number of further additions for the benefit of the new test suite,
3972 including a fake gethostbyname() that interfaces to the fake DNS resolver
3975 TF/06 The fix for widen_domains has also been applied to qualify_single and
3976 search_parents which are the other dnslookup options that can cause
3979 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3982 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3984 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3985 service" if it encountered an MX record with an empty target hostname.
3986 The message is now "an MX or SRV record indicated no SMTP service".
3988 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3989 if the DNS of the sending site is misconfigured. This is quite a
3990 common situation. This change restores the behaviour of treating a
3991 helo verification defer as a failure.
3993 PH/51 If self=fail was set on a router, the bounce message did not include the
3994 actual error message.
4000 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
4002 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
4003 command itself ended in error, the underlying message about the transport
4004 filter timeout was being overwritten with the pipe command error. Now the
4005 underlying error message should be appended to the second error message.
4007 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4009 PH/02 Reduce the amount of output that "make" produces by default. Full output
4010 can still be requested.
4012 PH/03 The warning log line about a condition test deferring for a "warn" verb
4013 was being output only once per connection, rather than after each
4014 occurrence (because it was using the same function as for successful
4015 "warn" verbs). This seems wrong, so I have changed it.
4017 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4018 it should not have, which might have caused a crash in the right
4019 circumstances, but probably never did.
4021 PH/04 Installed a modified version of Tony Finch's patch to make submission
4022 mode fix the return path as well as the Sender: header line, and to
4023 add a /name= option so that you can make the user's friendly name appear
4026 TF/03 Added the control = fakedefer ACL modifier.
4028 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4029 Mark Lowes for thorough testing.
4031 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4033 TK/03 Merged latest SRS patch from Miles Wilton.
4035 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4036 with the definition in sysexits.h (which is #included earlier).
4037 Fortunately, Exim does not actually use EX_OK. The code used to try to
4038 preserve the sysexits.h value, by assuming that macro definitions were
4039 scanned for macro replacements. I have been disabused of this notion,
4040 so now the code just undefines EX_OK before #including unistd.h.
4042 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4043 in the smtp transport. When a block could not be written in a single
4044 write() function, the timeout was being re-applied to each part-write.
4045 This seems wrong - if the receiver was accepting one byte at a time it
4046 would take for ever. The timeout is now adjusted when this happens. It
4047 doesn't have to be particularly precise.
4049 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4050 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4052 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4053 without a domain if the check to <postmaster@domain> fails.
4055 SC/01 Eximstats: added -xls and the ability to specify output files
4056 (patch written by Frank Heydlauf).
4058 SC/02 Eximstats: use FileHandles for outputting results.
4060 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4062 SC/04 Eximstats: fixed display of large numbers with -nvr option
4064 SC/05 Eximstats: fixed merging of reports with empty tables.
4066 SC/06 Eximstats: added the -include_original_destination flag
4068 SC/07 Eximstats: removed tabs and trailing whitespace.
4070 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4072 TK/06 MBOX spool code: Add real "From " MBOX separator line
4073 so the .eml file is really in mbox format (even though
4074 most programs do not really care). Patch from Alex Miller.
4076 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4077 The latter is generated from $received_to and is only set if the
4078 message has one envelope recipient. SA can use these headers,
4079 obviously out-of-the-box. Patch from Alex Miller.
4081 PH/08 The ${def test on a variable was returning false if the variable's
4082 value was "0", contrary to what the specification has always said!
4083 The result should be true unless the variable is empty.
4085 PH/09 The syntax error of a character other than { following "${if
4086 def:variable_name" (after optional whitespace) was not being diagnosed.
4087 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4088 accidental colon was present, for example, could give incorrect results.
4090 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4091 result is used (not including appendfile, where other changes are about
4094 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4095 This involved changing a lot of size variables from int to off_t. It
4096 should work with maildirs and everything.
4098 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4099 spamd dying while we are connected to it.
4101 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4104 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4105 function for BDB 4.3.
4107 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4109 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4110 the sources for PCRE have been re-arranged and more files are now
4113 PH/15 The code I had for printing potentially long long variables in PH/11
4114 above was not the best (it lost precision). The length of off_t variables
4115 is now inspected at build time, and an appropriate printing format (%ld
4116 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4117 to be "long long int" or "long int". This is needed for the internal
4118 formatting function string_vformat().
4120 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4121 the configuration file to be ":syslog", then the script "guesses" where
4122 the logs files are, rather than using the compiled in default. In our
4123 case the guess is not the same as the compiled default, so the script
4124 suddenly stopped working when I started to use syslog. The patch checks
4125 to see if log_file_path is "". If so, it attempts to read it from exim
4126 with no configuration file to get the compiled in version, before it
4127 falls back to the previous guessing code."
4129 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4130 implementing BATV in an Exim configuration. See NewStuff for the gory
4133 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4134 Makefile that are specific to HP-UX.
4136 PH/18 If the "use_postmaster" option was set for a recipient callout together
4137 with the "random" option, the postmaster address was used as the MAIL
4138 FROM address for the random test, but not for the subsequent recipient
4139 test. It is now used for both.
4141 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4142 patch removes a few documentation additions to RFC 3028, because the
4143 latest draft now contains them. It adds the new en;ascii-case comparator
4144 and a new error check for 8bit text in MIME parts. Comparator and
4145 require names are now matched exactly. I enabled the subaddress
4146 extension, but it is not well tested yet (read: it works for me)."
4148 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4149 rework some of the code of TK/09 above to avoid the hardwired use of
4150 "%lld" and "long long". Replaced the call to snprintf() with a call to
4153 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4154 records point to non-existent hosts", "retry timeout exceeded", and
4155 "retry time not reached for any host after a long failure period".
4157 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4158 experimental DomainKeys support:
4160 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4161 (2) On an error such as an illegally used "control", the wrong name for
4162 the control was given.
4164 These problems did NOT occur unless DomainKeys support was compiled.
4166 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4168 PH/24 Added ${if match_ip condition.
4170 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4171 databases so that it will be absolutely obvious if a crash occurs in the
4172 DB library. This is a regular occurrence (often caused by mis-matched
4175 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4176 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4177 picked up on a user's system that detects such things. There doesn't seem
4178 to be a gcc warning option for this - only an attribute that has to be
4179 put on the function's prototype. It seems that in Fedora Core 4 they have
4180 set this on a number of new functions. No doubt there will be more in due
4183 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4184 specify a transport. However, if an address that was verified by such a
4185 router was the subject of a callout, Exim crashed because it tried to
4186 read the rcpt_include_affixes from the non-existent transport. Now it
4187 just assumes that the setting of that option is false. This bug was
4188 introduced by 4.51/PH/31.
4190 PH/28 Changed -d+all to exclude +memory, because that information is very
4191 rarely of interest, but it makes the output a lot bigger. People tend to
4192 do -d+all out of habit.
4194 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4195 code in os-type was giving problems when libc.so lives in lib64, like on
4198 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4199 aren't the modern standard, and it seems that some systems' include files
4200 don't always have them. Exim was already checking for some of the newer
4201 ones like T_AAAA, and defining it itself. I've added checks for all the
4202 record types that Exim uses.
4204 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4205 not automatically generating a new one, as it is supposed to. This
4206 prevented TLS from working. If the file did exist, but contained invalid
4207 data, a new version was generated, as expected. It was only the case of a
4208 non-existent file that was broken.
4210 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4211 with a change in libdomainkeys > 0.64.
4213 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4214 from DNS. If the selector record carries the flag, it now has
4215 precedence over the domain-wide flag.
4217 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4219 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4220 the use of a port name that isn't defined in /etc/services) occurred, the
4221 message was deferred as in a normal delivery, and thus remained on the
4222 spool, instead of being failed because of the mua_wrapper setting. This
4223 is now fixed, and I tidied up some of the mua_wrapper messages at the
4226 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4227 the messages in a hash of arrays rather than using individual hashes.
4228 This is a bit cleaner and results in dramatic memory savings, albeit
4229 at a slight CPU cost.
4231 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4232 as requested by Marc Sherman.
4234 SC/10 Eximstats: added histograms for user specified patterns as requested
4237 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4239 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4240 fopen() in the content-scanning modules that did not already have it.
4246 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4247 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4249 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4251 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4253 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4254 to test Sieve filters that use "vacation".
4256 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4257 that changes the way the GnuTLS parameters are stored in the cache file.
4258 The new format can be generated externally. For backward compatibility,
4259 if the data in the cache doesn't make sense, Exim assumes it has read an
4260 old-format file, and it generates new data and writes a new file. This
4261 means that you can't go back to an older release without removing the
4264 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4265 work if there are any delivery delays because "one_time" forces the
4266 parent to be marked "delivered", so its unseen clone is never tried
4267 again. For this reason, Exim now forbids the simultaneous setting of
4270 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4271 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4272 entry if you want to know the details. The fix, however, neglected to
4273 consider the case when local delivery batching is involved. The test for
4274 "previously delivered" was not happening when checking to see if an
4275 address could be batched with a previous (undelivered) one; under
4276 certain circumstances this could lead to multiple deliveries to the same
4279 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4280 in its include files, and this causes problems building Exim.
4282 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4283 header_syntax) but Exim was just ignoring anything given after a slash.
4284 In particular, this caused confusion with an attempt to use "verify =
4285 reverse_host_lookup/defer_ok". An error is now given when options are
4286 supplied for verify items that do not have them. (Maybe reverse_host_
4287 lookup should have a defer_ok option, but that's a different point.)
4289 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4290 defined by RFC 821) to 2048, because there were problems with some AUTH
4291 commands, and RFC 1869 says the size should be increased for extended
4292 SMTP commands that take arguments.
4294 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4297 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4298 "unknown" error; now it says that the functionality isn't in the binary.
4300 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4301 an address' error message when a string expansion fails (syntax or
4302 whatever). Otherwise the password may appear in the log. Following change
4303 PH/42 below, there is no longer a chance of it appearing in a bounce
4306 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4308 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4309 of its addresses was ever tried. (Bugzilla bug #2.)
4311 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4312 the result incorrectly in the debug output. (It correctly added a newline
4313 to what was transported.)
4315 TF/01 Added $received_time.
4317 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4318 commented out examples of how to interface to a virus scanner and to
4319 SpamAssassin. Also added commented examples of av_scanner and
4320 spamd_address settings.
4322 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4323 and controls are allowed in which ACLs. There were a couple of minor
4324 errors. Some of the entries in the conditions table (which is a table of
4325 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4326 negation of where the condition IS allowed.
4328 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4330 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4331 header file does not have a version number, so I've had to invent a new
4332 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4333 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4334 radiusclient), but it was contributed by a Radius user.
4336 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4337 files or directories, for OpenSSL.
4339 PH/19 When an Exim process that is running as root has to create an Exim log
4340 file, it does so in a subprocess that runs as exim:exim so as to get the
4341 ownership right at creation (otherwise, other Exim processes might see
4342 the file with the wrong ownership). There was no test for failure of this
4343 fork() call, which would lead to the process getting stuck as it waited
4344 for a non-existent subprocess. Forks do occasionally fail when resources
4345 run out. I reviewed all the other calls to fork(); they all seem to check
4348 PH/20 When checking for unexpected SMTP input at connect time (before writing
4349 the banner), Exim was not dealing correctly with a non-positive return
4350 from the read() function. If the client had disconnected by this time,
4351 the result was a log entry for a synchronization error with an empty
4352 string after "input=" when read() returned zero. If read() returned -1
4353 (an event I could not check), uninitialized data bytes were printed.
4354 There were reports of junk text (parts of files, etc) appearing after
4357 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4359 PH/22 Added support for macro redefinition, and (re)definition in between
4360 driver and ACL definitions.
4362 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4363 forgetting to use the resulting value; it was using the unexpanded value.
4365 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4366 hadn't been configured. The fix is from Juergen Kreileder, who
4367 understands it better than I do:
4369 "Here's what I see happening with three configured cyrus_sasl
4370 authenticators configured (plain, login, cram-md5):
4372 On startup auth_cyrus_sasl_init() gets called for each of these.
4373 This means three calls to sasl_listmech() without a specified mech_list.
4374 => SASL tests which mechs of all available mechs actually work
4375 => three warnings about OTP not working
4376 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4378 With the patch, sasl_listmech() also gets called three times. But now
4379 SASL's mech_list option is set to the server_mech specified in the the
4380 authenticator. Or in other words, the answer from sasl_listmech()
4381 gets limited to just the mech you're testing for (which is different
4383 => the return list contains just 'plain' or 'login', 'cram-md5' or
4384 nothing depending on the value of ob->server_mech.
4386 I've just tested the patch: Authentication still works fine,
4387 unavailable mechs specified in the exim configuration are still
4388 caught, and the auth.log warnings about OTP are gone."
4390 PH/25 When debugging is enabled, the contents of the command line are added
4391 to the debugging output, even when log_selector=+arguments is not
4394 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4395 answer is "GNU", and only if the return is "GNU/something" is the answer
4398 PH/27 $acl_verify_message is now set immediately after the failure of a
4399 verification in an ACL, and so is available in subsequent modifiers. In
4400 particular, the message can be preserved by coding like this:
4402 warn !verify = sender
4403 set acl_m0 = $acl_verify_message
4405 Previously, $acl_verify_message was set only while expanding "message"
4406 and "log_message" when a very denied access.
4408 PH/28 Modified OS/os.c-Linux with
4410 -#ifndef OS_LOAD_AVERAGE
4411 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4413 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4414 nomenclature these days.)
4416 PH/29 Installed patch from the Sieve maintainer that adds the options
4417 sieve_useraddress and sieve_subaddress to the redirect router.
4419 PH/30 In these circumstances:
4420 . Two addresses routed to the same list of hosts;
4421 . First host does not offer TLS;
4422 . First host accepts first address;
4423 . First host gives temporary error to second address;
4424 . Second host offers TLS and a TLS session is established;
4425 . Second host accepts second address.
4426 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4427 and peerdn, if requested) that were in fact used only for the second
4430 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4431 attention to any local part prefix or suffix that was matched by the
4432 router that accepted the address. It now behaves in the same way as it
4433 does for delivery: the affixes are removed from the local part unless
4434 rcpt_include_affixes is set on the transport.
4436 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4437 timeout during the DATA phase of an incoming message.
4439 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4440 applied a patch sanctioned by the Sieve maintainer.
4442 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4443 the uid or gid is negative. A case of a negative gid caused this to be
4444 noticed. The fix allows for either to be negative.
4446 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4447 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4450 PH/36 The change PH/12 above was broken. Fixed it.
4452 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4453 the grounds that routing the same address twice would always produce the
4454 same answer. This might have been true once, but it is certainly no
4455 longer true now. Routing a child address may depend on the previous
4456 routing that produced that child. Some complicated redirection strategies
4457 went wrong when messages had multiple recipients, and made Exim's
4458 behaviour dependent on the order in which the addresses were given.
4460 I have moved the duplicate checking until after the routing is complete.
4461 Exim scans the addresses that are assigned to local and remote
4462 transports, and removes any duplicates. This means that more work will be
4463 done, as duplicates will always all be routed, but duplicates are
4464 presumably rare, so I don't expect this is of any significance.
4466 For deliveries to pipes, files, and autoreplies, the duplicate checking
4467 still happens during the routing process, since they are not going to be
4470 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4471 It corrects a timeout issue with spamd. This is Ian's comment: "The
4472 background is that sometimes spamd either never reads data from a
4473 connection it has accepted, or it never writes response data. The exiscan
4474 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4475 blindly assumes that writes won't block so it may never time out."
4477 PH/39 Allow G after quota size as well as K and M.
4479 PH/40 The value set for $authenticated_id in an authenticator may not contain
4480 binary zeroes or newlines because the value is written to log lines and
4481 to spool files. There was no check on this. Now the value is run through
4482 the string_printing() function so that such characters are converted to
4483 printable escape sequences.
4485 PH/41 $message_linecount is a new variable that contains the total number of
4486 lines in the message. Compare $body_linecount, which is the count for the
4489 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4490 bounce and delay warning messages, except in certain special cases, which
4493 (a) An SMTP error message from a remote host;
4494 (b) A message specified in a :fail: redirection;
4495 (c) A message specified in a "fail" command in a system filter;
4496 (d) A message specified in a FAIL return from the queryprogram router;
4497 (e) A message specified by the cannot_route_message router option.
4499 In these cases only, Exim does include the error details in bounce and
4500 warning messages. There are also a few cases where bland messages such
4501 as "unrouteable address" or "local delivery error" are given.
4503 PH/43 $value is now also set for the "else" part of a ${run expansion.
4505 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4506 being worked on, but at least Exim now implements the latest version to
4509 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4510 process to complete was treated as a delivery failure, a timeout while
4511 writing the message to the pipe was logged, but erroneously treated as a
4512 successful delivery. Such timeouts include transport filter timeouts. For
4513 consistency with the overall process timeout, these timeouts are now
4514 treated as errors, giving rise to delivery failures by default. However,
4515 there is now a new Boolean option for the pipe transport called
4516 timeout_defer, which, if set TRUE, converts the failures into defers for
4517 both kinds of timeout. A transport filter timeout is now identified in
4520 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4521 systems where "make" and "gmake" are different, calling "gmake" at top
4522 level broke things. I've arranged for the value of $(MAKE) to be passed
4523 from the Makefile to this script so that it can call the same version of
4527 A note about Exim versions 4.44 and 4.50
4528 ----------------------------------------
4530 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4531 changes of various kinds. As a consequence, a big documentation update was
4532 needed. This delayed the release for rather longer than seemed good, especially
4533 in the light of a couple of (minor) security issues. Therefore, the changes
4534 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4535 release. So 4.44 and 4.50 are in effect two different branches that both start
4538 I have left the 4.50 change log unchanged; it contains all the changes since
4539 4.43. The change log for 4.44 is below; many of its items are identical to
4540 those for 4.50. This seems to be the most sensible way to preserve the
4541 historical information.
4547 1. Minor wording change to the doc/README.SIEVE file.
4549 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4550 computation of the current number of files was incorrect.
4552 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4553 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4556 4. Give more explanation in the error message when the command for a transport
4557 filter fails to execute.
4559 5. There are several places where Exim runs a non-Exim command in a
4560 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4561 was being done only for the command run by the queryprogram router. It is
4562 now done for all such subprocesses. The other cases are: ${run, transport
4563 filters, and the commands run by the lmtp and pipe transports.
4565 6. Added CONFIGURE_GROUP build-time option.
4567 7. Some older OS have a limit of 256 on the maximum number of file
4568 descriptors. Exim was using setrlimit() to set 1000 as a large value
4569 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4570 systems. I've change it so that if it can't get 1000, it tries for 256.
4572 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4573 was an oversight, and furthermore, ever since the addition of extra
4574 controls (e.g. 4.43/32), the checks on when to allow different forms of
4575 "control" were broken. There should now be diagnostics for all cases when a
4576 control that does not make sense is encountered.
4578 9. Added the /retain_sender option to "control=submission".
4580 10. $recipients is now available in the predata ACL (oversight).
4582 11. Tidy the search cache before the fork to do a delivery from a message
4583 received from the command line. Otherwise the child will trigger a lookup
4584 failure and thereby defer the delivery if it tries to use (for example) a
4585 cached ldap connection that the parent has called unbind on.
4587 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4588 of $address_data from the recipient verification was clobbered by the
4589 sender verification.
4591 13. The value of address_data from a sender verification is now available in
4592 $sender_address_data in subsequent conditions in the ACL statement.
4594 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4596 15. Added a new option "connect=<time>" to callout options, to set a different
4599 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4600 was its contents. (It was OK if the option was not defined at all.)
4602 17. A "Completed" log line is now written for messages that are removed from
4603 the spool by the -Mrm option.
4605 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4606 information about exactly what failed.
4608 19. Added -dd to debug only the daemon process.
4610 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4611 handles timeouts, both on the server side and network timeouts. Renamed the
4612 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4614 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4615 It is now set to "smtps".
4617 22. $host_address is now set to the target address during the checking of
4618 ignore_target_hosts.
4620 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4621 being passed; this would have caused $sender_host_name to have been used if
4622 matching the list had actually called for a host name (not very likely,
4623 since this list is usually IP addresses). A host name is now passed as
4626 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4627 code that specifies a non-restarting handler (typically sigaction() in
4628 modern systems) in an attempt to fix a rare and obscure crash bug.
4630 25. Narrowed the window for a race in the daemon that could cause it to ignore
4631 SIGCHLD signals. This is not a major problem, because they are used only to
4632 wake it up if nothing else does.
4634 26. A malformed maildirsize file could cause Exim to calculate negative values
4635 for the mailbox size or file count. Odd effects could occur as a result.
4636 The maildirsize information is now recalculated if the size or filecount
4639 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4640 support for a long time. Removed HAVE_SYS_VFS_H.
4642 28. Installed the latest version of exipick from John Jetmore.
4644 29. In an address list, if the pattern was not a regular expression, an empty
4645 subject address (from a bounce message) matched only if the pattern was an
4646 empty string. Non-empty patterns were not even tested. This was the wrong
4647 because it is perfectly reasonable to use an empty address as part of a
4648 database query. An empty address is now tested by patterns that are
4649 lookups. However, all the other forms of pattern expect the subject to
4650 contain a local part and a domain, and therefore, for them, an empty
4651 address still always fails if the pattern is not itself empty.
4653 30. Exim went into a mad DNS loop when attempting to do a callout where the
4654 host was specified on an smtp transport, and looking it up yielded more
4655 than one IP address.
4657 31. Re-factored the code for checking spool and log partition space into a
4658 function that finds that data and another that does the check. The former
4659 is then used to implement four new variables: $spool_space, $log_space,
4660 $spool_inodes, and $log_inodes.
4662 32. The RFC2047 encoding function was originally intended for short strings
4663 such as real names; it was not keeping to the 75-character limit for
4664 encoded words that the RFC imposes. It now respects the limit, and
4665 generates multiple encoded words if necessary. To be on the safe side, I
4666 have increased the buffer size for the ${rfc2047: expansion operator from
4669 33. It is now permitted to omit both strings after an "if" condition; if the
4670 condition is true, the result is "true". As before, when the second string
4671 is omitted, a false condition yields an empty string. This makes it less
4672 cumbersome to write custom ACL and router conditions.
4674 34. Failure to deliver a bounce message always caused it to be frozen, even if
4675 there was an errors_to setting on the router. The errors_to setting is now
4678 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4679 canonical form (fully expanded) before being placed in
4680 $sender_host_address.
4682 36. The table in the code that translates DNS record types into text (T_A to
4683 "A" for instance) was missing entries for NS and CNAME. It is just possible
4684 that this could have caused confusion if both these types were looked up
4685 for the same domain, because the text type is used as part of Exim's
4686 per-process caching. But the chance of anyone hitting this buglet seems
4689 37. The dnsdb lookup has been extended in a number of ways.
4691 (1) There is a new type, "zns", which walks up the domain tree until it
4692 finds some nameserver records. It should be used with care.
4694 (2) There is a new type, "mxh", which is like "mx" except that it returns
4695 just the host names, not the priorities.
4697 (3) It is now possible to give a list of domains (or IP addresses) to be
4698 looked up. The behaviour when one of the lookups defers can be
4699 controlled by a keyword.
4701 (4) It is now possible to specify the separator character for use when
4702 multiple records are returned.
4704 38. The dnslists ACL condition has been extended: it is now possible to supply
4705 a list of IP addresses and/or domains to be looked up in a particular DNS
4708 39. Added log_selector=+queue_time_overall.
4710 40. When running the queue in the test harness, wait just a tad after forking a
4711 delivery process, to get repeatability of debugging output.
4713 41. Include certificate and key file names in error message when GnuTLS fails
4714 to set them up, because the GnuTLS error message doesn't include the name
4715 of the failing file when there is a problem reading it.
4717 42. Allow both -bf and -bF in the same test run.
4719 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4721 44. The "Exiscan patch" is now merged into the mainline Exim source.
4723 45. Sometimes the final signoff response after QUIT could fail to get
4724 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4725 < 0 before doing a fflush(). This bug looks as though it goes back to the
4726 introduction of TLS in release 3.20, but "sometimes" must have been rare
4727 because the tests only now provoked it.
4729 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4730 (this can affect the format of dates).
4732 47. exim_tidydb, when checking for the continued existence of a message for
4733 which it has found a message-specific retry record, was not finding
4734 messages that were in split spool directories. Consequently, it was
4735 deleting retry records that should have stayed in existence.
4737 48. Steve fixed some bugs in eximstats.
4739 49. The SPA authentication driver was not abandoning authentication and moving
4740 on to the next authenticator when an expansion was forced to fail,
4741 contradicting the general specification for all authenticators. Instead it
4742 was generating a temporary error. It now behaves as specified.
4744 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4745 (the order specifies the preference for clients). The order is now AES256,
4746 AES128, 3DES, ARCFOUR128.
4748 51. Small patch to Sieve code - explicitly set From: when generating an
4751 52. Exim crashed if a remote delivery caused a very long error message to be
4752 recorded - for instance if somebody sent an entire SpamAssassin report back
4753 as a large number of 550 error lines. This bug was coincidentally fixed by
4754 increasing the size of one of Exim's internal buffers (big_buffer) that
4755 happened as part of the Exiscan merge. However, to be on the safe side, I
4756 have made the code more robust (and fixed the comments that describe what
4759 53. Now that there can be additional text after "Completed" in log lines (if
4760 the queue_time_overall log selector is set), a one-byte patch to exigrep
4761 was needed to allow it to recognize "Completed" as not the last thing in
4764 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4765 patch that reportedly fixes this has been added. I am not expert enough to
4766 create a test for it. This is what the patch creator wrote:
4768 "I found a little strange behaviour of ldap code when working with
4769 Windows 2003 AD Domain, where users was placed in more than one
4770 Organization Units. When I tried to give exim partial DN, the exit code
4771 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4772 But simultaneously result of request was absolutely normal ldap result,
4773 so I produce this patch..."
4775 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4776 REFERENCE, so I have modified the code to exclude the patch when that macro
4779 55. Some experimental protocols are using DNS PTR records for new purposes. The
4780 keys for these records are domain names, not reversed IP addresses. The
4781 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4782 leaves it alone. Component reversal etc. now happens only for IP addresses.
4785 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4787 57. Double the size of the debug message buffer (to 2048) so that more of very
4788 long debug lines gets shown.
4790 58. The exicyclog utility now does better if the number of log files to keep
4791 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4793 59. Two changes related to the smtp_active_hostname option:
4795 (1) $smtp_active_hostname is now available as a variable.
4796 (2) The default for smtp_banner uses $smtp_active_hostname instead
4797 of $primary_hostname.
4799 60. The host_aton() function is supposed to be passed a string that is known
4800 to be a valid IP address. However, in the case of IPv6 addresses, it was
4801 not checking this. This is a hostage to fortune. Exim now panics and dies
4802 if the condition is not met. A case was found where this could be provoked
4803 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4804 components; fortuitously, this particular loophole had already been fixed
4805 by change 4.50/55 above.
4807 If there are any other similar loopholes, the new check in host_aton()
4808 itself should stop them being exploited. The report I received stated that
4809 data on the command line could provoke the exploit when Exim was running as
4810 exim, but did not say which command line option was involved. All I could
4811 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4812 running as the user.
4815 61. There was a buffer overflow vulnerability in the SPA authentication code
4816 (which came originally from the Samba project). I have added a test to the
4817 spa_base64_to_bits() function which I hope fixes it.
4820 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4821 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4823 63. The daemon start-up calls getloadavg() while still root for those OS that
4824 need the first call to be done as root, but it missed one case: when
4825 deliver_queue_load_max is set with deliver_drop_privilege. This is
4826 necessary for the benefit of the queue runner, because there is no re-exec
4827 when deliver_drop_privilege is set.
4829 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4830 This has been fixed.
4832 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4833 was in use, was not putting the data itself into the right store pool;
4834 consequently, it could be overwritten for a subsequent message in the same
4835 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4838 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4840 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4841 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4842 the function were treating the return as a boolean value, which happened to
4843 work because 0=false and not-0=true, but is not correct code.
4845 68. The host_aton() function was not handling scoped IPv6 addresses (those
4846 with, for example, "%eth0" on the end) correctly.
4848 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4849 negated items (that is, ~something) in unsigned ints. Some compilers
4850 apparently mutter when there is no cast.
4852 70. If an address verification called from an ACL failed, and did not produce a
4853 user-specific message (i.e. there was only a "system" message), nothing was
4854 put in $acl_verify_message. In this situation, it now puts the system
4857 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4858 session; change 4.31/43 added the unwanted input to the log line - except
4859 that it did not do this in the start of session case. It now does.
4861 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4862 This is wrong and can cause the other end to generate a synchronization
4863 error if it is another Exim or anything else that does the synchronization
4864 check. A QUIT command is no longer sent after a timeout.
4866 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4867 during host lookups.
4869 74. The defer_ok option of callout verification was not working if it was used
4870 when verifying addresses in header lines, that is, for this case:
4872 verify = header_sender/callout=defer_ok
4874 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4875 those file descriptors could be used for SMTP connections. If anything
4876 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4877 could be sent to the SMTP client, causing chaos. The daemon now opens
4878 stdin, stdout, and stderr to /dev/null when it puts itself into the
4881 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4882 log by default. The user can override this with suitable Perl magic.
4884 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4885 the log message when discard triggers, was not working for the DATA ACL or
4886 for the non-SMTP ACL.
4888 78. Error message wording change in sieve.c.
4890 79. If smtp_accept_max_per_host was set, the number of connections could be
4891 restricted to fewer than expected, because the daemon was trying to set up
4892 a new connection before checking whether the processes handling previous
4893 connections had finished. The check for completed processes is now done
4894 earlier. On busy systems, this bug wouldn't be noticed because something
4895 else would have woken the daemon, and it would have reaped the completed
4898 80. If a message was submitted locally by a user whose login name contained one
4899 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4900 It caused a spool format error. I have fixed the spool reading code. A
4901 related problem was that the "from" clause in the Received: line became
4902 illegal because of the space(s). It is now covered by ${quote_local_part.
4904 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4907 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4909 83. Patch from Sieve maintainer to fix unterminated string problem in
4910 "vacation" handling.
4912 84. Some minor changes to the Linux configuration files to help with other
4913 OS variants using glibc.
4915 85. One more patch for Sieve to update vacation handling to latest spec.
4918 ----------------------------------------------------
4919 See the note above about the 4.44 and 4.50 releases.
4920 ----------------------------------------------------
4926 1. Change 4.43/35 introduced a bug that caused file counts to be
4927 incorrectly computed when quota_filecount was set in an appendfile
4930 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4931 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4934 3. Give more explanation in the error message when the command for a transport
4935 filter fails to execute.
4937 4. There are several places where Exim runs a non-Exim command in a
4938 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4939 was being done only for the command run by the queryprogram router. It is
4940 now done for all such subprocesses. The other cases are: ${run, transport
4941 filters, and the commands run by the lmtp and pipe transports.
4943 5. Some older OS have a limit of 256 on the maximum number of file
4944 descriptors. Exim was using setrlimit() to set 1000 as a large value
4945 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4946 systems. I've change it so that if it can't get 1000, it tries for 256.
4948 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4949 was an oversight, and furthermore, ever since the addition of extra
4950 controls (e.g. 4.43/32), the checks on when to allow different forms of
4951 "control" were broken. There should now be diagnostics for all cases when a
4952 control that does not make sense is encountered.
4954 7. $recipients is now available in the predata ACL (oversight).
4956 8. Tidy the search cache before the fork to do a delivery from a message
4957 received from the command line. Otherwise the child will trigger a lookup
4958 failure and thereby defer the delivery if it tries to use (for example) a
4959 cached ldap connection that the parent has called unbind on.
4961 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4962 of $address_data from the recipient verification was clobbered by the
4963 sender verification.
4965 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4966 was its contents. (It was OK if the option was not defined at all.)
4968 11. A "Completed" log line is now written for messages that are removed from
4969 the spool by the -Mrm option.
4971 12. $host_address is now set to the target address during the checking of
4972 ignore_target_hosts.
4974 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4975 being passed; this would have caused $sender_host_name to have been used if
4976 matching the list had actually called for a host name (not very likely,
4977 since this list is usually IP addresses). A host name is now passed as
4980 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4981 code that specifies a non-restarting handler (typically sigaction() in
4982 modern systems) in an attempt to fix a rare and obscure crash bug.
4984 15. Narrowed the window for a race in the daemon that could cause it to ignore
4985 SIGCHLD signals. This is not a major problem, because they are used only to
4986 wake it up if nothing else does.
4988 16. A malformed maildirsize file could cause Exim to calculate negative values
4989 for the mailbox size or file count. Odd effects could occur as a result.
4990 The maildirsize information is now recalculated if the size or filecount
4993 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4994 support for a long time. Removed HAVE_SYS_VFS_H.
4996 18. Updated exipick to current release from John Jetmore.
4998 19. Allow an empty sender to be matched against a lookup in an address list.
4999 Previously the only cases considered were a regular expression, or an
5002 20. Exim went into a mad DNS lookup loop when doing a callout where the
5003 host was specified on the transport, if the DNS lookup yielded more than
5006 21. The RFC2047 encoding function was originally intended for short strings
5007 such as real names; it was not keeping to the 75-character limit for
5008 encoded words that the RFC imposes. It now respects the limit, and
5009 generates multiple encoded words if necessary. To be on the safe side, I
5010 have increased the buffer size for the ${rfc2047: expansion operator from
5013 22. Failure to deliver a bounce message always caused it to be frozen, even if
5014 there was an errors_to setting on the router. The errors_to setting is now
5017 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5018 canonical form (fully expanded) before being placed in
5019 $sender_host_address.
5021 24. Updated eximstats to version 1.33
5023 25. Include certificate and key file names in error message when GnuTLS fails
5024 to set them up, because the GnuTLS error message doesn't include the name
5025 of the failing file when there is a problem reading it.
5027 26. Expand error message when OpenSSL has problems setting up cert/key files.
5030 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5031 (this can affect the format of dates).
5033 28. exim_tidydb, when checking for the continued existence of a message for
5034 which it has found a message-specific retry record, was not finding
5035 messages that were in split spool directories. Consequently, it was
5036 deleting retry records that should have stayed in existence.
5038 29. eximstats updated to version 1.35
5039 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5040 1.35 - bugfix such that pie charts by volume are generated correctly
5042 30. The SPA authentication driver was not abandoning authentication and moving
5043 on to the next authenticator when an expansion was forced to fail,
5044 contradicting the general specification for all authenticators. Instead it
5045 was generating a temporary error. It now behaves as specified.
5047 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5048 (the order specifies the preference for clients). The order is now AES256,
5049 AES128, 3DES, ARCFOUR128.
5051 31. Small patch to Sieve code - explicitly set From: when generating an
5054 32. Exim crashed if a remote delivery caused a very long error message to be
5055 recorded - for instance if somebody sent an entire SpamAssassin report back
5056 as a large number of 550 error lines. This bug was coincidentally fixed by
5057 increasing the size of one of Exim's internal buffers (big_buffer) that
5058 happened as part of the Exiscan merge. However, to be on the safe side, I
5059 have made the code more robust (and fixed the comments that describe what
5062 33. Some experimental protocols are using DNS PTR records for new purposes. The
5063 keys for these records are domain names, not reversed IP addresses. The
5064 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5065 leaves it alone. Component reversal etc. now happens only for IP addresses.
5068 34. The host_aton() function is supposed to be passed a string that is known
5069 to be a valid IP address. However, in the case of IPv6 addresses, it was
5070 not checking this. This is a hostage to fortune. Exim now panics and dies
5071 if the condition is not met. A case was found where this could be provoked
5072 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5073 components; fortuitously, this particular loophole had already been fixed
5074 by change 4.50/55 or 4.44/33 above.
5076 If there are any other similar loopholes, the new check in host_aton()
5077 itself should stop them being exploited. The report I received stated that
5078 data on the command line could provoke the exploit when Exim was running as
5079 exim, but did not say which command line option was involved. All I could
5080 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5081 running as the user.
5084 35. There was a buffer overflow vulnerability in the SPA authentication code
5085 (which came originally from the Samba project). I have added a test to the
5086 spa_base64_to_bits() function which I hope fixes it.
5089 36. The daemon start-up calls getloadavg() while still root for those OS that
5090 need the first call to be done as root, but it missed one case: when
5091 deliver_queue_load_max is set with deliver_drop_privilege. This is
5092 necessary for the benefit of the queue runner, because there is no re-exec
5093 when deliver_drop_privilege is set.
5095 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5096 was in use, was not putting the data itself into the right store pool;
5097 consequently, it could be overwritten for a subsequent message in the same
5098 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5101 38. Sometimes the final signoff response after QUIT could fail to get
5102 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5103 < 0 before doing a fflush(). This bug looks as though it goes back to the
5104 introduction of TLS in release 3.20, but "sometimes" must have been rare
5105 because the tests only now provoked it.
5111 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5112 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5113 Now it returns an integer. A number of calls were still expecting a T/F
5114 return. Fortuitously, in all cases, the tests worked in OK situations,
5115 which is the norm. However, things would have gone wrong on any write
5116 failures on the smtp file descriptor. This function is used when sending
5117 messages over SMTP and also when doing verify callouts.
5119 2. When Exim is called to do synchronous delivery of a locally submitted
5120 message (the -odf or -odi options), it no longer closes stderr before doing
5123 3. Implemented the mua_wrapper option.
5125 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5127 5. Implemented the functions header_remove(), header_testname(),
5128 header_add_at_position(), and receive_remove_recipient(), and exported them
5131 6. If an ACL "warn" statement specified the addition of headers, Exim already
5132 inserted X-ACL-Warn: at the start if there was no header name. However, it
5133 was not making this test for the second and subsequent header lines if
5134 there were newlines in the string. This meant that an invalid header could
5135 be inserted if Exim was badly configured.
5137 7. Allow an ACL "warn" statement to add header lines at the start or after all
5138 the Received: headers, as well as at the end.
5140 8. Added the rcpt_4xx retry error code.
5142 9. Added postmaster_mailfrom=xxx to callout verification option.
5144 10. Added mailfrom=xxxx to the callout verification option, for verify=
5147 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5148 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5150 12. Inserted some casts to stop certain compilers warning when using pointer
5151 differences as field lengths or precisions in printf-type calls (mostly
5152 affecting debugging statements).
5154 13. Added optional readline() support for -be (dynamically loaded).
5156 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5157 same clock tick as a message's arrival, so that its received time was the
5158 same as the "first fail" time on the retry record, and that message
5159 remained on the queue past the ultimate address timeout, every queue runner
5160 would try a delivery (because it was past the ultimate address timeout) but
5161 after another failure, the ultimate address timeout, which should have then
5162 bounced the address, did not kick in. This was a "< instead of <=" error;
5163 in most cases the first failure would have been in the next clock tick
5164 after the received time, and all would be well.
5166 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5167 being recognized when the domain list was tested by the match_domain
5168 condition in an expansion string.
5170 16. Added the ${str2b64: operator.
5172 17. Exim was always calling setrlimit() to set a large limit for the number of
5173 processes, without checking whether the existing limit was already
5174 adequate. (It did check for the limit on file descriptors.) Furthermore,
5175 errors from getrlimit() and setrlimit() were being ignored. Now they are
5176 logged to the main and panic logs, but Exim does carry on, to try to do its
5177 job under whatever limits there are.
5179 18. Imported PCRE 5.0.
5181 19. Trivial typo in log message " temporarily refused connection" (the leading
5184 20. If the log selector return_path_on_delivery was set and an address was
5185 redirected to /dev/null, the delivery process crashed because it assumed
5186 that a return path would always be set for a "successful" delivery. In this
5187 case, the whole delivery is bypassed as an optimization, and therefore no
5190 21. Internal re-arrangement: the function for sending a challenge and reading
5191 a response while authentication was assuming a zero-terminated challenge
5192 string. It's now changed to take a pointer and a length, to allow for
5193 binary data in such strings.
5195 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5197 23. Exim was not respecting finduser_retries when seeking the login of the
5198 uid under which it was called; it was always trying 10 times. (The default
5199 setting of finduser_retries is zero.) Also, it was sleeping after the final
5200 failure, which is pointless.
5202 24. Implemented tls_on_connect_ports.
5204 25. Implemented acl_smtp_predata.
5206 26. If the domain in control=submission is set empty, Exim assumes that the
5207 authenticated id is a complete email address when it generates From: or
5208 Sender: header lines.
5210 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5211 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5212 chown and chgrp in /bin and hostname in /usr/bin.
5214 28. Exim was keeping the "process log" file open after each use, just as it
5215 does for the main log. This opens the possibility of it remaining open for
5216 long periods when the USR1 signal hits a daemon. Occasional processlog
5217 errors were reported, that could have been caused by this. Anyway, it seems
5218 much more sensible not to leave this file open at all, so that is what now
5221 29. The long-running daemon process does not normally write to the log once it
5222 has entered its main loop, and it closes the log before doing so. This is
5223 so that log files can straightforwardly be renamed and moved. However,
5224 there are a couple of unusual error situations where the daemon does write
5225 log entries, and I had neglected to close the log afterwards.
5227 30. The text of an SMTP error response that was received during a remote
5228 delivery was being truncated at 512 bytes. This is too short for some of
5229 the long messages that one sometimes sees. I've increased the limit to
5232 31. It is now possible to make retry rules that apply only when a message has a
5233 specific sender, in particular, an empty sender.
5235 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5236 it possible to be selective about when SMTP synchronization is enforced.
5238 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5240 32. Implemented hosts_connection_nolog.
5242 33. Added an ACL for QUIT.
5244 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5247 35. Added mailbox_size and mailbox_filecount to appendfile.
5249 36. Added control = no_multiline_responses to ACLs.
5251 37. There was a bug in the logic of the code that waits for the clock to tick
5252 in the case where the clock went backwards by a substantial amount such
5253 that the microsecond fraction of "now" was more than the microsecond
5254 fraction of "then" (but the whole seconds number was less).
5256 38. Added support for the libradius Radius client library this is found on
5257 FreeBSD (previously only the radiusclient library was supported).
5263 1. When certain lookups returned multiple values in the form name=value, the
5264 quoting of the values was not always being done properly. Specifically:
5265 (a) If the value started with a double quote, but contained no whitespace,
5267 (b) If the value contained whitespace other than a space character (i.e.
5268 tabs or newlines or carriage returns) it was not quoted.
5269 This fix has been applied to the mysql and pgsql lookups by writing a
5270 separate quoting function and calling it from the lookup code. The fix
5271 should probably also be applied to nisplus, ibase and oracle lookups, but
5272 since I cannot test any of those, I have not disturbed their existing code.
5274 2. A hit in the callout cache for a specific address caused a log line with no
5275 reason for rejecting RCPT. Now it says "Previous (cached) callout
5276 verification failure".
5278 3. There was an off-by-one bug in the queryprogram router. An over-long
5279 return line was truncated at 256 instead of 255 characters, thereby
5280 overflowing its buffer with the terminating zero. As well as fixing this, I
5281 have increased the buffer size to 1024 (and made a note to document this).
5283 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5284 when Exim is waiting for an SMTP response from a remote server, Exim
5285 restarts its select() call on the socket, thereby resetting its timeout.
5286 This is not a problem when such interrupts are rare. Somebody set up a cron
5287 job to run exiwhat every 2 minutes, which is less than the normal select()
5288 timeout (5 or 10 minutes). This meant that the select() timeout never
5289 kicked in because it was always reset. I have fixed this by comparing the
5290 time when an interrupt arrives with the time at the start of the first call
5291 to select(). If more time than the timeout has elapsed, the interrupt is
5292 treated as a timeout.
5294 5. Some internal re-factoring in preparation for the addition of Sieve
5295 extensions (by MH). In particular, the "personal" test is moved to a
5296 separate function, and given an option for scanning Cc: and Bcc: (which is
5297 not set for Exim filters).
5299 6. When Exim created an email address using the login of the caller as the
5300 local part (e.g. when creating a From: or Sender: header line), it was not
5301 quoting the local part when it contained special characters such as @.
5303 7. Installed new OpenBSD configuration files.
5305 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5306 try to make them clearer.
5308 9. Callout options, other than the timeout value, were being ignored when
5309 verifying sender addresses in header lines. For example, when using
5311 verify = header_sender/callout=no_cache
5313 the cache was (incorrectly) being used.
5315 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5316 only the Cygwin environment.
5318 11. When return_path_on_delivery was set as a log selector, if different remote
5319 addresses in the same message used different return paths and parallel
5320 remote delivery occurred, the wrong values would sometimes be logged.
5321 (Whenever a remote delivery process finished, the return path value from
5322 the most recently started remote delivery process was logged.)
5324 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5325 lines when AUTH and/or TLS are in use. This is the "received protocol"
5326 field. Exim used to use "asmtp" for authenticated SMTP, without any
5327 indication (in the protocol name) for TLS use. Now it follows the RFC and
5328 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5329 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5330 names appear in log lines as well as in Received: header lines.
5332 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5333 extensions, and comparison tests, and to fix some bugs.
5335 14. Changes to the "personal" filter test:
5337 (1) The test was buggy in that it was just doing the equivalent of
5338 "contains" tests on header lines. For example, if a user's address was
5339 anne@some.where, the "personal" test would incorrectly be true for
5341 To: susanne@some.where
5343 This test is now done by extracting each address from the header in turn,
5344 and checking the entire address. Other tests that are part of "personal"
5345 are now done using regular expressions (for example, to check local parts
5346 of addresses in From: header lines).
5348 (2) The list of non-personal local parts in From: addresses has been
5349 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5350 taken from the Sieve specification recommendations.
5352 (3) If the message contains any header line starting with "List-" it is
5353 treated as non-personal.
5355 (4) The test for "circular" in the Subject: header line has been removed
5356 because it now seems ill-conceived.
5358 15. Minor typos in src/EDITME comments corrected.
5360 16. Installed latest exipick from John Jetmore.
5362 17. If headers_add on a router specified a text string that was too long for
5363 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5364 of string_sprintf() is now avoided.
5366 18. $message_body_size was not set (it was always zero) when running the DATA
5367 ACL and the local_scan() function.
5369 19. For the "mail" command in an Exim filter, no default was being set for
5370 the once_repeat time, causing a random time value to be used if "once" was
5371 specified. (If the value happened to be <= 0, no repeat happened.) The
5372 default is now 0s, meaning "never repeat". The "vacation" command was OK
5373 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5374 (I found it when inspecting the code).
5376 20. There is now an overall timeout for performing a callout verification. It
5377 defaults to 4 times the callout timeout, which applies to individual SMTP
5378 commands during the callout. The overall timeout applies when there is more
5379 than one host that can be tried. The timeout is checked before trying the
5380 next host. This prevents very long delays if there are a large number of
5381 hosts and all are timing out (e.g. when the network connections are timing
5382 out). The value of the overall timeout can be changed by specifying an
5383 additional sub-option for "callout", called "maxwait". For example:
5385 verify = sender/callout=5s,maxwait=20s
5387 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5388 to the end before writing, but this should make it even safer).
5390 22. Exim was forgetting that it had advertised PIPELINING for the second and
5391 subsequent messages on an SMTP connection. It was also not resetting its
5392 memory on STARTTLS and an internal HELO.
5394 23. When Exim logs an SMTP synchronization error within a session, it now
5395 records whether PIPELINING has been advertised or not.
5397 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5398 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5399 rather than long int.
5401 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5403 26. Added the never_mail option to autoreply.
5409 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5410 crash if the getsockname() call failed; this can happen if a connection is
5411 closed very soon after it is established. The problem was simply in the
5412 order in which certain operations were done, causing Exim to try to write
5413 to the SMTP stream before it had set up the file descriptor. The bug has
5414 been fixed by making things happen in the correct order.
5420 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5421 before the connection was closed, thus losing the rejection response.
5423 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5424 some early Solaris releases, but causes trouble in current releases where
5425 socklen_t is defined.
5427 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5430 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5433 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5434 that releasing the top of it at the end releases what was used for sub-
5435 expansions (unless the block got too big). However, discard this block if
5436 the first thing is a variable or header, so that we can use its block when
5437 it is dynamic (useful for very large $message_headers, for example).
5439 6. Lookups now cache *every* query, not just the most recent. A new, separate
5440 store pool is used for this. It can be recovered when all lookup caches are
5441 flushed. Lookups now release memory at the end of their result strings.
5442 This has involved some general refactoring of the lookup sources.
5444 7. Some code has been added to the store_xxx() functions to reduce the amount
5445 of flapping under certain conditions.
5447 8. log_incoming_interface used to affect only the <= reception log lines. Now
5448 it causes the local interface and port to be added to several more SMTP log
5449 lines, for example "SMTP connection from", and rejection lines.
5451 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5453 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5455 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5456 could be overwritten at the end of the current message (or the start of a
5457 new message if it was set in a HELO ACL). The value is now preserved for
5458 the duration of the SMTP connection.
5460 12. If a transport had a headers_rewrite setting, and a matching header line
5461 contained an unqualified address, that address was qualified, even if it
5462 did not match any rewriting rules. The underlying bug was that the values
5463 of the flags that permit the existence of unqualified sender and recipient
5464 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5465 non-local messages, and by -bnq for local messages) were not being
5466 preserved with the message after it was received.
5468 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5469 "next input=" as part of the text comprising the host identity instead of
5470 the correct text. The code was using the same buffer for two different
5471 strings. However, depending on which order the printing function evaluated
5472 its arguments, the bug did not always show up. Under Linux, for example, my
5473 test suite worked just fine.
5475 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5476 doesn't work with some older versions of Perl. It has been changed to "my",
5477 which in any case is probably the better facility to use.
5479 15. A really picky compiler found some instances of statements for creating
5480 error messages that either had too many or two few arguments for the format
5483 16. The size of the buffer for calls to the DNS resolver has been increased
5484 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5485 for addresses that have a lot of PTR records. This alleviates a problem; it
5486 does not fully solve it.
5488 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5489 buffer now truncates the list and logs the incident, which is the same
5490 action as happens when Exim is looking up a host name and its aliases.
5491 Previously in this situation something unpredictable would happen;
5492 sometimes it was "internal error: store_reset failed".
5494 18. If a server dropped the connection unexpectedly when an Exim client was
5495 using GnuTLS and trying to read a response, the client delivery process
5496 crashed while trying to generate an error log message.
5498 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5499 string, for example:
5501 warn message = H1: something\nH2: something
5503 the text was added as a single header line from Exim's point of view
5504 though it ended up OK in the delivered message. However, searching for the
5505 second and subsequent header lines using $h_h2: did not work. This has been
5506 fixed. Similarly, if a system filter added multiple headers in this way,
5507 the routers could not see them.
5509 20. Expanded the error message when iplsearch is called with an invalid key to
5510 suggest using net-iplsearch in a host list.
5512 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5513 ACLs are no longer actually imposed (and a message to that effect is
5516 22. If a "gecos" field in a passwd entry contained escaped characters, in
5517 particular, if it contained a \" sequence, Exim got it wrong when building
5518 a From: or a Sender: header from that name. A second bug also caused
5519 incorrect handling when an unquoted " was present following a character
5520 that needed quoting.
5522 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5523 was not being matched caselessly.
5525 24. Arranged for all hyphens in the exim.8 source to be escaped with
5528 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5529 properly. Recipient callout cache records were still being keyed to include
5530 the sender, even when use_sender was set false. This led to far more
5531 callouts that were necessary. The sender is no longer included in the key
5532 when use_sender is false.
5534 26. Added "control = submission" modifier to ACLs.
5536 27. Added the ${base62d: operator to decode base 62 numbers.
5538 28. dnsdb lookups can now access SRV records.
5540 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5541 the configuration file.
5543 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5544 (-v) mode. This makes the output for a verbose queue run more intelligible.
5546 31. Added a use_postmaster feature to recipient callouts.
5548 32. Added the $body_zerocount variable, containing the number of binary zero
5549 bytes in the message body.
5551 33. The time of last modification of the "new" subdirectory is now used as the
5552 "mailbox time last read" when there is a quota error for a maildir
5555 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5557 35. Added +ignore_unknown as a special item in host lists.
5559 36. Code for decoding IPv6 addresses in host lists is now included, even if
5560 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5561 address was recognized as an IP address, but was then not correctly decoded
5562 into binary, causing unexpected and incorrect effects when compared with
5569 1. Very minor rewording of debugging text in manualroute to say "list of
5570 hosts" instead of "hostlist".
5572 2. If verify=header_syntax was set, and a header line with an unqualified
5573 address (no domain) and a large number of spaces between the end of the
5574 name and the colon was received, the reception process suffered a buffer
5575 overflow, and (when I tested it) crashed. This was caused by some obsolete
5576 code that should have been removed. The fix is to remove it!
5578 3. When running in the test harness, delay a bit after writing a bounce
5579 message to get a bit more predictability in the log output.
5581 4. Added a call to search_tidyup() just before forking a reception process. In
5582 theory, someone could use a lookup in the expansion of smtp_accept_max_
5583 per_host which, without the tidyup, could leave open a database connection.
5585 5. Added the variables $recipient_data and $sender_data which get set from a
5586 lookup success in an ACL "recipients" or "senders" condition, or a router
5587 "senders" option, similar to $domain_data and $local_part_data.
5589 6. Moved the writing of debug_print from before to after the "senders" test
5592 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5593 problems for message scanning, either using a data ACL, or using
5594 local_scan() because the Received: header was not generated till after they
5595 were called (in order to set the time as the time of reception completion).
5596 I have revised the way this works. The header is now generated after the
5597 body is received, but before the ACL or local_scan() are called. After they
5598 are run, the timestamp in the header is updated.
5604 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5605 before starting a queue runner without re-exec. This happened only when
5606 deliver_drop_privilege was set or when the Exim user was set to root. The
5607 effect of the bug was that timeouts during subsequent deliveries caused
5608 crashes instead of being properly handled. The handler is now left at its
5609 default (and expected) setting.
5611 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5612 message, again when deliver_drop_privilege is set or Exim is run as root.
5613 The bug described in (1) was not present in this case, but the tidying up
5614 of the other signals was missing. I have made the two cases consistent.
5616 3. The ignore_target_hosts setting on a manualroute router was being ignored
5617 for hosts that were looked up using the /MX notation.
5619 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5622 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5623 operated on the sender address. After changing the $sender_address to <>
5624 for the sender address verify, Exim was re-instated it as the original
5625 (before rewriting) address, but remembering that it had rewritten it, so it
5626 wasn't rewriting it again. This bug also had the effect of breaking the
5627 sender address verification caching when the sender address was rewritten.
5629 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5630 This has been changed so that if the ip literal address matches
5631 ignore_target_hosts, the router declines.
5633 7. Added expansion conditions match_domain, match_address, and match_local_
5634 part (NOT match_host).
5636 8. The placeholder for the Received: header didn't have a length field set.
5638 9. Added code to Exim itself and to exim_lock to test for a specific race
5639 condition that could lead to file corruption when using MBX delivery. The
5640 issue is with the lockfile that is created in /tmp. If this file is removed
5641 after a process has opened it but before that process has acquired a lock,
5642 there is the potential for a second process to recreate the file and also
5643 acquire a lock. This could lead to two Exim processes writing to the file
5644 at the same time. The added code performs the same test as UW imapd; it
5645 checks after acquiring the lock that its file descriptor still refers to
5646 the same named file.
5648 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5649 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5652 11. Added the smtp_active_hostname option. If used, this will typically be made
5653 to depend on the incoming interface address. Because $interface_address is
5654 not set up until the daemon has forked a reception process, error responses
5655 that can happen earlier (such as "too many connections") no longer contain
5658 12. If an expansion in a condition on a "warn" statement fails because a lookup
5659 defers, the "warn" statement is abandoned, and the next ACL statement is
5660 processed. Previously this caused the whole ACL to be aborted.
5662 13. Added the iplsearch lookup type.
5664 14. Added ident_timeout as a log selector.
5666 15. Added tls_certificate_verified as a log selector.
5668 16. Added a global option tls_require_ciphers (compare the smtp transport
5669 option of the same name). This controls incoming TLS connections.
5671 17. I finally figured out how to make tls_require_ciphers do a similar thing
5672 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5673 before starting the TLS session.
5675 18. Tabs are now shown as \t in -bP output.
5677 19. If the log selector return_path_on_delivery was set, Exim crashed when
5678 bouncing a message because it had too many Received: header lines.
5680 20. If two routers both had headers_remove settings, and the first one included
5681 a superfluous trailing colon, the final name in the first list and the
5682 first name in the second list were incorrectly joined into one item (with a
5683 colon in the middle).
5689 1. Added -C and -D options to the exinext utility, mainly to make it easier
5690 to include in the automated testing, but these could be helpful when
5691 multiple configurations are in use.
5693 2. The exinext utility was not formatting the output nicely when there was
5694 an alternate port involved in the retry record key, nor when there was a
5695 message id as well (for retries that were specific to a specific message
5696 and a specific host). It was also confused by IPv6 addresses, because of
5697 the additional colons they contain. I have fixed the IPv4 problem, and
5698 patched it up to do a reasonable job for IPv6.
5700 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5701 delivery, the log line now contains "pipelined" if PIPELINING was used.
5703 4. An SMTP transport process used to panic and die if the bind() call to set
5704 an explicit outgoing interface failed. This has been changed; it is now
5705 treated in the same way as a connect() failure.
5707 5. A reference to $sender_host_name in the part of a conditional expansion
5708 that was being skipped was still causing a DNS lookup. This no longer
5711 6. The def: expansion condition was not recognizing references to header lines
5712 that used bh_ and bheader_.
5714 7. Added the _cache feature to named lists.
5716 8. The code for checking quota_filecount in the appendfile transport was
5717 allowing one more file than it should have been.
5719 9. For compatibility with Sendmail, the command line option
5727 and sets the incoming protocol and host name (for trusted callers). The
5728 host name and its colon can be omitted when only the protocol is to be set.
5729 Note the Exim already has two private options, -pd and -ps, that refer to
5730 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5731 "s", but I don't think that's a major issue.
5733 10. A number of refactoring changes to the code, none of which should affect
5736 (a) The number of logging options was getting close to filling up the
5737 32-bit word that was used as a bit map. I have split them into two classes:
5738 those that are passed in the argument to log_write(), and those that are
5739 only ever tested independently outside of that function. These are now in
5740 separate 32-bit words, so there is plenty of room for expansion again.
5741 There is no change in the user interface or the logging behaviour.
5743 (b) When building, for example, log lines, the code previously used a
5744 macro that called string_cat() twice, in order to add two strings. This is
5745 not really sufficiently general. Furthermore, there was one instance where
5746 it was actually wrong because one of the argument was used twice, and in
5747 one call a function was used. (As it happened, calling the function twice
5748 did not affect the overall behaviour.) The macro has been replaced by a
5749 function that can join an arbitrary number of extra strings onto a growing
5752 (c) The code for expansion conditions now uses a table and a binary chop
5753 instead of a serial search (which was left over from when there were very
5754 few conditions). Also, it now recognizes conditions like "pam" even when
5755 the relevant support is not compiled in: a suitably worded error message is
5756 given if an attempt is made to use such a condition.
5758 11. Added ${time_interval:xxxxx}.
5760 12. A bug was causing one of the ddress fields not to be passed back correctly
5761 from remote delivery subprocesses. The field in question was not being
5762 subsequently used, so this caused to problems in practice.
5764 13. Added new log selectors queue_time and deliver_time.
5766 14. Might have fixed a bug in maildirsizefile handling that threw up
5767 "unexpected character" debug warnings, and recalculated the data
5768 unnecessarily. In any case, I expanded the warning message to give more
5771 15. Added the message "Restricted characters in address" to the statements in
5772 the default ACL that block characters like @ and % in local parts.
5774 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5775 Three changes have been made:
5777 (a) There was a serious bug; a negative response to MAIL caused the whole
5778 recipient domain to be cached as invalid, thereby blocking all messages
5779 to all local parts at the same domain, from all senders. This bug has
5780 been fixed. The domain is no longer cached after a negative response to
5781 MAIL if the sender used is not empty.
5783 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5786 (c) A new callout option, "use_sender" has been added for people who want
5787 the modified behaviour.
5793 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5796 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5797 indeed breaks things for older releases.
5799 3. Added additional logging to the case where there is a problem reading data
5800 from a filter that is running in a subprocess using a pipe, in order to
5801 try to track down a specific problem.
5803 4. Testing facility fudge: when running in the test harness and attempting
5804 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5805 getting "No route to host". Convert this to a timeout.
5807 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5810 6. Some OS don't have socklen_t but use size_t instead. This affects the
5811 fifth argument of getsockopt() amongst other things. This is now
5812 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5813 can be set for individual OS. I have set it for SunOS5, OSF1, and
5814 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5815 some earlier ones do not.
5817 7. Change 4.30/15 was not doing the test caselessly.
5819 8. The standard form for an IPv6 address literal was being rejected by address
5820 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5821 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5822 this, as well as the form without the "IPv6" on the front (but only when
5823 address literals are enabled, of course).
5825 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5827 10. Exim crashed if a message with an empty sender address specified by -f
5828 encountered a router with an errors_to setting. This could be provoked only
5829 by a command such as
5833 where an empty string was supplied; "<>" did not hit this bug.
5835 11. Installed PCRE release 4.5.
5837 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5838 remained set. It is now erased.
5840 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5841 times from message ids (which are base 36 rather than the normal 62).
5843 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5844 were being counted as actual protocol errors, and logged if the log
5845 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5846 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5847 and DATA following a set of rejected RCPTs do not count as protocol errors.
5848 In other words, Exim assumes they were pipelined, though this may not
5849 actually be the case. Of course, in all cases the client gets an
5850 appropriate error code.
5852 15. If a lookup fails in an ACL condition, a message about the failure may
5853 be available; it is used if testing the ACL cannot continue, because most
5854 such messages specify what the cause of the deferral is. However, some
5855 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5856 that caused an old message to be retained and used if a later statement
5857 caused a defer, replacing the real cause of the deferral.
5859 16. If an IP address had so many PTR records that the DNS lookup buffer
5860 was not large enough to hold them, Exim could crash while trying to process
5861 the truncated data. It now detects and logs this case.
5863 17. Further to 4.21/58, another change has been made: if (and only if) the
5864 first line of a message (the first header line) ends with CRLF, a bare LF
5865 in a subsequent header line has a space inserted after it, so as not to
5866 terminate the header.
5868 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5869 unnecessarily, used atoi() instead of strtol(), and didn't check the
5870 termination when getting file sizes from file names by regex.
5872 19. Completely re-implemented the support for maildirsize files, in the light
5873 of a number of problems with the previous contributed implementation
5874 (4.30/29). In particular:
5876 . If the quota is zero, the maildirsize file is maintained, but no quota is
5879 . If the maildir directory does not exist, it is created before any attempt
5880 to write a maildirsize file.
5882 . The quota value in the file is just a cache; if the quota is changed in
5883 the transport, the new value overrides.
5885 . A regular expression is available for excluding directories from the
5888 20. The autoreply transport checks the characters in options that define the
5889 message's headers; it allows continued headers, but it was checking with
5890 isspace() after an embedded newline instead of explicitly looking for a
5893 21. If all the "regular" hosts to which an address was routed had passed their
5894 expiry times, and had not reached their retry times, the address was
5895 bounced, even if fallback hosts were defined. Now Exim should go on to try
5898 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5899 equivalent code in the SMTP transport. Some hosts send humungous responses
5900 to HELO/EHLO, more than 1024 it seems.
5902 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5903 gives compiler warnings in some environments. I've now done it "properly",
5906 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5907 (because of problems with the built-in one) was declared to return uschar *
5908 instead of char *, causing compiler failure.
5910 25. Fixed a file descriptor leak when processing alias/forward files.
5912 26. Fixed a minor format string issue in dbfn.c.
5914 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5916 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5917 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5920 29. When a sender address is verified, it is cached, to save repeating the test
5921 when there is more than one recipient in a message. However, when the
5922 verification involves a callout, it is possible for different callout
5923 options to be set for different recipients. It is too complicated to keep
5924 track of this in the cache, so now Exim always runs a verification when a
5925 callout is required, relying on the callout cache for the optimization.
5926 The overhead is duplication of the address routing, but this should not be
5929 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5930 to be verified with callout=postmaster, and the main callout worked but the
5931 postmaster check failed, the verification correctly failed. However, if a
5932 subsequent RCPT command asked for sender verification *without* the
5933 postmaster check, incorrect caching caused this verification also to fail,
5936 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5937 it was not caching the DNS options (qualify_single, search_parents) that
5938 were used when the lookup failed. A subsequent lookup with different
5939 options therefore always gave the same answer, though there were cases
5940 where it should not have. (Example: a "domains = !$mx_any" option on a
5941 dnslookup router: the "domains" option is always processed without any
5942 widening, but the router might have qualify_single set.) Now Exim uses the
5943 cached value only when the same options are set.
5945 32. Added John Jetmore's "exipick" utility to the distribution.
5947 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5948 than a timeout (e.g. a certificate is required, and is not provided), an
5949 Exim server now closes the connection immediately. Previously it waited for
5950 the client to close - but if the client is SSL, it seems that they each
5951 wait for each other, leading to a delay before one of them times out.
5953 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5954 maintained 0.8.x compatibility because I don't think many are using it, and
5955 it is clearly obsolete.
5957 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5960 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5961 client certificate was expired. A simple patch fixes this, though I don't
5962 understand the full logic of why the verify callback is called multiple
5965 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5966 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5967 which causes problems with some clients (such as the Certicom SSL Plus
5968 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5969 disables the coutermeasure allowing Eudora to connect."
5971 38. Exim was not checking that a write() to a log file succeeded. This could
5972 lead to Bad Things if a log got too big, in particular if it hit a file
5973 size limit. Exim now panics and dies if it cannot write to a log file, just
5974 as it does if it cannot open a log file.
5976 39. Modified OS/Makefile-Linux so that it now contains
5978 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5980 The two -D definitions ensure that Exim is compiled with large file
5981 support, which makes it possible to handle log files that are bigger than
5984 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5985 instance) a domain was checked against a named list that involved a lookup,
5986 causing $domain_data to be set, then another domain was checked against the
5987 same list, then the first domain was re-checked, the value of $domain_data
5988 after the final check could be wrong. In particular, if the second check
5989 failed, it could be set empty. This bug probably also applied to
5992 41. The strip_trailing_dot option was not being applied to the address given
5993 with the -f command-line option.
5995 42. The code for reading a message's header from the spool was incrementing
5996 $received_count, but never initializing it. This meant that the value was
5997 incorrect (doubled) while delivering a message in the same process in which
5998 it was received. In the most common configuration of Exim, this never
5999 happens - a fresh exec is done - but it can happen when
6000 deliver_drop_privilege is set.
6002 43. When Exim logs an SMTP synchronization error - client data sent too soon -
6003 it now includes up to 150 characters of the unexpected data in the log
6006 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6007 and building data strings. The size of both of these buffers was 10 000
6008 bytes - far larger than anybody would *ever* want, thought I. Needless to
6009 say, somebody hit the limit. I have increased the maximum line length to
6010 20 000 and the maximum data length of concatenated lines to 100 000. I have
6011 also fixed two bugs, because there was no checking on these buffers. Tsk,
6012 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6013 buffer is too small.
6015 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6016 lsearch lookups. Now it does.
6018 46. When parsing a route_list item in a manualroute router, a fixed-length
6019 buffer was used for the list of hosts. I made this 1024 bytes long,
6020 thinking that nobody would ever have a list of hosts that long. Wrong.
6021 Somebody had a whole pile of complicated expansion conditions, and the
6022 string was silently truncated, leading to an expansion error. It turns out
6023 that it is easier to change to an unlimited length (owing to other changes
6024 that have happened since this code was originally written) than to build
6025 structure for giving a limitation error. The length of the item that
6026 expands into the list of hosts is now unlimited.
6028 47. The lsearch lookup could not handle data where the length of text line was
6029 more than 4095 characters. Such lines were truncated, leading to shortened
6030 data being returned. It should now handle lines of any length.
6032 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6033 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6036 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6037 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6038 no longer happens in comment lines. A list of replacements is now placed
6039 at the head of all of the source files, except those whose only change is
6040 to replace PERL_COMMAND in the very first #! line.
6042 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6043 messages on the queue, with a bottom-up merge sort, using code contributed
6044 by Michael Haardt. This should make operations like -bp somewhat faster on
6045 large queues. It won't affect queue runners, except when queue_run_in_order
6048 51. Installed eximstats 1.31 in the distribution.
6050 52. Added support for SRV lookups to the dnslookup router.
6052 53. If an ACL referred to $message_body or $message_body_end, the value was not
6053 reset for any messages that followed in the same SMTP session.
6055 54. The store-handling optimization for building very long strings was not
6056 differentiating between the different store pools. I don't think this
6057 actually made any difference in practice, but I've tidied it.
6059 55. While running the routers to verify a sender address, $sender_address
6060 was still set to the sender address. This is wrong, because when routing to
6061 send a bounce to the sender, it would be empty. Therefore, I have changed
6062 it so that, while verifying a sender address, $sender_address is set to <>.
6063 (There is no change to what happens when verifying a recipient address.)
6065 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6066 target A or AAAA records (if not already returned) without resetting the
6067 qualify_single or search_parents options of the DNS resolver. These are
6068 inappropriate in this case because the targets of MX and SRV records must
6069 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6070 target that, when qualified, matched something in the local domain. These
6071 two options are now turned off when doing these lookups.
6073 57. It seems that at least some releases of Reiserfs (which does not have the
6074 concept of a fixed number of inodes) returns zero and not -1 for the
6075 number of available inodes. This interacted badly with check_spool_inodes,
6076 which assumed that -1 was the "no such thing" setting. What I have done is
6077 to check that the total number of inodes is greater than zero before doing
6078 the test of how many are available.
6080 58. When a "warn" ACL statement has a log_message modifier, the message is
6081 remembered, and not repeated. This is to avoid a lot of repetition when a
6082 message has many recipients that cause the same warning to be written.
6083 However, Exim was preserving the list of already written lines for an
6084 entire SMTP session, which doesn't seem right. The memory is now reset if a
6085 new message is started.
6087 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6088 debugging output unless log_rewrite was also set.
6090 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6091 of (int)(handle) when we know that handle contains (void *)(-1).
6093 61. The Exim daemon panic-logs an error return when it closes the incoming
6094 connection. However "connection reset by peer" seems to be common, and
6095 isn't really an error worthy of noting specially, so that particular error
6098 62. When Exim is trying to find all the local interfaces, it used to panic and
6099 die if the ioctl to get the interface flags failed. However, it seems that
6100 on at least one OS (Solaris 9) it is possible to have an interface that is
6101 included in the list of interfaces, but for which you get a failure error
6102 for this call. This happens when the interface is not "plumbed" into a
6103 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6104 failure of the "get flags" call assumes that the interface is down.
6106 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6107 makes life easier for people who are doing arithmetic on fields extracted
6108 from dates, where you often get leading zeros that should not be
6109 interpreted as octal.
6111 64. Added qualify_domain to the redirect router, to override the global
6114 65. If a pathologically long header line contained very many addresses (the
6115 report of this problem mentioned 10 000) and each of them was rewritten,
6116 Exim could use up a very large amount of memory. (It kept on making new
6117 copies of the header line as it rewrote, and never released the old ones.)
6118 At the expense of a bit more processing, the header rewriting function has
6119 been changed so that it no longer eats memory in this way.
6121 66. The generation of the Received: header has been moved from the time that a
6122 message starts to be received, to the time that it finishes. The timestamp
6123 in the Received: header should now be very close to that of the <= log
6124 line. There are two side-effects of this change:
6126 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6127 logged header lines no longer include the local Received: line, because
6128 it has not yet been created. The same applies to a copy of the message
6129 that is returned to a non-SMTP sender when a message is rejected.
6131 (b) When a filter file is tested using -bf, no additional Received: header
6132 is added to the test message. After some thought, I decided that this
6135 This change does not affect the value of $received_for. It is still set
6136 after address rewriting, but before local_scan() is called.
6138 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6140 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6141 gave an unhelpful panic error message, and a defer error. I have managed to
6142 change this behaviour so that it now rejects any supplied certificate,
6143 which seems right, as the list of acceptable certificates is empty.
6145 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6146 gave an unhelpful defer error. I have not managed to make this reject any
6147 supplied certificates, but the error message it gives is "no certificate
6148 supplied", which is not helpful.
6150 70. exigrep's output now also includes lines that are not associated with any
6151 message, but which match the given pattern. Implemented by a patch from
6152 Martin Sluka, which also tidied up the Perl a bit.
6154 71. Recipient callout verification, like sender verification, was using <> in
6155 the MAIL FROM command. This isn't really the right thing, since the actual
6156 sender may affect whether the remote host accepts the recipient or not. I
6157 have changed it to use the actual sender in the callout; this means that
6158 the cache record is now keyed on a recipient/sender pair, not just the
6159 recipient address. There doesn't seem to be a real danger of callout loops,
6160 since a callout by the remote host to check the sender would use <>.
6161 [SEE ABOVE: changed after hitting problems.]
6163 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6164 temporary errors. However, in the case of such a code being given after
6165 the end of a data transmission (i.e. after ".") Exim was failing to write
6166 a retry record for the message. (Yes, there was some broken host that was
6167 actually sending 8xx at this point.)
6169 73. An unknown lookup type in a host list could cause Exim to panic-die when
6170 the list was checked. (An example that provoked this was putting <; in the
6171 middle of a list instead of at the start.) If this happened during a DATA
6172 ACL check, a -D file could be left lying around. This kind of configuration
6173 error no longer causes Exim to die; instead it causes a defer error. The
6174 incident is still logged to the main and panic logs.
6176 74. Buglet left over from Exim 3 conversion. The message "too many messages
6177 in one connection" was written to the rejectlog but not the mainlog, except
6178 when address rewriting (yes!) was being logged.
6180 75. Added write_rejectlog option.
6182 76. When a system filter was run not as root (that is, when system_filter_user
6183 was set), the values of the $n variables were not being returned to the
6184 main process; thus, they were not subsequently available in the $sn
6187 77. Added +return_path_on_delivery log selector.
6189 78. A connection timeout was being treated differently from recipients deferred
6190 when testing hosts_max_try with a message that was older than the host's
6191 retry timeout. (The host should not be counted, thus allowing all hosts to
6192 be tried at least once before bouncing.) This may have been the cause of an
6193 occasionally reported bug whereby a message would remain on the queue
6194 longer than the retry timeout, but would be bounced if a delivery was
6195 forced. I say "may" because I never totally pinned down the problem;
6196 setting up timeout/retry tests is difficult. See also the next item.
6198 79. The ultimate address timeout was not being applied to errors that involved
6199 a combination of host plus message (for example, a timeout on a MAIL
6200 command). When an address resolved to a number of possible hosts, and they
6201 were not all tried for each delivery (e.g. because of hosts_max_try), a
6202 message could remain on the queue longer than the retry timeout.
6204 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6207 81. Fixed an obscure SMTP outgoing bug which required at least the following
6208 conditions: (a) there was another message waiting for the same server;
6209 (b) the server returned 5xx to all RCPT commands in the first message so
6210 that the message was not completed; (c) the server dropped the connection
6211 or gave a negative response to the RSET that Exim sends to abort the
6212 transaction. The observed case was a dropped connection after DATA that had
6213 been sent in pipelining mode. That is, the server had advertised PIPELINING
6214 but was not implementing it correctly. The effect of the bug was incorrect
6215 behaviour, such as trying another host, and this could lead to a crash.
6221 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6222 and daemon.c were passed as pointers to ints; they should have been
6223 pointers to socklen_t variables (which are typically unsigned ints).
6225 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6228 3. Fixed a really odd bug that affected only the testing scheme; patching a
6229 certain fixed string in the binary changed the value of another string that
6230 happened to be identical to the end of the original first string.
6232 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6233 name", it returns that address as the IP address. On some operating
6234 systems (e.g. Solaris), it also passes back the IP address string as the
6235 "host name". However, on others (e.g. Linux), it passes back an empty
6236 string. Exim wasn't checking for this, and was changing the host name to an
6237 empty string, assuming it had been canonicalized.
6239 5. Although rare, it is permitted to have more than one PTR record for a given
6240 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6241 all the names associated with an address, because they do in Solaris.
6242 However, it seems that they do not in Linux for data that comes from the
6243 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6244 I found this out when I moved to a new Linux workstation and tried to run
6245 the Exim test suite.
6247 To get round this problem I have changed the code so that it now does its
6248 own call to the DNS to look up PTR records when searching for a host name.
6249 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6250 addresses that are only in /etc/hosts are still found.
6252 This behaviour is, however, controlled by an option called host_lookup_
6253 order, which defaults to "bydns:byaddr". If people want to use the other
6254 order, or indeed, just use one or the other means of lookup, they can
6255 specify it in this variable.
6257 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6258 some operating systems, this comes back from gethostbyaddr() as an empty
6259 string, and this is what Exim used to test for. However, it seems that in
6260 other systems, "." is yielded. Exim now tests for this case too.
6262 7. The values of check_spool_space and check_log_space are now held internally
6263 as a number of kilobytes instead of an absolute number of bytes. If a
6264 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6265 kilobyte. This means that much larger values can be stored.
6267 8. Exim monitor: an attempt to get the action menu when not actually pointing
6268 at a message produces an empty menu entitled "No message selected". This
6269 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6270 no entries in it ("Shell widget menu has zero width and/or height"). So I
6271 have added a single, blank menu entry in this case.
6273 9. Added ${quote_local_part.
6275 10. MIME decoding is now applied to the contents of Subject: header lines when
6278 11. Now that a reference to $sender_host_address automatically causes a reverse
6279 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6280 host lookup before query-style lookups in lists that might use this
6281 variable. This has therefore been abolished, and the "net-" prefix is no
6282 longer necessary for query-style lookups.
6284 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6285 have been a typo for LFLAGS, so it has been changed.
6287 13. The install script calls Exim with "-C /dev/null" in order to find the
6288 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6289 to be output. However, since Exim outputs its version number before the
6290 error, it didn't break the script. It just looked ugly. I fixed this by
6291 always allowing "-C /dev/null" if the caller is root.
6293 14. Ignore overlarge ACL variable number when reading spool file - insurance
6294 against a later release with more variables having written the file.
6296 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6297 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6298 this, as well as the form without the "IPv6" on the front.
6300 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6301 OS/Makefile-Darwin file.
6303 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6304 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6306 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6307 variable. However, some OS use ints for inodes. I've added cast to long int
6308 to get rid of the compiler warning.
6310 19. I had forgotten to lock out "/../" in configuration file names when
6311 ALT_CONFIG_PREFIX was set.
6313 20. Routers used for verification do not need to specify transports. However,
6314 if such a router generated a host list, and callout was configured, Exim
6315 crashed, because it could not find a port number from the (non-existent)
6316 transport. It now assumes port 25 in this circumstance.
6318 21. Added the -t option to exigrep.
6320 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6321 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6322 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6323 them is set, LOOKUP_LSEARCH is forced.
6325 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6327 24. Added sender and host information to the "rejected by local_scan()" log
6328 line; previously there was no indication of these.
6330 25. Added .include_if_exists.
6332 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6333 Linux. It turns out that not all file systems support this. Apparently some
6334 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6335 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6336 not supported on the file descriptor, is now ignored when Exim is trying to
6337 sync a directory. This applies only to Linux.
6339 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6341 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6342 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6345 29. Added support for maildirsize files from supplied patch (modified a bit).
6347 30. The use of :fail: followed by an empty string could lead Exim to respond to
6348 sender verification failures with (e.g.):
6350 550 Verification failed for <xxx>
6351 550 Sender verify failed
6353 where the first response line was missing the '-' that indicates it is not
6354 the final line of the response.
6356 31. The loop for finding the name of the user that called Exim had a hardwired
6357 limit of 10; it now uses the value of finduser_retries, which is used for
6358 all other user lookups.
6360 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6363 33. Exim was neglecting to zero errno before one call of strtol() when
6364 expanding a string and expecting an integer value. On some systems this
6365 resulted in spurious "integer overflow" errors. Also, it was casting the
6366 result into an int without checking.
6368 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6369 did not work. The code looks as if it has *never* worked, though it appears
6370 to have been documented since at least release 1.62. I have made it work.
6372 35. The "timeout_DNS" error in retry rules, also documented since at least
6373 1.62, also never worked. As it isn't clear exactly what this means, and
6374 clearly it isn't a major issue, I have abolished the feature by treating it
6375 as "timeout", and writing a warning to the main and panic logs.
6377 36. The display of retry rules for -brt wasn't always showing the error code
6380 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6381 timeout_connect_A, timeout_connect_MX.
6383 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6384 to the empty sender.
6386 39. The daemon was not analysing the content of -oX till after it had closed
6387 stderr and disconnected from the controlling terminal. This meant that any
6388 syntax errors were only noted on the panic log, and the return code from
6389 the command was 0. By re-arranging the code a little, I've made the
6390 decoding happen first, so such errors now appear on stderr, and the return
6391 code is 1. However, the actual setting up of the sockets still happens in
6392 the disconnected process, so errors there are still only recorded on the
6395 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6396 connections (as happens on some IP stacks) was logged at start up time as
6397 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6398 it from "IPv6 and IPv4", which means that two separate sockets are being
6401 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6402 says whether AF_INET or AF_INET6 was passed as an argument.
6404 42. Exiwhat output was messed up when time zones were included in log
6407 43. Exiwhat now gives more information about the daemon's listening ports,
6408 and whether -tls-on-connect was used.
6410 44. The "port" option of the smtp transport is now expanded.
6412 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6413 silently ignored. Now an error message is written to the main and panic
6416 46. There's a new ACL modifier called "logwrite" which writes to a log file
6417 as soon as it is encountered.
6419 47. Added $local_user_uid and $local_user_gid at routing time.
6421 48. Exim crashed when trying to verify a sender address that was being
6424 49. Exim was recognizing only a space character after ".include". It now also
6425 recognizes a tab character.
6427 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6428 extracting the relevant information from the specification. The man page no
6429 longer contains scrambled data for the -d option, and I've added a section
6430 at the front about calling Exim under different names.
6432 51. Added "extra_headers" argument to the "mail" command in filter files.
6434 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6437 53. Installed eximstats 1.29.
6439 54. Added transport_filter_timeout as a generic transport option.
6441 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6442 Cc: header lines. This was required by RFC 822, but it not required by RFC
6445 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6446 incoming messages that did not have them. Now it does so only if the
6447 message originates locally, that is, if there is no associated remote host
6448 address. When Resent- header lines are present, this applies to the Resent-
6449 lines rather than the non-Resent- lines.
6451 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6452 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6454 58. Messages for configuration errors now include the name of the main
6455 configuration file - useful now that there may be more than one file in a
6456 list (.included file names were always shown).
6458 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6459 for those rare installations that do not start the daemon as root or run it
6460 setuid root. I've cut out the call to initgroups() if the daemon is not
6463 60. The Exim user and group can now be bound into the binary as text strings
6464 that are looked up at the start of Exim's processing.
6466 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6468 62. Added $mailstore_basename variable.
6470 63. Installed patch to sieve.c from Michael Haardt.
6472 64. When Exim failed to open the panic log after failing to open the main log,
6473 the original message it was trying to log was written to stderr and debug
6474 output, but if they were not available (the usual case in production), it
6475 was lost. Now it is written to syslog before the two lines that record the
6476 failures to open the logs.
6478 65. Users' Exim filters run in subprocesses under the user's uid. It is
6479 possible for a "deliver" command or an alias in a "personal" command to
6480 provoke an address rewrite. If logging of address rewriting is configured,
6481 this fails because the process is not running as root or exim. There may be
6482 a better way of dealing with this, but for the moment (because 4.30 needs
6483 to be released), I have disabled address rewrite logging when running a
6484 filter in a non-root, non-exim process.
6490 1. The buildconfig auxiliary program wasn't quoting the value set for
6491 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6492 not defined. This bug was masked in 4.22 by the effect that was fixed in
6495 2. Some messages that were rejected after a message id was allocated were
6496 shown as "incomplete" by exigrep. It no longer does this for messages that
6497 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6499 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6500 have allow_domain_literals set, the ID did not get logged in the <= line.
6501 Domain literals are now always recognized in Message-ID: header lines.
6503 4. The first argument for a ${extract expansion item is the key name or field
6504 number. Leading and trailing spaces in this item were not being ignored,
6505 causing some misleading effects.
6507 5. When deliver_drop_privilege was set, single queue runner processes started
6508 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6509 same command in the process it spins off) were not dropping privilege.
6511 6. When the daemon running as "exim" started a queue runner, it always
6512 re-executed Exim in the spun-off process. This is a waste of effort when
6513 deliver_drop_privilege is set. The new process now just calls the
6514 queue-runner function directly.
6520 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6523 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6524 ignored. Though the use of -oP was forcing the writing of a pid file, it
6525 was always written to the default place.
6527 3. If the message "no IP address found for host xxxx" is generated during
6528 incoming verification, it is now followed by identification of the incoming
6529 connection (so you can more easily find what provoked it).
6531 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6533 5. Added some features to "harden" Exim a bit more against certain attacks:
6535 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6536 be put in Local/Makefile. This is like the never_users runtime option,
6537 but it cannot be overridden. The default setting is "root".
6539 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6540 prefix string with which any file named in a -C command line option
6543 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6544 is retained for -C and -D only if the caller of Exim is root. Without
6545 it, the exim user may also use -C and -D and retain privilege.
6547 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6548 command line option is disabled.
6550 6. Macro names set by the -D option must start with an upper case letter, just
6551 like macro names defined in the configuration file.
6553 7. Added "dereference=" facility to LDAP.
6555 8. Two instances of the typo "uknown" in the source files are fixed.
6557 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6558 the Configure-Makefile script screwed up while processing it.
6560 10. Incorporated PCRE 4.4.
6562 11. The SMTP synchronization check was not operating right at the start of an
6563 SMTP session. For example, it could not catch a HELO sent before the client
6564 waited for the greeting. There is now a check for outstanding input at the
6565 point when the greeting is written. Because of the duplex, asynchronous
6566 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6567 way, but not yet received, when the check is performed.
6569 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6570 on TCP/IP sockets, because this apparently causes some broken clients to
6573 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6574 unchanged) from the Cygwin maintainer.
6576 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6577 was supported instead of testing for mbx. Effectively a typo.
6579 15. The spa authenticator server code was not checking that the input it
6580 received was valid base64.
6582 16. The debug output line for the "set" modifier in ACLs was not showing the
6583 name of the variable that was being set.
6585 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6587 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6588 lookup on its own. Something else was needed to trigger the lookup. For
6589 example, a match in host_lookup or the need for a host name in a host list.
6590 Now, if $sender_host_name is referenced and the host name has not yet been
6591 looked up, a lookup is performed. If the lookup fails, the variable remains
6592 empty, and $host_lookup_failed is set to "1".
6594 19. Added "eqi" as a case-independent comparison operator.
6596 20. The saslauthd authentication condition could segfault if neither service
6597 nor realm was specified.
6599 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6600 error message that was logged was misleading, and incoming SMTP
6601 connections were dropped. The message is now more accurate, and temporary
6602 errors are given to SMTP connections.
6604 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6605 (except RSET) with a 421 error, until QUIT is received. However, it was
6606 failing to send a response to QUIT.
6608 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6609 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6610 ACL is now run after the helo_try_verify_hosts code.
6612 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6613 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6614 case-independent, so other case variants are also recognized). Apparently
6615 some systems use these upper case variants.
6617 25. If more than two messages were waiting for the same host, and a transport
6618 filter was specified for the transport, Exim sent two messages over the
6619 same TCP/IP connection, and then failed with "socket operation on non-
6620 socket" when it tried to send the third.
6622 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6624 27. The extern definition of crypt16() in expand.c was not being excluded when
6625 the OS had its own crypt16() function.
6627 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6628 as a preferred synonym for return_size_limit, both as an option and as an
6631 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6633 30. Changed the default configuration ACL to relax the local part checking rule
6634 for addresses that are not in any local domains. For these addresses,
6635 slashes and pipe symbols are allowed within local parts, but the sequence
6636 /../ is explicitly forbidden.
6638 31. SPA server authentication was not clearing the challenge buffer before
6641 32. log_message in a "warn" ACL statement was writing to the reject log as
6642 well as to the main log, which contradicts the documentation and doesn't
6643 seem right (because no rejection is happening). So I have stopped it.
6645 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6646 However, I am unable to do any testing of this.
6648 34. Fixed an infelicity in the appendfile transport. When checking directories
6649 for a mailbox, to see if any needed to be created, it was accidentally
6650 using path names with one or more superfluous leading slashes; tracing
6651 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6653 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6654 contents are added to the log line that is written for every discarded
6655 recipient. (Previously a log_message setting was ignored.)
6657 36. The ${quote: operator now quotes the string if it is empty.
6659 37. The install script runs exim in order to find its version number. If for
6660 some reason other than non-existence or emptiness, which it checks, it
6661 could not run './exim', it was installing it with an empty version number,
6662 i.e. as "exim-". This error state is now caught, and the installation is
6665 38. An argument was missing from the function that creates an error message
6666 when Exim fails to connect to the socket for saslauthd authentication.
6667 This could cause Exim to crash, or give a corrupted message.
6669 39. Added isip, isip4, and isip6 to ${if conditions.
6671 40. The ACL variables $acl_xx are now saved with the message, and can be
6672 accessed later in routers, transports, and filters.
6674 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6675 strings in the file are not string-expanded.
6677 42. If a MAIL command specified a SIZE value that was too large to fit into an
6678 int variable, the check against message_size_limit failed. Such values are
6679 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6680 day this will have to be increased, but I don't think I want to be around
6681 when emails are that large.
6688 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6689 iconv() is not standard in FreeBSD.
6691 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6692 IPv6 enabled. The observed symptom was a segmentation fault on return from
6693 the function os_common_find_running_interfaces() in src/os.c.
6695 3. In the check_special_case() function in daemon.c I had used "errno" as an
6696 argument name, which causes warnings on some systems. This was basically a
6697 typo, since it was named "eno" in the comments!
6699 4. The code that waits for the clock to tick (at a resolution of some fraction
6700 of a second) so as to ensure message-id uniqueness was always waiting for
6701 at least one whole tick, when it could have waited for less. [This is
6702 almost certainly not relevant at current processor speeds, where it is
6703 unlikely to ever wait at all. But we try to future-proof.]
6705 5. The function that sleeps for a time interval that includes fractions of a
6706 second contained a race. It did not block SIGALRM between setting the
6707 timer, and suspending (a couple of lines later). If the interval was short
6708 and the sigsuspend() was delayed until after it had expired, the suspension
6709 never ended. On busy systems this could lead to processes getting stuck for
6712 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6713 process, before it forks any delivery processes. The open lookup caching
6714 mechanism meant that the open file or database connection was passed into
6715 the delivery process. The problem was that delivery processes always tidy
6716 up cached lookup data. This could cause a problem for the next delivery
6717 process started by the queue runner, because the external queue runner
6718 process does not know about the closure. So the next delivery process
6719 still has data in the lookup cache. In the case of a file lookup, there was
6720 no problem because closing a file descriptor in a subprocess doesn't affect
6721 the parent. However, if the lookup was caching a connection to a database,
6722 the connection was closed, and the second delivery process was likely to
6723 see errors such as "PGSQL: query failed: server closed the connection
6724 unexpectedly". The problem has been fixed by closing all cached lookups
6725 in a queue runner before running a delivery process.
6727 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6728 seem to have the "const" qualifier which it has on other OS. I've
6731 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6732 *of the same type* (client or server) with the same public name that an
6733 error should be diagnosed.
6735 9. When Exim looked up a host name for an IP address, but failed to find the
6736 original IP address when looking up the host name (a safety check), it
6737 output the message "<ip address> does not match any IP for NULL", which was
6738 confusing, to say the least. The bug was that the host name should have
6739 appeared instead of "NULL".
6741 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6742 user that is built into the binary, and the -C or -D options is used, root
6743 privilege is dropped before the configuration file is read. In addition,
6744 logging is switched to stderr instead of the normal log files. If the
6745 configuration then re-defines the Exim user, the unprivileged environment
6746 is probably not what is expected, so Exim logs a panic warning message (but
6749 However, if deliver_drop_privilege is set, the unprivileged state may well
6750 be exactly what is intended, so the warning has been cut out in that case,
6751 and Exim is allowed to try to write to its normal log files.
6757 1. smtp_return_error_details was not giving details for temporary sender
6758 or receiver verification errors.
6760 2. Diagnose a configuration error if two authenticators have the same public
6763 3. Exim used not to create the message log file for a message until the first
6764 delivery attempt. This could be confusing when incoming messages were held
6765 for policy or load reasons. The message log file is now created at the time
6766 the message is received, and an initial "Received" line is written to it.
6768 4. The automatically generated man page for command line options had a minor
6769 bug that caused no ill effects; however, a more serious problem was that
6770 the procedure for building the man page automatically didn't always
6771 operate. Consequently, release 4.20 contains an out-of-date version. This
6772 shouldn't happen again.
6774 5. When building Exim with embedded Perl support, the script that builds the
6775 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6776 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6778 6. The freeze_tell option was not being used for messages that were frozen on
6779 arrival, either by an ACL or by local_scan().
6781 7. Added the smtp_incomplete_transaction log selector.
6783 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6784 was accepting AUTH without a new EHLO.
6786 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6787 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6790 10. Logging of TCP/IP connections (when configured) now happens in the main
6791 daemon process instead of the child process, so that the TCP/IP connection
6792 count is more accurate (but it can never be perfect).
6794 11. The use of "drop" in a nested ACL was not being handled correctly in the
6795 outer ACL. Now, if condition failure induced by the nested "drop" causes
6796 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6797 or "require"), the connection is dropped.
6799 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6800 that yield "discard" can now be used with an "accept" or a "discard" verb,
6801 but an error is generated for any others (because I can't see a useful way
6802 to define what should happen).
6804 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6805 are now processed in the same way as lines in the Exim configuration file.
6806 In particular, continuation lines are supported.
6808 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6810 15. Added -ti meaning -t -i.
6812 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6813 domains, and warn by logging if others are found.
6815 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6816 returned from GIFCONF when Exim is trying to find the list of interfaces on
6817 a host. The code in os.c has been modified to copy each ifreq to an aligned
6818 structure in all cases.
6820 Also, in some cases, the returned ifreq's were being copied to a 'struct
6821 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6822 means the last couple of bytes of an IPv6 address could be chopped if the
6823 ifreq contained only a normal sockaddr (14 bytes storage).
6825 18. Named domain lists were not supported in the hosts_treat_as_local option.
6826 An entry such as +xxxx was not recognized, and was treated as a literal
6829 19. Ensure that header lines added by a DATA ACL are included in the reject log
6830 if the ACL subsequently rejects the message.
6832 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6833 MD5 (which is deprecated).
6835 21. When testing a filter file using -bf, Exim was writing a message when it
6836 took the sender from a "From " line in the message, but it was not doing so
6837 when it took $return_path from a Return-Path: header line. It now does.
6839 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6840 with a valid header line field name (a series of printing characters
6841 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6843 23. Changed "disc" in the source to "disk" to conform to the documentation and
6844 the book and for uniformity.
6846 24. Ignore Sendmail's -Ooption=value command line item.
6848 25. When execve() failed while trying to run a command in a pipe transport,
6849 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6850 could be confused with a return value of 69 from the command itself. This
6851 has been changed to 127, the value the shell returns if it is asked to run
6852 a non-existent command. The wording for the related log line suggests a
6853 non-existent command as the problem.
6855 26. If received_header_text expands to an empty string, do not add a Received:
6856 header line to the message. (Well, it adds a token one on the spool, but
6857 marks it "old" so that it doesn't get used or transmitted.)
6859 27. Installed eximstats 1.28 (addition of -nt option).
6861 28. There was no check for failure on the call to getsockname() in the daemon
6862 code. This can fail if there is a shortage of resources on the system, with
6863 ENOMEM, for example. A temporary error is now given on failure.
6865 29. Contrary to the C standard, it seems that in some environments, the
6866 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6867 program. Exim now does this explicitly; it affects the formatting of
6868 timestamps using strftime().
6870 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6871 complaints. I've now initialized all the variables, to avoid this.
6873 32. Header lines added by a system filter were not being "seen" during
6874 transport-time rewrites.
6876 33. The info_callback() function passed to OpenSSL is set up with type void
6877 (*)(SSL *, int, int), as described somewhere. However, when calling the
6878 function (actually a macro) that sets it up, the type void(*)() is
6879 expected. I've put in a cast to prevent warnings from picky compilers.
6881 34. If a DNS black list lookup found a CNAME record, but there were no A
6882 records associated with the domain it pointed at, Exim crashed.
6884 35. If a DNS black list lookup returned more than one A record, Exim ignored
6885 all but the first. It now scans all returned addresses if a particular IP
6886 value is being sought. In this situation, the contents of the
6887 $dnslist_value variable are a list of all the addresses, separated by a
6890 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6891 to accept a host name and all its aliases if the forward lookup for any of
6892 them yielded the IP address of the incoming connection. Now it accepts only
6893 those names whose forward lookup yields the correct IP address. Any other
6894 names are discarded. This closes a loophole whereby a rogue DNS
6895 administrator could create reverse DNS records to break through a
6896 wildcarded host restriction in an ACL.
6898 37. If a user filter or a system filter that ran in a subprocess used any of
6899 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6900 the wrong values were passed to the pipe command ($thisaddress had the
6901 value of $0, $0 had the value of $1, etc). This bug was introduced by
6902 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6904 38. Improved the line breaking for long SMTP error messages from ACLs.
6905 Previously, if there was no break point between 40 and 75 characters, Exim
6906 left the rest of the message alone. Two changes have been made: (a) I've
6907 reduced the minimum length to 35 characters; (b) if it can't find a break
6908 point between 35 and 75 characters, it looks ahead and uses the first one
6909 that it finds. This may give the occasional overlong line, but at least the
6910 remaining text gets split now.
6912 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6913 file descriptors might be low, and that setting 1000 would always raise it.
6914 It turns out that in some environments, the limit is already over 1000 and
6915 that lowering it causes trouble. So now Exim takes care not to decrease it.
6917 40. When delivering a message, the value of $return_path is set to $sender_
6918 address at the start of routing (routers may change the value). By an
6919 oversight, this default was not being set up when an address was tested by
6920 -bt or -bv, which affected the outcome if any router or filter referred to
6923 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6924 log only when "message" or "log_message" are set. However, if one of the
6925 conditions was an address verification, or a call to a nested ACL, the
6926 messages generated by the underlying test were being passed through. This
6927 no longer happens. The underlying message is available in $acl_verify_
6928 message for both "message" and "log_message" expansions, so it can be
6929 passed through if needed.
6931 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6932 new expansion $bh_ to give the encoded byte string without charset
6933 translation. Translation happens only if iconv() is available; HAVE_ICONV
6934 indicates this at build time. HEADERS_CHARSET gives the charset to
6935 translate to; headers_charset can change it in the configuration, and
6936 "headers charset" can change it in an individual filter file.
6938 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6939 that creates RFC 2047 encoded "words" labels them as that charset instead
6940 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6941 expansion operator; (ii) when Exim creates a From: line for a local
6942 message; (iii) when a header line is rewritten to include a "phrase" part.
6944 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6945 buggy, causing it to skip the first lines of messages whose message ID
6946 ended in 'D'. This would not have bitten before Exim release 4.14, because
6947 message IDs were unlikely to end in 'D' before then. The effect was to have
6948 incorrect size information for certain domains.
6950 45. #include "config.h" was missing at the start of the crypt16.c module. This
6951 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6954 46. If there was a timeout during a "random" callout check, Exim treated it as
6955 a failure of the random address, and carried on sending RSET and the real
6956 address. If the delay was just some slowness somewhere, the response to the
6957 original RCPT would be taken as a response to RSET and so on, causing
6958 mayhem of various kinds.
6960 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6961 when I implemented it. It didn't allow for the fact that some option values
6962 may legitimately be negative (e.g. size_addition), and it didn't even do
6963 the right test for positive values.
6965 48. Domain names in DNS records are case-independent. Exim always looks them up
6966 in lower case. Some resolvers return domain names in exactly the case they
6967 appear in the zone file, that is, they may contain uppercase letters. Not
6968 all resolvers do this - some return always lower case. Exim was treating a
6969 change of case by a resolver as a change of domain, similar to a widening
6970 of a domain abbreviation. This triggered its re-routing code and so it was
6971 trying to route what was effectively the same domain again. This normally
6972 caused routing to fail (because the router wouldn't handle the domain
6973 twice). Now Exim checks for this case specially, and just changes the
6974 casing of the domain that it ultimately uses when it transmits the message
6977 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6980 50. If a filter generated a file delivery with a non-absolute name (possible if
6981 no home directory exists for the router), the forbid_file option was not
6984 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6985 the existing equality matching.
6987 52. Exim was using ints instead of ino_t variables in some places where it was
6988 dealing with inode numbers.
6990 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6991 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6992 called TMPDIR, and if it finds it is different, it changes its value.
6994 54. The smtp_printf() function is now made available to local_scan() so
6995 additional output lines can be written before returning. There is also an
6996 smtp_fflush() function to enable the detection of a dropped connection.
6997 The variables smtp_input and smtp_batched_input are exported to
7000 55. Changed the default runtime configuration: the message "Unknown user"
7001 has been removed from the ACL, and instead placed on the localuser router,
7002 using the cannot_route_message feature. This means that any verification
7003 failures that generate their own messages won't get overridden. Similarly,
7004 the "Unrouteable address" message that was in the ACL for unverifiable
7005 relay addresses has also been removed.
7007 56. Added hosts_avoid_esmtp to the smtp transport.
7009 57. The exicyclog script was not checking for the esoteric option
7010 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7011 will work only if exicyclog is run under the appropriate euid.
7013 58. Following a discussion on the list, the rules by which Exim recognises line
7014 endings on incoming messages have been changed. The -dropcr and drop_cr
7015 options are now no-ops, retained only for backwards compatibility. The
7016 following line terminators are recognized: LF CRLF CR. However, special
7017 processing applies to CR:
7019 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7020 nor a local message in the state where . is a terminator.
7022 (ii) If a bare CR is encountered in a header line, an extra space is added
7023 after the line terminator so as not to end the header. The reasoning
7024 behind this is that bare CRs in header lines are most likely either
7025 to be mistakes, or people trying to play silly games.
7027 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7028 was being incorrectly given as 18 bytes larger than it should have been.
7029 This is a VOB (very old bug).
7031 60. This may never have affected anything current, but just in case it has:
7032 When the local host is found other than at the start of a list of hosts,
7033 the local host, those with the same MX, and any that follow, are discarded.
7034 When the list in question was part of a longer list of hosts, the following
7035 hosts (not currently being processed) were also being discarded. This no
7036 longer happens. I'm not sure if this situation could ever has previously
7039 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7042 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7043 header. This is something that is recommended in a new Internet Draft, and
7044 is something that is documented as being done by Sendmail. There are two
7045 possible values. For messages generated by the autoreply transport, Exim
7048 Auto-Submitted: auto-replied
7050 whereas for all other generated messages (e.g. bounces) it adds
7052 Auto-Submitted: auto-generated
7054 63. The "personal" condition in filters now includes a test for the
7055 Auto-Submitted: header. If it contains the string "auto-" the message it
7056 not considered personal.
7058 64. Added rcpt_include_affixes as a generic transport option.
7060 65. Added queue_only_override (default true).
7062 66. Added the syslog_duplication option.
7064 67. If what should have been the first header line of a message consisted of
7065 a space followed by a colon, Exim was mis-interpreting it as a header line.
7066 It isn't of course - it is syntactically invalid and should therefore be
7067 treated as the start of the message body. The misbehaviour could have
7068 caused a number of strange effects, including loss of data in subsequent
7069 header lines, and spool format errors.
7071 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7072 client host had authenticated. This control can now be exercised by an ACL
7073 for more flexibility.
7075 69. By default, callouts do not happen when testing with -bh. There is now a
7076 variant, -bhc, which does actually run the callout code, including
7077 consulting and updating the callout cache.
7079 70. Added support for saslauthd authentication, courtesy of Alexander
7082 71. If statvfs() failed on the spool or log directories while checking their
7083 size for availability, Exim confusingly gave the error "space shortage".
7084 Furthermore, in debugging mode it crashed with a floating point exception.
7085 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7086 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7087 serious problem, Exim now writes to the main and panic logs when this
7088 happens, with details of the failure. It then refuses to accept the
7089 incoming message, giving the message "spool directory problem" or "log
7090 directory problem" with a 421 code for SMTP messages.
7092 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7093 0, 1, and 2 exist, because some OS complain for execs without them (see
7094 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7095 descriptors. However, the code omitted to check that the open succeeded,
7096 causing mysterious errors if for some reason the permissions on /dev/null
7097 got screwed. Now Exim writes a message to the main and panic logs, and
7098 bombs out if it can't open /dev/null.
7100 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7101 interact so that it is all more flexible. It is supposed to remain
7102 backwards compatible. Also added extra_local_interfaces.
7104 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7105 to bomb out with an assertion failure - to the client this appears as a
7106 connection drop. This problem occurs in the part of the code that was taken
7107 from the Samba project. Fortunately, the assertion is in a very simple
7108 function, so I have fixed this by reproducing the function inline in the
7109 one place where it is called, and arranging for authentication to fail
7110 instead of killing the process with assert().
7112 75. The SPA client code was not working when the server requested OEM rather
7113 than Unicode encoding.
7115 76. Added code to make require_files with a specific uid setting more usable in
7116 the case where statting the file as root fails - usually a non-root-mounted
7117 NFS file system. When this happens and the failure is EACCES, Exim now
7118 forks a subprocess and does the per-uid checking as the relevant uid.
7120 77. Added process_log_path.
7122 78. If log_file_path was not explicitly set, a setting of check_log_space or
7123 check_log_inodes was ignored.
7125 79. If a space check for the spool or log partitions fails, the incident is now
7126 logged. Of course, in the latter case the data may get lost...
7128 80. Added the %p formatting code to string_format() so that it can be used to
7129 print addresses in debug_print(). Adjusted all the address printing in the
7130 debugging in store.c to use %p rather than %d.
7132 81. There was a concern that a line of code in smtp_in.c could overflow a
7133 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7134 initially expressed, the concern was not well-founded, because trailing
7135 spaces are removed early. However, if the trailing spaces were followed by
7136 a NULL, they did not get removed, so the overflow was possible. Two fixes
7139 (a) I re-wrote the offending code in a cleaner fashion.
7140 (b) If an incoming SMTP command contains a NULL character, it is rejected
7143 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7144 runs initgroups(), so that if the Exim user is in any additional groups,
7145 they will be used during message reception.
7151 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob