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.
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.
193 HS/03 Malware Avast: Better match the Avast multiline protocol. Add
194 "pass_unscanned". Only tmpfails from the scanner are written to
195 the paniclog, as they may require admin intervention (permission
196 denied, license issues). Other scanner errors (like decompression
197 bombs) do not cause a paniclog entry.
199 JH/36 Fix reinitialisation of DKIM logging variable between messages.
200 Previously it was possible to log spurious information in receive log
203 JH/37 Bug 2255: Revert the disable of the OpenSSL session caching. This
204 triggered odd behaviour from Outlook Express clients.
206 PP/03 Add util/renew-opendmarc-tlds.sh script for safe renewal of public
213 JH/01 Rework error string handling in TLS interface so that the caller in
214 more cases is responsible for logging. This permits library-sourced
215 string to be attached to addresses during delivery, and collapses
216 pairs of long lines into single ones.
218 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
219 during configuration. Wildcards are allowed and expanded.
221 JH/02 Rework error string handling in DKIM to pass more info back to callers.
222 This permits better logging.
224 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
225 do not close it down and have the child transport start it up again on
226 the passed-on TCP connection. Instead, proxy the child (and any
227 subsequent ones) for TLS via a unix-domain socket channel. Logging is
228 affected: the continued delivery log lines do not have any DNSSEC, TLS
229 Certificate or OCSP information. TLS cipher information is still logged.
231 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
232 identical IP addresses on different listening ports. Will also affect
235 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
236 add noisy ifdef guards to special-case this sillyness.
237 Patch from Bernd Kuhls.
239 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
240 than 255 are no longer allowed.
242 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
243 Disable the session-cache too, which might reduce our load. Since we
244 currrectly use a new context for every connection, both as server and
245 client, there is no benefit for these.
246 GnuTLS appears to not support tickets server-side by default (we don't
247 call gnutls_session_ticket_enable_server()) but client side is enabled
248 by default on recent versions (3.1.3 +) unless the PFS priority string
251 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
252 <https://reproducible-builds.org/specs/source-date-epoch/>.
254 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
255 the check for any unsuccessful recipients did not notice the limit, and
256 erroneously found still-pending ones.
258 JH/08 Pipeline CHUNKING command and data together, on kernels that support
259 MSG_MORE. Only in-clear (not on TLS connections).
261 JH/09 Avoid using a temporary file during transport using dkim. Unless a
262 transport-filter is involved we can buffer the headers in memory for
263 creating the signature, and read the spool data file once for the
264 signature and again for transmission.
266 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
267 4.77 as the kernel support then wasn't solid, having issues in 64bit
268 mode. Now, it's been long enough. Add support for FreeBSD also.
270 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
271 case where the routing stage had gathered several addresses to send to
272 a host before calling the transport for the first, we previously failed
273 to close down TLS in the old transport process before passing the TCP
274 connection to the new process. The new one sent a STARTTLS command
275 which naturally failed, giving a failed delivery and bloating the retry
276 database. Investigation and fix prototype from Wolfgang Breyha.
278 JH/12 Fix check on SMTP command input synchronisation. Previously there were
279 false-negatives in the check that the sender had not preempted a response
280 or prompt from Exim (running as a server), due to that code's lack of
281 awareness of the SMTP input buffering.
283 PP/04 Add commandline_checks_require_admin option.
284 Exim drops privileges sanely, various checks such as -be aren't a
285 security problem, as long as you trust local users with access to their
286 own account. When invoked by services which pass untrusted data to
287 Exim, this might be an issue. Set this option in main configuration
288 AND make fixes to the calling application, such as using `--` to stop
291 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
292 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
293 responses to those, into a single TLS record each way (this usually means
294 a single packet). As a side issue, smtp_enforce_sync now works on TLS
297 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
298 affects you only if you're dancing at the edge of the param size limits.
299 If you are, and this message makes sense to you, then: raise the
300 configured limit or use OpenSSL 1.1. Nothing we can do for older
303 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
304 cmdline element to get the documented default one. Previously it was
307 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
308 are unsuable for non-message-specific reasons - eg. connection timeout,
309 banner-time rejection.
311 JH/16 Fix logging of delivery remote port, when specified by router, under
314 PP/06 Repair manualroute's ability to take options in any order, even if one
315 is the name of a transport.
318 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
320 JH/17 Change the list-building routines interface to use the expanding-string
321 triplet model, for better allocation and copying behaviour.
323 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
324 Previously it was constructed the first time a possibly-matching string
325 was met in the configuration file input during startup; now it is done
328 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
329 compatible one, to avoid the (poorly documented) possibility of a config
330 file in the working directory redirecting the DB files, possibly correpting
331 some existing file. CVE-2017-10140 assigned for BDB.
333 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
334 cache-hot. Previously, although the result was properly cached, the
335 initial verify call returned a defer.
337 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
338 the main verify for receipient in uncached-mode.
340 JH/22 Retire historical build files to an "unsupported" subdir. These are
341 defined as "ones for which we have no current evidence of testing".
343 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
344 if present. Previously it was ignored.
346 JH/24 Start using specified-initialisers in C structure init coding. This is
347 a C99 feature (it's 2017, so now considered safe).
349 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
350 if was a fixed-sized field and bitmask ops via macros; it is now more
353 PP/07 GitHub PR 56: Apply MariaDB build fix.
354 Patch provided by Jaroslav Škarvada.
356 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
357 during Coverity cleanups [4.87 JH/47]
358 Diagnosis and fix provided by Michael Fischer v. Mollard.
360 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
361 the right size to place the terminating semicolon on its own folded
362 line, the header hash was calculated to an incorrect value thanks to
363 the (relaxed) space the fold became.
365 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
366 and confused the parent.
368 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
369 which could crash as a result. This could lead to undeliverable messages.
371 JH/28 Logging: "next input sent too soon" now shows where input was truncated
374 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
375 matters on fast-turnover and PID-randomising systems, which were getting
376 out-of-order delivery.
378 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
379 a possibly-overlapping copy. The symptom was that "Remote host closed
380 connection in response to HELO" was logged instead of the actual 4xx
383 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
384 Previously only that bufferd was discarded, resulting in SYMTP command
387 JH/32 DKIM: when a message has multiple signatures matching an identity given
388 in dkim_verify_signers, run the dkim acl once for each. Previously only
389 one run was done. Bug 2189.
391 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
392 "panic the current process" to "deliberately defer". The panic log is
393 still written with the problem list name; the mail and reject logs now
394 get a temp-reject line for the message that was being handled, saying
395 something like "domains check lookup or other defer". The SMTP 451
396 message is still "Temporary local problem".
398 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
399 A crafted sequence of BDAT commands could result in in-use memory beeing
400 freed. CVE-2017-16943.
402 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
403 from SMTP input. Previously it was always done; now only done for DATA
404 and not BDAT commands. CVE-2017-16944.
406 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
407 to the message (such as an overlong header line). Previously this was
408 not done and we did not exit BDAT mode. Followon from the previous item
409 though a different problem.
415 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
416 than -2003 did; needs libidn2 in addition to libidn.
418 JH/02 The path option on a pipe transport is now expanded before use.
420 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
421 Patch provided by "Björn", documentation fix added too.
423 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
424 missing a wire-to-host endian conversion.
426 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
427 close after a BDAT command line could be taken as a following command,
428 giving a synch failure. Fix by only checking for synch immediately
429 before acknowledging the chunk.
431 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
432 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
433 macro. Patches provided by Josh Soref.
435 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
436 Previously we did not; the RFC seems ambiguous and VRFY is not listed
437 by IANA as a service extension. However, John Klensin suggests that we
440 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
441 the dkim code may be unix-mode line endings rather than smtp wire-format
442 CRLF, so prepend a CR to any bare LF.
444 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
445 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
447 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
448 the first were themselves being wrongly included in the feed into dkim
449 processing; with most chunk sizes in use this resulted in an incorrect
450 body hash calculated value.
452 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
453 DKIM signature block, for verification. Although advised against by
454 standards it is specifically not ruled illegal.
456 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
458 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
459 missing a body hash (the bh= tag).
461 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
462 It seems that HAProxy sends the Proxy Protocol information in clear and
463 only then does a TLS startup, so do the same.
465 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
466 TCP connections (such as for Spamd) unless the daemon successfully set
467 Fast Open mode on its listening sockets. This fixes breakage seen on
468 too-old kernels or those not configured for Fast Open, at the cost of
469 requiring both directions being enabled for TFO, and TFO never being used
470 by non-daemon-related Exim processes.
472 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
473 endings, at least on the first header line. Try to canonify any that get
474 past that check, despite the cost.
476 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
477 now limited to an arbitrary five deep, while parsing addresses with the
478 strip_excess_angle_brackets option enabled.
480 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
481 instead leave the unprompted TLS handshake in socket buffer for the
482 TLS library to consume.
484 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
486 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
488 JH/16 Drop variables when they go out of scope. Memory management drops a whole
489 region in one operation, for speed, and this leaves assigned pointers
490 dangling. Add checks run only under the testsuite which checks all
491 variables at a store-reset and panics on a dangling pointer; add code
492 explicitly nulling out all the variables discovered. Fixes one known
493 bug: a transport crash, where a dangling pointer for $sending_ip_address
494 originally assigned in a verify callout, is re-used.
496 PP/06 Drop '.' from @INC in various Perl scripts.
498 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
500 PP/08 Reduce a number of compilation warnings under clang; building with
501 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
502 should be warning-free.
504 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
506 HS/01 Fix portability problems introduced by PP/08 for platforms where
507 realloc(NULL) is not equivalent to malloc() [SunOS et al].
509 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
510 chunk. This allows us to accept broken chunked messages. We need a more
511 general solution here.
513 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
514 already-broken messages in the queue.
516 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
518 JH/19 Fix reference counting bug in routing-generated-address tracking.
524 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
525 supports it and a size is available (ie. the sending peer gave us one).
527 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
528 years of being deprecated). The replacements are the ACLs
529 acl_smtp_mime and acl_not_smtp_mime.
531 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
532 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
533 or even an in-clear connection were permitted. Now, if the host lookup
534 was dnssec and dane was requested then the host is only used if the
535 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
536 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
537 if one fails this test.
538 This means that a poorly-configured remote DNS will make it incommunicado;
539 but it protects against a DNS-interception attack on it.
541 JH/04 Bug 1810: make continued-use of an open smtp transport connection
542 non-noisy when a race steals the message being considered.
544 JH/05 If main configuration option tls_certificate is unset, generate a
545 self-signed certificate for inbound TLS connections.
547 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
548 in rewrites and routers.
550 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
551 and logged a warning sing 4.83; now they are a configuration file error.
553 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
554 (lacking @domain). Apply the same qualification processing as RCPT.
556 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
558 JH/10 Support ${sha256:} applied to a string (as well as the previous
561 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
562 a cutthrough deliver is pending, as we always want to make a connection.
563 This also avoids re-routing the message when later placing the cutthrough
564 connection after a verify cache hit.
565 Do not update it with the verify result either.
567 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
568 when routing results in more than one destination address.
570 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
571 signing (which inhibits the cutthrough capability). Previously only
572 the presence of an option was tested; now an expansion evaluating as
573 empty is permissible (obviously it should depend only on data available
574 when the cutthrough connection is made).
576 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
577 the relevant preceding SMTP command did not note the pipelining mode.
579 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
580 Previously they were not counted.
582 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
583 as one having no matching records. Previously we deferred the message
584 that needed the lookup.
586 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
587 distinguished as "(=".
589 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
590 for missing MX records. Previously it only worked for missing A records.
592 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
594 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
595 after the data-go-ahead and data-ack. Patch from Jason Betts.
597 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
598 even for a "none" policy. Patch from Tony Meyer.
600 JH/22 Fix continued use of a connection for further deliveries. If a port was
601 specified by a router, it must also match for the delivery to be
604 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
605 When one of the recipients of a message was unsuitable for the connection
606 (has no matching addresses), we lost track of needing to mark it
607 deferred. As a result mail would be lost.
609 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
611 JH/25 Decoding ACL controls is now done using a binary search; the source code
612 takes up less space and should be simpler to maintain. Merge the ACL
613 condition decode tables also, with similar effect.
615 JH/26 Fix problem with one_time used on a redirect router which returned the
616 parent address unchanged. A retry would see the parent address marked as
617 delivered, so not attempt the (identical) child. As a result mail would
620 JH/27 Fix a possible security hole, wherein a process operating with the Exim
621 UID can gain a root shell. Credit to http://www.halfdog.net/ for
622 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
625 JH/28 Enable {spool,log} filesystem space and inode checks as default.
626 Main config options check_{log,spool}_{inodes,space} are now
627 100 inodes, 10MB unless set otherwise in the configuration.
629 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
630 Previously it only applied to the main-section connection policy
633 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
635 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
636 by me. Added RFC7919 DH primes as an alternative.
638 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
639 are not in the system include path.
641 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
642 GnuTLS, when a session startup failed (eg because the client disconnected)
643 Exim did stdio operations after fclose. This was exposed by a recent
644 change which nulled out the file handle after the fclose.
646 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
647 signed directly by the cert-signing cert, rather than an intermediate
648 OCSP-signing cert. This is the model used by LetsEncrypt.
650 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
652 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
653 an incoming connection.
655 HS/02 Bug 1802: Do not half-close the connection after sending a request
658 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
659 fallback to "prime256v1".
661 JH/34 SECURITY: Use proper copy of DATA command in error message.
662 Could leak key material. Remotely exploitable. CVE-2016-9963.
668 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
669 and 3.4.4 - once the server is enabled to respond to an OCSP request
670 it does even when not requested, resulting in a stapling non-aware
671 client dropping the TLS connection.
673 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
674 support variable-length bit vectors. No functional change.
676 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
677 The I= interface field on outgoing lines is now after the H= remote
678 host field, same as incoming lines. There is a separate
679 outgoing_interface log selector which allows you to disable the
682 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
683 If not running log_selector +smtp_connection the mainlog would be held
684 open indefinitely after a "too many connections" event, including to a
685 deleted file after a log rotate. Leave the per net connection logging
686 leaving it open for efficiency as that will be quickly detected by the
687 check on the next write.
689 HS/01 Bug 1671: Fix post transport crash.
690 Processing the wait-<transport> messages could crash the delivery
691 process if the message IDs didn't exist for some reason. When
692 using 'split_spool_directory=yes' the construction of the spool
693 file name failed already, exposing the same netto behaviour.
695 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
696 mime_regex ACL conditions.
698 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
699 to DSN fail messages (bounces): remote IP, remote greeting, remote response
700 to HELO, local diagnostic string.
702 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
703 log line to debug. Even when configured with a tls authenticator many
704 client connections are expected to not authenticate in this way, so
705 an authenticate fail is not an error.
707 HS/02 Add the Exim version string to the process info. This way exiwhat
708 gives some more detail about the running daemon.
710 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
711 matter for fast-change records such as DNSBLs.
713 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
714 retry record, even if constant. There may be multiple transports with
715 different interface settings and the retry behaviour needs to be kept
718 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
720 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
722 JH/10 Bug 840: fix log_defer_output option of pipe transport
724 JH/11 Bug 830: use same host for all RCPTS of a message, even under
725 hosts_randomize. This matters a lot when combined with mua_wrapper.
727 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
728 ${quote_pgsql:<string>} operator.
730 JH/13 Bug 1708: avoid misaligned access in cached lookup.
732 JH/14 Change header file name for freeradius-client. Relevant if compiling
733 with Radius support; from the Gentoo tree and checked under Fedora.
735 JH/15 Bug 1712: Introduce $prdr_requested flag variable
737 JH/16 Bug 1714: Permit an empty string as expansion result for transport
738 option transport_filter, meaning no filtering.
740 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
742 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
743 defaults to "*" (all hosts). The variable is now available when not built
744 with TLS, default unset, mainly to enable keeping the testsuite sane.
745 If a server certificate is not supplied (via tls_certificate) an error is
746 logged, and clients will find TLS connections fail on startup. Presumably
747 they will retry in-clear.
748 Packagers of Exim are strongly encouraged to create a server certificate
749 at installation time.
751 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
752 with the $config_file variable.
754 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
755 in transport context, after the attempt, and per-recipient. The latter type
756 is per host attempted. The event data is the error message, and the errno
757 information encodes the lookup type (A vs. MX) used for the (first) host,
758 and the trailing two digits of the smtp 4xx response.
760 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
761 to write to mainlog (or rejectlog, paniclog) in the window between file
762 creation and permissions/ownership being changed. Particularly affects
763 installations where exicyclog is run as root, rather than exim user;
764 result is that the running daemon panics and dies.
766 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
768 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
769 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
770 "pri" and "weight". Note that the previous implicit priority given by the
771 list order is no longer honoured.
773 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
776 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
777 by defining SUPPORT_SOCKS.
779 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
780 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
781 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
782 variables are renamed to proxy_{local,external}_{address,port}.
784 JH/27 Move Internationalisation support from Experimental to mainline, enabled
785 for a build by defining SUPPORT_I18N
787 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
788 of the query string, and make ${quote_redis:} do that quoting.
790 JH/29 Move Events support from Experimental to mainline, enabled by default
791 and removable for a build by defining DISABLE_EVENT.
793 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
795 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
796 cached by the daemon.
798 JH/32 Move Redis support from Experimental to mainline, enabled for a build
799 by defining LOOKUP_REDIS. The libhiredis library is required.
801 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
802 keys are given for lookup.
804 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
805 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
806 only supported when built with TLS support. The PolarSSL SHA routines
807 are still used when the TLS library is too old for convenient support.
809 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
810 openssl_options), for security. OpenSSL forces this from version 1.1.0
811 server-side so match that on older versions.
813 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
814 allocation for $value could be released as the expansion processing
815 concluded, but leaving the global pointer active for it.
817 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
818 and to use the domains and local_parts ACL conditions.
820 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
821 incorrectly not doubled on cutthrough transmission, hence seen as a
822 body-termination at the receiving system - resulting in truncated mails.
823 Commonly the sender saw a TCP-level error, and retransmitted the message
824 via the normal store-and-forward channel. This could result in duplicates
825 received - but deduplicating mailstores were liable to retain only the
826 initial truncated version.
828 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
830 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
832 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
833 we're in there, support oversigning also; bug 1309.
835 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
837 HS/04 Add support for keep_environment and add_environment options.
839 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
840 either intentional arithmetic overflow during PRNG, or testing config-
843 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
844 delivery resulted in actual delivery. Cancel cutthrough before DATA
847 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
848 rejects a recipient: pass the reject to the originator.
850 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
851 Many were false-positives and ignorable, but it's worth fixing the
854 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
855 for the new environment-manipulation done at startup. Move the routines
856 from being local to tls.c to being global via the os.c file.
858 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
859 an extract embedded as result-arg for a map, the first arg for extract
860 is unavailable so we cannot tell if this is a numbered or keyed
861 extraction. Accept either.
867 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
870 JH/02 The smtp transport option "multi_domain" is now expanded.
872 JH/03 The smtp transport now requests PRDR by default, if the server offers
875 JH/04 Certificate name checking on server certificates, when exim is a client,
876 is now done by default. The transport option tls_verify_cert_hostnames
877 can be used to disable this per-host. The build option
878 EXPERIMENTAL_CERTNAMES is withdrawn.
880 JH/05 The value of the tls_verify_certificates smtp transport and main options
881 default to the word "system" to access the system default CA bundle.
882 For GnuTLS, only version 3.0.20 or later.
884 JH/06 Verification of the server certificate for a TLS connection is now tried
885 (but not required) by default. The verification status is now logged by
886 default, for both outbound TLS and client-certificate supplying inbound
889 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
892 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
893 Status Notification (bounce) messages are now MIME format per RFC 3464.
894 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
895 under the control of the dsn_advertise_hosts option, and routers may
896 have a dsn_lasthop option.
898 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
899 default, modifiable by a malware= option. The list separator for
900 the options can now be changed in the usual way. Bug 68.
902 JH/10 The smtp_receive_timeout main option is now expanded before use.
904 JH/11 The incoming_interface log option now also enables logging of the
905 local interface on delivery outgoing connections.
907 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
908 if the interface and destination host and port all match.
910 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
913 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
914 Patch from Andrew Lewis.
916 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
917 now supports optional time-restrictions, weighting, and priority
918 modifiers per server. Patch originally by <rommer@active.by>.
920 JH/16 The spamd_address main option now supports a mixed list of local
921 and remote servers. Remote servers can be IPv6 addresses, and
922 specify a port-range.
924 JH/17 Bug 68: The spamd_address main option now supports an optional
925 timeout value per server.
927 JH/18 Bug 1581: Router and transport options headers_add/remove can
928 now have the list separator specified.
930 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
933 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
936 JH/21 Support for the A6 type of dns record is withdrawn.
938 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
939 rather than the verbs used.
941 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
942 from 255 to 1024 chars.
944 JH/24 Verification callouts now attempt to use TLS by default.
946 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
947 are generic router options now. The defaults didn't change.
949 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
950 Original patch from Alexander Shikoff, worked over by JH.
952 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
953 files if ZCAT_COMMAND is not executable.
955 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
957 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
959 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
960 Normally benign, it bites when the pair was led to by a CNAME;
961 modern usage is to not canonicalize the domain to a CNAME target
962 (and we were inconsistent anyway for A-only vs AAAA+A).
964 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
966 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
967 when evaluating $sender_host_dnssec.
969 JH/31 Check the HELO verification lookup for DNSSEC, adding new
970 $sender_helo_dnssec variable.
972 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
974 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
976 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
978 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
979 documented as working, but never had. Support all but $spam_report.
981 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
982 added for tls authenticator.
984 HS/03 Add perl_taintmode main config option
990 TL/01 When running the test suite, the README says that variables such as
991 no_msglog_check are global and can be placed anywhere in a specific
992 test's script, however it was observed that placement needed to be near
993 the beginning for it to behave that way. Changed the runtest perl
994 script to read through the entire script once to detect and set these
995 variables, reset to the beginning of the script, and then run through
996 the script parsing/test process like normal.
998 TL/02 The BSD's have an arc4random API. One of the functions to induce
999 adding randomness was arc4random_stir(), but it has been removed in
1000 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
1001 function when detected.
1003 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
1004 cause callback expansion.
1006 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
1007 syntax errors in an expansion can be treated as a string instead of
1008 logging or causing an error, due to the internal use of bool_lax
1009 instead of bool when processing it.
1011 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
1012 server certificates when making smtp deliveries.
1014 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
1016 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
1018 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
1020 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
1021 Merged patch from Sebastian Wiedenroth.
1023 JH/05 Fix results-pipe from transport process. Several recipients, combined
1024 with certificate use, exposed issues where response data items split
1025 over buffer boundaries were not parsed properly. This eventually
1026 resulted in duplicates being sent. This issue only became common enough
1027 to notice due to the introduction of connection certificate information,
1028 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1030 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1031 size buffer was used, resulting in syntax errors when an expansion
1034 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1035 version 3.3.6 or later.
1037 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1038 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1039 both become "event_action", the variables become $event_name, $event_data
1040 and $event_defer_errno. There is a new variable $verify_mode, usable in
1041 routers, transports and related events. The tls:cert event is now also
1042 raised for inbound connections, if the main configuration event_action
1045 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1046 early OCSP support, but no stapling (appears to be less than 1.0.0).
1048 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1049 server certificate names available under the smtp transport option
1050 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1053 JH/10 Time-related extraction expansions from certificates now use the main
1054 option "timezone" setting for output formatting, and are consistent
1055 between OpenSSL and GnuTLS compilations. Bug 1541.
1057 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1058 encoded parameter in the incoming message. Bug 1558.
1060 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1061 include certificate info, eximon was claiming there were spoolfile
1064 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1066 JH/14 Log delivery-related information more consistently, using the sequence
1067 "H=<name> [<ip>]" wherever possible.
1069 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1070 are problematic for Debian distribution, omit them from the release
1073 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1075 JH/16 Fix string representation of time values on 64bit time_t architectures.
1078 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1079 output list separator was used.
1084 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1085 checkers that were complaining about end of non-void function with no
1088 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1089 This was a regression introduced in 4.83 by another bugfix.
1091 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1093 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1094 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1100 TF/01 Correctly close the server side of TLS when forking for delivery.
1102 When a message was received over SMTP with TLS, Exim failed to clear up
1103 the incoming connection properly after forking off the child process to
1104 deliver the message. In some situations the subsequent outgoing
1105 delivery connection happened to have the same fd number as the incoming
1106 connection previously had. Exim would try to use TLS and fail, logging
1107 a "Bad file descriptor" error.
1109 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1110 utilities have not been installed.
1112 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1113 temporary space as the ACL may create new global variables.
1115 TL/01 LDAP support uses per connection or global context settings, depending
1116 upon the detected version of the libraries at build time.
1118 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1119 to extract and use the src ip:port in logging and expansions as if it
1120 were a direct connection from the outside internet. PPv2 support was
1121 updated based on HAProxy spec change in May 2014.
1123 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1125 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1126 Properly escape header and check for NULL return.
1128 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1131 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1133 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1134 characters in header names, implemented as a verify condition.
1135 Contributed by Michael Fischer v. Mollard.
1137 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1138 results permerror and temperror. Previous values are deprecated but
1139 still accepted. In a future release, err_perm and err_temp will be
1140 completely removed, which will be a backward incompatibility if the
1141 ACL tests for either of these two old results. Patch contributed by
1142 user bes-internal on the mailing list.
1144 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1146 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1147 selectors, in both main and reject logs.
1149 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1152 JH/07 Add malware type "sock" for talking to simple daemon.
1154 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1156 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1157 routers/transports under cutthrough routing.
1159 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1160 numbers. Touch up "bool" conditional to keep the same definition.
1162 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1164 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1166 JH/12 Expand items in router/transport headers_add or headers_remove lists
1167 individually rather than the list as a whole. Bug 1452.
1169 Required for reasonable handling of multiple headers_ options when
1170 they may be empty; requires that headers_remove items with embedded
1171 colons must have them doubled (or the list-separator changed).
1173 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1174 view the policy declared in the DMARC record. Currently, $dmarc_status
1175 is a combined value of both the record presence and the result of the
1178 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1180 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1181 dnslookup router and the smtp transport (applying to the forward
1184 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1185 of ldap servers used for a specific lookup. Patch provided by Heiko
1188 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1189 New variable $lookup_dnssec_authenticated for observability.
1191 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1192 Patch submitted by Lars Timman.
1194 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1196 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1197 Requires trusted mode and valid format message id, aborts otherwise.
1198 Patch contributed by Heiko Schlichting.
1200 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1201 certextract with support for various fields. Bug 1358.
1203 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1204 is requested by default, modifiable by smtp transport option
1207 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1208 operate on certificate variables to give certificate fingerprints
1209 Also new ${sha256:cert_variable}.
1211 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1213 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1214 Christian Aistleitner.
1216 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1218 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1219 file. Patch from Wolfgang Breyha.
1221 JH/25 Expand the coverage of the delivery $host and $host_address to
1222 client authenticators run in verify callout. Bug 1476.
1224 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1225 align with daemon_smtp_ports. Bug 72.
1227 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1228 support and error reporting did not work properly.
1230 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1231 and is readable. Patch from Andrew Colin Kissa.
1233 TL/14 Enhance documentation of ${run expansion and how it parses the
1234 commandline after expansion, particularly in the case when an
1235 unquoted variable expansion results in an empty value.
1237 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1239 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1240 by renaming away from it. Addresses GH issue 15, reported by
1243 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1245 TL/15 SECURITY: prevent double expansion in math comparison functions
1246 (can expand unsanitized data). Not remotely exploitable.
1253 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1255 PP/02 Make -n do something, by making it not do something.
1256 When combined with -bP, the name of an option is not output.
1258 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1261 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1262 $sender_host_name and config options to manage this, and basic check
1265 PP/05 DSCP support for outbound connections and control modifier for inbound.
1267 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1268 (Only plugin which currently uses this is kerberos4, which nobody should
1269 be using, but we should make it available and other future plugins might
1270 conceivably use it, even though it would break NAT; stuff *should* be
1271 using channel bindings instead).
1273 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1274 name; added for Sendmail compatibility; requires admin caller.
1275 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1276 just ignore it); requires trusted caller.
1277 Also parse but ignore: -Ac -Am -X<logfile>
1280 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1282 TL/02 Add +smtp_confirmation as a default logging option.
1284 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1285 Patch by Magnus Holmgren from 2007-02-20.
1287 TL/04 Bugzilla 1281 - Spec typo.
1288 Bugzilla 1283 - Spec typo.
1289 Bugzilla 1290 - Spec grammar fixes.
1291 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1293 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1295 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1296 mailing list by by Dmitry Isaikin.
1298 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1300 JH/02 Support "G" suffix to numbers in ${if comparisons.
1302 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1304 NM/01 Bugzilla 1197 - Spec typo
1305 Bugzilla 1196 - Spec examples corrections
1307 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1309 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1310 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1313 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1314 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1316 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1317 "acl {{name}{arg}...}", and optional args on acl condition
1320 JH/05 Permit multiple router/transport headers_add/remove lines.
1322 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1324 JH/07 Avoid using a waiting database for a single-message-only transport.
1325 Performance patch from Paul Fisher. Bugzilla 1262.
1327 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1330 JH/09 Add $headers_added variable, with content from use of ACL modifier
1331 add_header (but not yet added to the message). Bugzilla 199.
1333 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1334 Pulled from Bugzilla 817 by Wolfgang Breyha.
1336 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1338 (nb: this is the same fix as in Exim 4.80.1)
1340 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1343 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1346 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1348 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1349 advertise SMTP AUTH mechanism to us, instead of a generic
1350 protocol violation error. Also, make Exim more robust to bad
1351 data from the Dovecot auth socket.
1353 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1355 When a queue runner is handling a message, Exim first routes the
1356 recipient addresses, during which it prunes them based on the retry
1357 hints database. After that it attempts to deliver the message to
1358 any remaining recipients. It then updates the hints database using
1361 So if a recipient address works intermittently, it can get repeatedly
1362 deferred at routing time. The retry hints record remains fresh so the
1363 address never reaches the final cutoff time.
1365 This is a fairly common occurrence when a user is bumping up against
1366 their storage quota. Exim had some logic in its local delivery code
1367 to deal with this. However it did not apply to per-recipient defers
1368 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1370 This change adds a proper retry rule check during routing so that the
1371 final cutoff time is checked against the message's age. We only do
1372 this check if there is an address retry record and there is not a
1373 domain retry record; this implies that previous attempts to handle
1374 the address had the retry_use_local_parts option turned on. We use
1375 this as an approximation for the destination being like a local
1376 delivery, as in LMTP.
1378 I suspect this new check makes the old local delivery cutoff check
1379 redundant, but I have not verified this so I left the code in place.
1381 TF/02 Correct gecos expansion when From: is a prefix of the username.
1383 Test 0254 submits a message to Exim with the header
1387 When I ran the test suite under the user fanf2, Exim expanded
1388 the header to contain my full name, whereas it should have added
1389 a Resent-Sender: header. It erroneously treats any prefix of the
1390 username as equal to the username.
1392 This change corrects that bug.
1394 GF/01 DCC debug and logging tidyup
1395 Error conditions log to paniclog rather than rejectlog.
1396 Debug lines prefixed by "DCC: " to remove any ambiguity.
1398 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1400 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1401 Bug spotted by Jeremy Harris; was flawed since initial commit.
1402 Would have resulted in OCSP responses post-SNI triggering an Exim
1403 NULL dereference and crash.
1405 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1407 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1408 Bug detection, analysis and fix by Samuel Thibault.
1409 Bugzilla 1331, Debian bug #698092.
1411 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1413 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1414 Server implementation by Todd Lyons, client by JH.
1415 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1416 config variable "prdr_enable" controls whether the server
1417 advertises the facility. If the client requests PRDR a new
1418 acl_data_smtp_prdr ACL is called once for each recipient, after
1419 the body content is received and before the acl_smtp_data ACL.
1420 The client is controlled by both of: a hosts_try_prdr option
1421 on the smtp transport, and the server advertisement.
1422 Default client logging of deliveries and rejections involving
1423 PRDR are flagged with the string "PRDR".
1425 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1426 fclose(). Diagnosis by Todd Lyons.
1428 PP/17 Update configure.default to handle IPv6 localhost better.
1429 Patch by Alain Williams (plus minor tweaks).
1432 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1433 This is now consistent with GnuTLS, and is now documented: the
1434 previous undocumented portable approach to treating the option as
1435 unset was to force an expansion failure. That still works, and
1436 an empty string is now equivalent.
1438 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1439 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1440 not performing validation itself.
1442 PP/20 Added force_command boolean option to pipe transport.
1443 Patch from Nick Koston, of cPanel Inc.
1445 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1448 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1450 PP/21 Fix eximon continuous updating with timestamped log-files.
1451 Broken in a format-string cleanup in 4.80, missed when I repaired the
1452 other false fix of the same issue.
1453 Report and fix from Heiko Schlichting.
1456 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1457 Report from Prashanth Katuri.
1459 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1460 It's SecureTransport, so affects any MacOS clients which use the
1461 system-integrated TLS libraries, including email clients.
1463 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1464 using a MIME ACL for non-SMTP local injection.
1465 Report and assistance in diagnosis by Warren Baker.
1467 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1469 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1471 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1472 last id that failed so it may be referenced in subsequent ACL's.
1474 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1477 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1478 ldap library initialization, allowing self-signed CA's to be
1479 used. Also properly sets require_cert option later in code by
1480 using NULL (global ldap config) instead of ldap handle (per
1481 session). Bug diagnosis and testing by alxgomz.
1483 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1484 the src/util/ subdirectory.
1486 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1487 renamed to Transport Post Delivery Action by Jeremy Harris, as
1490 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1491 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1492 redis_servers = needs to be configured which will be used by the redis
1493 lookup. Patch from Warren Baker, of The Packet Hub.
1495 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1497 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1498 hostname or reverse DNS when processing a host list. Used suggestions
1499 from multiple comments on this bug.
1501 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1503 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1504 Missed a few lines, added it to make the runtest require no keyboard
1507 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1508 contains upper case chars. Make router use caseful_local_part.
1510 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1511 support when GnuTLS has been built with p11-kit.
1517 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1519 This, or similar/improved, will also be change PP/11 of 4.82.
1525 PP/01 Handle short writes when writing local log-files.
1526 In practice, only affects FreeBSD (8 onwards).
1527 Bugzilla 1053, with thanks to Dmitry Isaikin.
1529 NM/01 Bugzilla 949 - Documentation tweak
1531 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1534 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1536 PP/02 Implemented gsasl authenticator.
1538 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1540 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1541 `pkg-config foo` for cflags/libs.
1543 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1544 with rest of GSASL and with heimdal_gssapi.
1546 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1547 `pkg-config foo` for cflags/libs for the TLS implementation.
1549 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1550 properties get this fed in as external SSF. A number of robustness
1551 and debugging improvements to the cyrus_sasl authenticator.
1553 PP/08 cyrus_sasl server now expands the server_realm option.
1555 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1556 Patch by Jeremy Harris.
1558 PP/10 Added dbmjz lookup type.
1560 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1562 PP/12 MAIL args handles TAB as well as SP, for better interop with
1563 non-compliant senders.
1564 Analysis and variant patch by Todd Lyons.
1566 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1567 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1568 Patch from Dirk Mueller <dmueller@suse.com>
1570 PP/13 tls_peerdn now print-escaped for spool files.
1571 Observed some $tls_peerdn in wild which contained \n, which resulted
1572 in spool file corruption.
1574 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1575 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1576 or write after TLS renegotiation, which otherwise led to messages
1579 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1580 as a tracking header (ie: a signed header comes before the signature).
1581 Patch from Wolfgang Breyha.
1583 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1584 comma-sep list; embedded commas doubled.
1586 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1588 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1590 Report and patch from Dmitry Banschikov.
1592 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1593 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1594 failures. We appear to now support TLS1.1+ with Exim.
1596 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1597 lets Exim select keys and certificates based upon TLS SNI from client.
1598 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1599 before an outbound SMTP session. New log_selector, +tls_sni.
1601 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1602 NULL dereference. Report and patch from Alun Jones.
1604 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1605 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1606 Not seeing resolver debug output on NetBSD, but suspect this is a
1607 resolver implementation change.
1609 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1610 Left warnings. Added "eximon gdb" invocation mode.
1612 PP/21 Defaulting "accept_8bitmime" to true, not false.
1614 PP/22 Added -bw for inetd wait mode support.
1616 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1617 locate the relevant includes and libraries. Made this the default.
1619 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1620 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1622 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1623 This may cause build issues on older platforms.
1625 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1626 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1627 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1628 Added SNI support via GnuTLS too.
1629 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1631 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1633 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1634 Applied second patch from Janne, implementing suggestion to default
1635 multiple-strings-in-record handling to match SPF spec.
1637 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1639 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1640 read-only, out of scope).
1641 Patch from Wolfgang Breyha, report from Stuart Northfield.
1643 PP/29 Fix three issues highlighted by clang analyser static analysis.
1644 Only crash-plausible issue would require the Cambridge-specific
1645 iplookup router and a misconfiguration.
1646 Report from Marcin Mirosław.
1648 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1650 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1651 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1652 As part of this, removing so much warning spew let me fix some minor
1653 real issues in debug logging.
1655 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1656 assignment on my part. Fixed.
1658 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1659 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1660 Janne Snabb (who went above and beyond: thank you).
1662 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1663 string otherwise requires a connection and a bunch more work and it's
1664 relatively easy to get wrong. Should also expose TLS library linkage
1667 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1668 64-bit ${eval} (JH/03).
1670 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1671 GNU libc to support some of the 64-bit stuff, should not lead to
1672 conflicts. Defined before os.h is pulled in, so if a given platform
1673 needs to override this, it can.
1675 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1676 protection layer was required, which is not implemented.
1677 Bugzilla 1254, patch from Wolfgang Breyha.
1679 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1680 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1681 tls_dhparam take prime identifiers. Also unbreak combination of
1682 OpenSSL+DH_params+TLSSNI.
1684 PP/39 Disable SSLv2 by default in OpenSSL support.
1690 PP/01 Solaris build fix for Oracle's LDAP libraries.
1691 Bugzilla 1109, patch from Stephen Usher.
1693 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1695 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1698 TF/02 Fix a couple more cases where we did not log the error message
1699 when unlink() failed. See also change 4.74-TF/03.
1701 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1702 lock up or crash if it happened to be inside a call to libc when it
1703 got a SIGUSR1 from exiwhat.
1705 The SIGUSR1 handler appends the current process status to the process
1706 log which is later printed by exiwhat. It used to use the general
1707 purpose logging code to do this, but several functions it calls are
1708 not safe for signals.
1710 The new output code in the SIGUSR1 handler is specific to the process
1711 log, and simple enough that it's easy to inspect for signal safety.
1712 Removing some special cases also simplifies the general logging code.
1713 Removing the spurious timestamps from the process log simplifies
1716 TF/04 Improved ratelimit ACL condition.
1718 The /noupdate option has been deprecated in favour of /readonly which
1719 has clearer semantics. The /leaky, /strict, and /readonly update modes
1720 are mutually exclusive. The update mode is no longer included in the
1721 database key; it just determines when the database is updated. (This
1722 means that when you upgrade Exim will forget old rate measurements.)
1724 Exim now checks that the per_* options are used with an update mode that
1725 makes sense for the current ACL. For example, when Exim is processing a
1726 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1727 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1728 must specify per_mail/readonly. If you omit the update mode it defaults to
1729 /leaky where that makes sense (as before) or /readonly where required.
1731 The /noupdate option is now undocumented but still supported for
1732 backwards compatibility. It is equivalent to /readonly except that in
1733 ACLs where /readonly is required you may specify /leaky/noupdate or
1734 /strict/noupdate which are treated the same as /readonly.
1736 A useful new feature is the /count= option. This is a generalization
1737 of the per_byte option, so that you can measure the throughput of other
1738 aggregate values. For example, the per_byte option is now equivalent
1739 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1741 The per_rcpt option has been generalized using the /count= mechanism
1742 (though it's more complicated than the per_byte equivalence). When it is
1743 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1744 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1745 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1746 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1747 Note that using per_rcpt with a non-readonly update mode in more than
1748 one ACL will cause the recipients to be double-counted. (The per_mail
1749 and per_byte options don't have this problem.)
1751 The handling of very low rates has changed slightly. If the computed rate
1752 is less than the event's count (usually one) then this event is the first
1753 after a long gap. In this case the rate is set to the same as this event's
1754 count, so that the first message of a spam run is counted properly.
1756 The major new feature is a mechanism for counting the rate of unique
1757 events. The new per_addr option counts the number of different
1758 recipients that someone has sent messages to in the last time period. It
1759 behaves like per_rcpt if all the recipient addresses are different, but
1760 duplicate recipient addresses do not increase the measured rate. Like
1761 the /count= option this is a general mechanism, so the per_addr option
1762 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1763 example, measure the rate that a client uses different sender addresses
1764 with the options per_mail/unique=$sender_address. There are further
1765 details in the main documentation.
1767 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1769 TF/06 Removed a few PCRE remnants.
1771 TF/07 Automatically extract Exim's version number from tags in the git
1772 repository when doing development or release builds.
1774 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1775 Bugzilla 879. Patch from Paul Fisher.
1777 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1778 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1781 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1783 PP/05 Variable $av_failed, true if the AV scanner deferred.
1784 Bugzilla 1078. Patch from John Horne.
1786 PP/06 Stop make process more reliably on build failure.
1787 Bugzilla 1087. Patch from Heiko Schlittermann.
1789 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1790 Bugzilla 1089. Patch from Heiko Schlittermann.
1792 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1793 Bugzilla 1131. Patch from Holger Weiß.
1795 PP/09 Handle IPv6 addresses with SPF.
1796 Bugzilla 860. Patch from Wolfgang Breyha.
1798 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1800 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1803 PP/11 match_* no longer expand right-hand-side by default.
1804 New compile-time build option, EXPAND_LISTMATCH_RHS.
1805 New expansion conditions, "inlist", "inlisti".
1807 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1809 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1811 PP/14 fix log_write() format string regression from TF/03.
1812 Bugzilla 1152. Patch from Dmitry Isaikin.
1818 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1820 PP/02 Harmonised TLS library version reporting; only show if debugging.
1821 Layout now matches that introduced for other libraries in 4.74 PP/03.
1823 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1825 PP/04 New "dns_use_edns0" global option.
1827 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1830 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1831 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1833 TK/01 Updated PolarSSL code to 0.14.2.
1834 Bugzilla 1097. Patch from Andreas Metzler.
1836 PP/07 Catch divide-by-zero in ${eval:...}.
1837 Fixes bugzilla 1102.
1839 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1842 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1843 format-string attack -- SECURITY: remote arbitrary code execution.
1845 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1846 time unintentionally subject to list matching rules, letting the header
1847 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1848 arbitrary string expansion). This allowed for information disclosure.
1850 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1851 INT_MIN/-1 -- value coerced to INT_MAX.
1857 NM/01 Workaround for PCRE version dependency in version reporting
1860 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1861 This fixes portability to compilers other than gcc, notably
1862 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1864 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1865 makefiles for portability to HP-UX and POSIX correctness.
1867 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1868 Also via indented variable definition in the Makefile.
1869 (Debugging by Oliver Heesakkers).
1871 PP/02 Restore caching of spamd results with expanded spamd_address.
1872 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1874 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1875 Improves build reliability. Fix from: Frank Elsner
1877 NM/02 Fix wide character breakage in the rfc2047 coding
1878 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1880 NM/03 Allow underscore in dnslist lookups
1881 Fixes bug 1026. Patch from Graeme Fowler
1883 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1884 Code patches from Adam Ciarcinski of NetBSD.
1886 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1889 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1890 is logged, to avoid truncation. Patch from John Horne.
1892 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1893 Patch from Jakob Hirsch.
1895 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1896 SQL string expansion failure details.
1897 Patch from Andrey Oktyabrski.
1899 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1900 Patch from Simon Arlott.
1902 PP/09 New lookups functionality failed to compile on old gcc which rejects
1903 extern declarations in function scope.
1904 Patch from Oliver Fleischmann
1906 PP/10 Use sig_atomic_t for flags set from signal handlers.
1907 Check getgroups() return and improve debugging.
1908 Fixed developed for diagnosis in bug 927 (which turned out to be
1911 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1912 Patch from Mark Zealey.
1914 PP/12 Bugzilla 1056: Improved spamd server selection.
1915 Patch from Mark Zealey.
1917 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1918 Based on patch from Heiko Schlittermann.
1920 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1921 Patch from Uwe Doering, sign-off by Michael Haardt.
1923 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1924 variable declaration deep within a block. Bug and patch from
1927 PP/15 lookups-Makefile IRIX compatibility coercion.
1929 PP/16 Make DISABLE_DKIM build knob functional.
1931 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1932 Patch by Simon Arlott
1934 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1935 variable argument macros. Our copy now differs from upstream.
1941 TF/01 Failure to get a lock on a hints database can have serious
1942 consequences so log it to the panic log.
1944 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1945 controlled using the smtp_confirmation log selector.
1947 TF/03 Include the error message when we fail to unlink a spool file.
1949 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1950 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1951 for maintaining out-of-tree patches for some time.
1953 PP/01 Bugzilla 139: Documentation and portability issues.
1954 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1955 Handle per-OS dynamic-module compilation flags.
1957 PP/02 Let /dev/null have normal permissions.
1958 The 4.73 fixes were a little too stringent and complained about the
1959 permissions on /dev/null. Exempt it from some checks.
1960 Reported by Andreas M. Kirchwitz.
1962 PP/03 Report version information for many libraries, including
1963 Exim version information for dynamically loaded libraries. Created
1964 version.h, now support a version extension string for distributors
1965 who patch heavily. Dynamic module ABI change.
1967 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1968 privilege escalation vulnerability whereby the Exim run-time user
1969 can cause root to append content of the attacker's choosing to
1972 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1975 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1976 If dropping privileges for untrusted macros, we disabled normal logging
1977 on the basis that it would fail; for the Exim run-time user, this is not
1978 the case, and it resulted in successful deliveries going unlogged.
1979 Fixed. Reported by Andreas Metzler.
1985 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1986 only prepend for the Resent-* case. Fixes regression introduced in
1987 Exim 4.70 by NM/22 for Bugzilla 607.
1989 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1990 increasing numbers of administrators be bitten by this.
1992 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1994 PP/03 Bugzilla 994: added openssl_options main configuration option.
1996 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1998 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
2000 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
2002 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
2003 without a peer certificate, leading to a segfault because of an
2004 assumption that peers always have certificates. Be a little more
2005 paranoid. Problem reported by Martin Tscholak.
2007 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
2008 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
2009 NB: ClamAV planning to remove STREAM in "middle of 2010".
2010 CL also introduces -bmalware, various -d+acl logging additions and
2011 more caution in buffer sizes.
2013 PP/09 Implemented reverse_ip expansion operator.
2015 PP/10 Bugzilla 937: provide a "debug" ACL control.
2017 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
2019 PP/12 Bugzilla 973: Implement --version.
2021 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2023 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2025 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2027 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2028 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2029 ignore trailing whitespace.
2031 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2033 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2036 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2037 Notification from Dr Andrew Aitchison.
2039 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2040 ExtendedDetectionInfo response format.
2041 Notification from John Horne.
2043 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2046 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2047 XSL and documented dependency on system catalogs, with examples of how
2050 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2053 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2054 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2055 configuration file which is writeable by the Exim user or group.
2057 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2058 of configuration files to cover files specified with the -C option if
2059 they are going to be used with root privileges, not just the default
2062 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2063 option (effectively making it always true).
2065 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2066 files to be used while preserving root privileges.
2068 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2069 that rogue child processes cannot use them.
2071 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2072 run-time user, instead of root.
2074 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2075 Exim run-time user without dropping privileges.
2077 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2078 result string, instead of calling string_vformat() twice with the same
2081 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2082 for other users. Others should always drop root privileges if they use
2083 -C on the command line, even for a whitelisted configure file.
2085 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2087 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2093 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2094 $data_path, and $header_path variables; fixed documentation bugs and
2097 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2098 exipick to access non-standard spools, including the "frozen" queue
2101 NM/01 Bugzilla 965: Support mysql stored procedures.
2102 Patch from Alain Williams
2104 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2106 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2107 Patch from Andreas Metzler
2109 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2110 Patch from Kirill Miazine
2112 NM/05 Bugzilla 671: Added umask to procmail example.
2114 JJ/03 installed exipick 20100323.0, fixing doc bug
2116 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2117 directory. Notification and patch from Dan Rosenberg.
2119 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2121 TK/02 Improve log output when DKIM signing operation fails.
2123 MH/01 Treat the transport option dkim_domain as a colon separated
2124 list, not as a single string, and sign the message with each element,
2125 omitting multiple occurences of the same signer.
2127 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2128 Bugzilla 985, 986. Patch by Simon Arlott
2130 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2131 Patch by Simon Arlott
2133 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2134 MBX locking. Notification from Dan Rosenberg.
2140 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2142 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2144 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2146 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2148 NM/04 Bugzilla 915: Flag broken perl installation during build.
2154 TK/01 Added patch by Johannes Berg that expands the main option
2155 "spamd_address" if it starts with a dollar sign.
2157 TK/02 Write list of recipients to X-Envelope-Sender header when building
2158 the mbox-format spool file for content scanning (suggested by Jakob
2161 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2162 (http://www.dcc-servers.net/) support via dccifd. Activated by
2163 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2165 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2166 by Mark Daniel Reidel <mr@df.eu>.
2168 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2169 When building exim an external PCRE library is now needed -
2170 PCRE is a system library on the majority of modern systems.
2171 See entry on PCRE_LIBS in EDITME file.
2173 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2174 conversation. Added nologin parameter to request.
2175 Patch contributed by Kirill Miazine.
2177 TF/01 Do not log submission mode rewrites if they do not change the address.
2179 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2181 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2182 log files in place. Contributed by Roberto Lima.
2184 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2186 TF/03 Bugzilla 615: When checking the local_parts router precondition
2187 after a local_part_suffix or local_part_prefix option, Exim now
2188 does not use the address's named list lookup cache, since this
2189 contains cached lookups for the whole local part.
2191 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2192 Robert Millan. Documentation is in experimental-spec.txt.
2194 TF/04 Bugzilla 668: Fix parallel build (make -j).
2196 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2198 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2199 Patch provided by Jan Srzednicki.
2201 TF/05 Leading white space used to be stripped from $spam_report which
2202 wrecked the formatting. Now it is preserved.
2204 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2205 that they are available at delivery time.
2207 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2209 TF/08 TLS error reporting now respects the incoming_interface and
2210 incoming_port log selectors.
2212 TF/09 Produce a more useful error message if an SMTP transport's hosts
2213 setting expands to an empty string.
2215 NM/06 Bugzilla 744: EXPN did not work under TLS.
2216 Patch provided by Phil Pennock.
2218 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2219 Patch provided by Richard Godbee.
2221 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2222 acl_smtp_notquit, added index entry.
2224 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2225 Patch provided by Eugene Bujak.
2227 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2228 accept(). Patch provided by Maxim Dounin.
2230 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2231 Patch provided by Phil Pennock.
2233 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2235 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2236 Patch provided by Brad "anomie" Jorsch.
2238 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2239 Patch provided by Dean Brooks.
2241 TK/05 Add native DKIM support (does not depend on external libraries).
2243 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2244 Patch provided by Graeme Fowler.
2246 NM/16 Bugzilla 851: Documentation example syntax fix.
2248 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2250 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2253 NM/19 Bugzilla 745: TLS version reporting.
2254 Patch provided by Phil Pennock.
2256 NM/20 Bugzilla 167: bool: condition support.
2257 Patch provided by Phil Pennock.
2259 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2260 clients. Patch provided by Phil Pennock.
2262 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2263 Patch provided by Brad "anomie" Jorsch.
2265 NM/23 Bugzilla 687: Fix misparses in eximstats.
2266 Patch provided by Heiko Schlittermann.
2268 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2269 Patch provided by Heiko Schlittermann.
2271 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2272 plus update to original patch.
2274 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2276 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2277 Patch provided by David Brownlee.
2279 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2281 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2283 NM/30 Bugzilla 888: TLS documentation bugfixes.
2285 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2287 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2288 Unlike the original bugzilla I have changed all shell scripts in src tree.
2290 NM/33 Bugzilla 898: Transport filter timeout fix.
2291 Patch by Todd Rinaldo.
2293 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2294 Patch by Serge Demonchaux.
2296 NM/35 Bugzilla 39: Base64 decode bug fixes.
2297 Patch by Jakob Hirsch.
2299 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2301 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2303 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2305 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2311 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2312 ALT-N's libdkim that I have put here:
2313 http://duncanthrax.net/exim-experimental/
2315 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2316 were called 'true' and it seems that C99 defines that as a reserved
2317 keyword to be used with 'bool' variable types. That means you could
2318 not include C99-style headers which use bools without triggering
2319 build errors in sieve.c.
2321 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2322 as mailq or other aliases. Changed the --help handling significantly
2323 to do whats expected. exim_usage() emits usage/help information.
2325 SC/01 Added the -bylocaldomain option to eximstats.
2327 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2329 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2331 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2337 PH/01 Another patch from the Sieve maintainer.
2339 PH/02 When an IPv6 address is converted to a string for single-key lookup
2340 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2341 dots are used instead of colons so that keys in lsearch files need not
2342 contain colons. This was done some time before quoting was made available
2343 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2344 (notated using the quote facility) so as to distinguish them from IPv4
2345 keys. This meant that lookups for IP addresses in host lists did not work
2346 for iplsearch lookups.
2348 This has been fixed by arranging for IPv6 addresses to be expressed with
2349 colons if the lookup type is iplsearch. This is not incompatible, because
2350 previously such lookups could never work.
2352 The situation is now rather anomalous, since one *can* have colons in
2353 ordinary lsearch keys. However, making the change in all cases is
2354 incompatible and would probably break a number of configurations.
2356 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2359 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2360 conversion specification without a maximum field width, thereby enabling
2361 a rogue spamd server to cause a buffer overflow. While nobody in their
2362 right mind would setup Exim to query an untrusted spamd server, an
2363 attacker that gains access to a server running spamd could potentially
2364 exploit this vulnerability to run arbitrary code as the Exim user.
2366 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2367 $primary_hostname instead of what libspf2 thinks the hosts name is.
2369 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2370 a directory entry by the name of the lookup key. Previously, if a
2371 symlink pointed to a non-existing file or a file in a directory that
2372 Exim lacked permissions to read, a lookup for a key matching that
2373 symlink would fail. Now it is enough that a matching directory entry
2374 exists, symlink or not. (Bugzilla 503.)
2376 PH/03 The body_linecount and body_zerocount variables are now exported in the
2379 PH/04 Added the $dnslist_matched variable.
2381 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2382 This means they are set thereafter only if the connection becomes
2385 PH/06 Added the client_condition to authenticators so that some can be skipped
2386 by clients under certain conditions.
2388 PH/07 The error message for a badly-placed control=no_multiline_responses left
2389 "_responses" off the end of the name.
2391 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2393 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2394 (without spaces) instead of just copying the configuration text.
2396 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2398 PH/11 Added $max_received_linelength.
2400 PH/12 Added +ignore_defer and +include_defer to host lists.
2402 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2403 way in which PCRE > 7.0 is built.
2405 PH/14 Implemented queue_only_load_latch.
2407 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2408 MAIL command. The effect was to mangle the value on 64-bit systems.
2410 PH/16 Another patch from the Sieve maintainer.
2412 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2414 PH/18 If a system quota error occurred while trying to create the file for
2415 a maildir delivery, the message "Mailbox is full" was not appended to the
2416 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2417 only to a quota excession during the actual writing of the file.
2419 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2420 characters?) which causes problems in log lines. The DN values are now
2421 passed through string_printing() before being added to log lines.
2423 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2424 and InterBase are left for another time.)
2426 PH/21 Added message_body_newlines option.
2428 PH/22 Guard against possible overflow in moan_check_errorcopy().
2430 PH/23 POSIX allows open() to be a macro; guard against that.
2432 PH/24 If the recipient of an error message contained an @ in the local part
2433 (suitably quoted, of course), incorrect values were put in $domain and
2434 $local_part during the evaluation of errors_copy.
2440 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2441 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2444 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2445 issue a MAIL command.
2447 PH/02 In an ACL statement such as
2449 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2451 if a client was not listed at all, or was listed with a value other than
2452 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2453 the condition was not true (as it should be), so access was not denied.
2454 The bug was that the ! inversion was incorrectly passed on to the second
2455 item. This has been fixed.
2457 PH/03 Added additional dnslists conditions == and =& which are different from
2458 = and & when the dns lookup returns more than one IP address.
2460 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2461 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2463 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2464 FSYNC, which compiles an option called disable_fsync that allows for
2465 bypassing fsync(). The documentation is heavily laced with warnings.
2467 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2469 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2470 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2471 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2472 including adding "make clean"; (3) Added -fPIC when compiling the test
2473 dynamically loaded module, to get rid of a warning.
2475 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2476 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2477 = 0s. The bug is otherwise harmless.
2479 PH/07 There was a bug in the dovecot authenticator such that the value of
2480 $auth1 could be overwritten, and so not correctly preserved, after a
2481 successful authentication. This usually meant that the value preserved by
2482 the server_setid option was incorrect.
2484 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2486 PH/09 Installed PCRE release 7.0.
2488 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2489 run for batched SMTP input. It is now run at the start of every message
2490 in the batch. While fixing this I discovered that the process information
2491 (output by running exiwhat) was not always getting set for -bs and -bS
2492 input. This is fixed, and it now also says "batched" for BSMTP.
2494 PH/11 Added control=no_pipelining.
2496 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2497 patch, slightly modified), and move the expansion of helo_data till after
2498 the connection is made in the smtp transport (so it can use these
2501 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2503 PH/14 Added log_selector = +pid.
2505 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2507 PH/16 Add ${if forany and ${if forall.
2509 PH/17 Added dsn_from option to vary the From: line in DSNs.
2511 PH/18 Flush SMTP output before performing a callout, unless control =
2512 no_callout_flush is set.
2514 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2515 was true (the default) a successful delivery failed to delete the retry
2516 item, thus causing premature timeout of the address. The bug is now
2519 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2521 PH/21 Long custom messages for fakedefer and fakereject are now split up
2522 into multiline responses in the same way that messages for "deny" and
2523 other ACL rejections are.
2525 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2526 with slight modification.
2528 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2529 draft, changing the syntax and factoring some duplicate code".
2531 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2532 for deliveries of the second and subsequent messages over the same SMTP
2535 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2536 ${reduce, with only minor "tidies".
2538 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2540 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2541 expansion side effects.
2543 PH/27 When a message times out after an over-quota error from an Exim-imposed
2544 quota, the bounce message says "mailbox is full". This message was not
2545 being given when it was a system quota that was exceeded. It now should
2548 MH/03 Made $recipients available in local_scan(). local_scan() already has
2549 better access to the recipient list through recipients_list[], but
2550 $recipients can be useful in postmaster-provided expansion strings.
2552 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2553 in the case of a MAIL command with additional options following the
2554 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2555 were accidentally chopped off.
2557 PH/29 SMTP synchronization checks are implemented when a command is read -
2558 there is a check that no more input is waiting when there shouldn't be
2559 any. However, for some commands, a delay in an ACL can mean that it is
2560 some time before the response is written. In this time, more input might
2561 arrive, invalidly. So now there are extra checks after an ACL has run for
2562 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2563 pipelining has not been advertised.
2565 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2567 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2568 returned in the SMTP response when smtp_return_error_details was false.
2569 This has been fixed.
2571 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2572 instead of the C I/O that was originally supplied, because problems were
2573 reported on Solaris.
2575 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2576 Exim which did not show up earlier: it was assuming that a call to
2577 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2578 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2579 was a macro that became an assignment, so it seemed to work. This has
2580 changed to a proper function call with a void return, hence the compile
2581 error. Exim's code has been fixed.
2583 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2586 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2588 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2590 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2591 because it turns out to be too big or there is a timeout) there is no
2592 "Completed" line in the log. When some messages of this type were
2593 selected by exigrep, they were listed as "not completed". Others were
2594 picked up by some special patterns. I have improved the selection
2595 criteria to be more general.
2597 PH/38 The host_find_failed option in the manualroute router can now be set
2598 to "ignore", to completely ignore a host whose IP address cannot be
2599 found. If all hosts are ignored, the behaviour is controlled by the new
2600 host_all_ignored option.
2602 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2603 homing or because of multiple MX records with /mx) generated more than
2604 one IP address, and the following item turned out to be the local host,
2605 all the secondary addresses of the first item were incorrectly removed
2606 from the list, along with the local host and any following hosts (which
2607 is what is supposed to happen).
2609 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2610 whoever called Exim into the -H file. In the case of the daemon it was
2611 behaving confusingly. When first started, it used values for whoever
2612 started the daemon, but after a SIGHUP it used the Exim user (because it
2613 calls itself on a restart). I have changed the code so that it now always
2616 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2617 message are rejected with the same error (e.g. no authentication or bad
2618 sender address), and a DATA command is nevertheless sent (as can happen
2619 with PIPELINING or a stupid MUA), the error message that was given to the
2620 RCPT commands is included in the rejection of the DATA command. This is
2621 intended to be helpful for MUAs that show only the final error to their
2624 PH/42 Another patch from the Sieve maintainer.
2626 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2627 Eximstats - Fixed some broken HTML links and added missing column headers
2629 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2630 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2632 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2634 PH/43 Yet another patch from the Sieve maintainer.
2636 PH/44 I found a way to check for a TCP/IP connection going away before sending
2637 the response to the final '.' that terminates a message, but only in the
2638 case where the client has not sent further data following the '.'
2639 (unfortunately, this is allowed). However, in many cases there won't be
2640 any further data because there won't be any more messages to send. A call
2641 to select() can be used: if it shows that the input is "ready", there is
2642 either input waiting, or the socket has been closed. An attempt to read
2643 the next input character can distinguish the two cases. Previously, Exim
2644 would have sent an OK response which the client would never have see.
2645 This could lead to message repetition. This fix should cure that, at
2646 least in a lot of common cases.
2648 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2649 advertised in response to EHLO.
2655 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2656 fixed by 4.65/MH/01 (is this a record?) are fixed:
2658 (i) An empty string was always treated as zero by the numeric comparison
2659 operators. This behaviour has been restored.
2661 (ii) It is documented that the numeric comparison operators always treat
2662 their arguments as decimal numbers. This was broken in that numbers
2663 starting with 0 were being interpreted as octal.
2665 While fixing these problems I realized that there was another issue that
2666 hadn't been noticed. Values of message_size_limit (both the global option
2667 and the transport option) were treated as octal if they started with 0.
2668 The documentation was vague. These values are now always treated as
2669 decimal, and I will make that clear in the documentation.
2675 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2676 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2679 MH/01 Don't check that the operands of numeric comparison operators are
2680 integers when their expansion is in "skipping" mode (fixes bug
2681 introduced by 4.64-PH/07).
2683 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2684 child addresses, Exim now panics and dies. Previously, because the count
2685 is held in a short int, deliveries were likely to be lost. As such a
2686 large number of recipients for a single message is ridiculous
2687 (performance will be very, very poor), I have chosen to impose a limit
2688 rather than extend the field.
2694 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2695 leftover -K file (the existence of which was triggered by #402).
2696 While we were at it, introduced process PID as part of the -K
2697 filename. This should rule out race conditions when creating
2700 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2701 processing considerably. Previous code took too long for large mails,
2702 triggering a timeout which in turn triggers #401.
2704 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2705 in the DK code in transports.c. sendfile() is not really portable,
2706 hence the _LINUX specificness.
2708 TF/01 In the add_headers option to the mail command in an Exim filter,
2709 there was a bug that Exim would claim a syntax error in any
2710 header after the first one which had an odd number of characters
2713 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2714 callout verification, Exim cached a "reject" for the entire domain. This
2715 is correct for most verifications, but it is not correct for a recipient
2716 verification with use_sender or use_postmaster set, because in that case
2717 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2718 case of MAIL FROM:<> rejection from other early rejections (e.g.
2719 rejection of HELO). When verifying a recipient using a non-null MAIL
2720 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2721 Whatever the result of the callout, the value of the domain cache is
2722 left unchanged (for any other kind of callout, getting as far as trying
2723 RCPT means that the domain itself is ok).
2725 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2728 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2729 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2730 session, the daemon ignores ECONNECT errors and logs others; it now
2731 ignores EPIPE as well.
2733 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2734 (quoted-printable decoding).
2736 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2737 later the small subsequent patch to fix an introduced bug.
2739 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2741 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2743 PH/08 An error is now given if message_size_limit is specified negative.
2745 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2746 to be given (somewhat) arbitrary names.
2748 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2751 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2752 miscellaneous code fixes
2754 PH/10 Added the log_reject_target ACL modifier to specify where to log
2757 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2758 hostname. This is wrong, because it relates to the incoming message (and
2759 probably the interface on which it is arriving) and not to the outgoing
2760 callout (which could be using a different interface). This has been
2761 changed to use the value of the helo_data option from the smtp transport
2762 instead - this is what is used when a message is actually being sent. If
2763 there is no remote transport (possible with a router that sets up host
2764 addresses), $smtp_active_hostname is used.
2766 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2767 tweaks were necessary in order to get it to work (see also 21 below):
2768 (a) The code assumed that strncpy() returns a negative number on buffer
2769 overflow, which isn't the case. Replaced with Exim's string_format()
2771 (b) There were several signed/unsigned issues. I just did the minimum
2772 hacking in of casts. There is scope for a larger refactoring.
2773 (c) The code used strcasecmp() which is not a standard C function.
2774 Replaced with Exim's strcmpic() function.
2775 (d) The code set only $1; it now sets $auth1 as well.
2776 (e) A simple test gave the error "authentication client didn't specify
2777 service in request". It would seem that Dovecot has changed its
2778 interface. Fortunately there's a specification; I followed it and
2779 changed what the client sends and it appears to be working now.
2781 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2784 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2785 address A is aliased to B and C, where B exists and C does not. Without
2786 -v the output is "A verified" because verification stops after a
2787 successful redirection if more than one address is generated. However,
2788 with -v the child addresses are also verified. Exim was outputting "A
2789 failed to verify" and then showing the successful verification for C,
2790 with its parentage. It now outputs "B failed to verify", showing B's
2791 parentage before showing the successful verification of C.
2793 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2794 look up a TXT record in a specific list after matching in a combined
2797 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2798 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2799 they consult the DNS. I had assumed they would set it the way they
2800 wanted; and indeed my experiments on Linux seem to show that in some
2801 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2802 To be on the safe side, however, I have now made the interface to
2803 host_find_byname() similar to host_find_bydns(), with an argument
2804 containing the DNS resolver options. The host_find_byname() function now
2805 sets these options at its start, just as host_find_bydns() does. The smtp
2806 transport options dns_qualify_single and dns_search_parents are passed to
2807 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2808 of host_find_byname() use the default settings of RES_DEFNAMES
2809 (qualify_single) but not RES_DNSRCH (search_parents).
2811 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2812 spool_read_header() do less string testing, by means of a preliminary
2813 switch on the second character of optional "-foo" lines. (This is
2814 overdue, caused by the large number of possibilities that now exist.
2815 Originally there were few.) While I was there, I also converted the
2816 str(n)cmp tests so they don't re-test the leading "-" and the first
2817 character, in the hope this might squeeze out yet more improvement.
2819 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2820 flag allowing group syntax was set by the header_syntax check but not
2821 turned off, possible causing trouble later; (2) The flag was not being
2822 set at all for the header_verify test, causing "group"-style headers to
2823 be rejected. I have now set it in this case, and also caused header_
2824 verify to ignore an empty address taken from a group. While doing this, I
2825 came across some other cases where the code for allowing group syntax
2826 while scanning a header line wasn't quite right (mostly, not resetting
2827 the flag correctly in the right place). These bugs could have caused
2828 trouble for malformed header lines. I hope it is now all correct.
2830 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2831 with the "reply" argument non-NULL. The code, however (which originally
2832 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2833 but it didn't always do it. This confused somebody who was copying the
2834 code for some other use. I have removed all the tests.
2836 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2837 feature that was used to support insecure browsers during the U.S. crypto
2838 embargo. It requires special client support, and Exim is probably the
2839 only MTA that supported it -- and would never use it because real RSA is
2840 always available. This code has been removed, because it had the bad
2841 effect of slowing Exim down by computing (never used) parameters for the
2842 RSA_EXPORT functionality.
2844 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2845 authenticator to fail if there's a tab character in the incoming data
2846 (there should never be unless someone is messing about, as it's supposed
2847 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2848 if the connection is using TLS or if the remote IP is the same as the
2849 local IP, and the "valid-client-cert option" if a client certificate has
2852 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2853 authenticators. This can be used for authorization after authentication
2854 succeeds. (In the case of plaintext, it servers for both authentication
2857 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2858 if any retry times were supplied.
2860 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2861 connection, where there is no client IP address to check. In this
2862 situation, the verify now always succeeds.
2864 PH/25 Applied John Jetmore's -Mset patch.
2866 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2868 PH/27 In a string expansion for a processed (not raw) header when multiple
2869 headers of the same name were present, leading whitespace was being
2870 removed from all of them, but trailing whitespace was being removed only
2871 from the last one. Now trailing whitespace is removed from each header
2872 before concatenation. Completely empty headers in a concatenation (as
2873 before) are ignored.
2875 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2876 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2878 PH/29 [Removed. This was a change that I later backed out, and forgot to
2879 correct the ChangeLog entry (that I had efficiently created) before
2880 committing the later change.]
2882 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2883 address errors (4xx response to RCPT) over the same connection as other
2884 messages routed to the same hosts. Such deliveries are always "forced",
2885 so retry times are not inspected. This resulted in far too many retries
2886 for the affected addresses. The effect occurred only when there were more
2887 hosts than the hosts_max_try setting in the smtp transport when it had
2888 the 4xx errors. Those hosts that it had tried were not added to the list
2889 of hosts for which the message was waiting, so if all were tried, there
2890 was no problem. Two fixes have been applied:
2892 (i) If there are any address or message errors in an SMTP delivery, none
2893 of the hosts (tried or untried) are now added to the list of hosts
2894 for which the message is waiting, so the message should not be a
2895 candidate for sending over the same connection that was used for a
2896 successful delivery of some other message. This seems entirely
2897 reasonable: after all the message is NOT "waiting for some host".
2898 This is so "obvious" that I'm not sure why it wasn't done
2899 previously. Hope I haven't missed anything, but it can't do any
2900 harm, as the worst effect is to miss an optimization.
2902 (ii) If, despite (i), such a delivery is accidentally attempted, the
2903 routing retry time is respected, so at least it doesn't keep
2904 hammering the server.
2906 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2907 in ${readsocket because some servers need this prod.
2909 PH/32 Added some extra debug output when updating a wait-xxx database.
2911 PH/33 The hint "could be header name not terminated by colon", which has been
2912 given for certain expansion errors for a long time, was not being given
2913 for the ${if def:h_colon_omitted{... case.
2915 PH/34 The spec says: "With one important exception, whenever a domain list is
2916 being scanned, $domain contains the subject domain." There was at least
2917 one case where this was not true.
2919 PH/35 The error "getsockname() failed: connection reset by peer" was being
2920 written to the panic log as well as the main log, but it isn't really
2921 panic-worthy as it just means the connection died rather early on. I have
2922 removed the panic log writing for the ECONNRESET error when getsockname()
2925 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2926 runs only) independently of the message's sender address. This meant
2927 that, if the 4xx error was in fact related to the sender, a different
2928 message to the same recipient with a different sender could confuse
2929 things. In particular, this can happen when sending to a greylisting
2930 server, but other circumstances could also provoke similar problems.
2931 I have changed the default so that the retry time for these errors is now
2932 based a combination of the sender and recipient addresses. This change
2933 can be overridden by setting address_retry_include_sender=false in the
2936 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2937 remote server are returned as part of bounce messages. This was not
2938 happening for LMTP over a pipe (the lmtp transport), but now it is the
2939 same for both kinds of LMTP.
2941 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2942 in header lines that were in fact CNAMEs. This is no longer the case.
2944 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2945 and queue runs started by the daemon processed all messages. This has
2946 been fixed so that -R and -S can now usefully be given with -q<time>.
2948 PH/40 Import PCRE release 6.7 (fixes some bugs).
2950 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2952 PH/42 Give an error if -q is specified more than once.
2954 PH/43 Renamed the variables $interface_address and $interface_port as
2955 $received_ip_address and $received_port, to make it clear that these
2956 values apply to message reception, and not to the outgoing interface when
2957 a message is delivered. (The old names remain recognized, of course.)
2959 PH/44 There was no timeout on the connect() call when using a Unix domain
2960 socket in the ${readsocket expansion. There now is.
2962 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2963 be meaningful with "accept".
2965 SC/01 Eximstats V1.43
2966 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2968 SC/02 Eximstats V1.44
2969 Use a glob alias rather than an array ref in the generated
2970 parser. This improves both readability and performance.
2972 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2973 Collect SpamAssassin and rejection statistics.
2974 Don't display local sender or destination tables unless
2975 there is data to show.
2976 Added average volumes into the top table text output.
2978 SC/04 Eximstats V1.46
2979 Collect data on the number of addresses (recipients)
2980 as well as the number of messages.
2982 SC/05 Eximstats V1.47
2983 Added 'Message too big' to the list of mail rejection
2984 reasons (thanks to Marco Gaiarin).
2986 SC/06 Eximstats V1.48
2987 Mainlog lines which have GMT offsets and are too short to
2988 have a flag are now skipped.
2990 SC/07 Eximstats V1.49 (Alain Williams)
2991 Added the -emptyok flag.
2993 SC/08 Eximstats V1.50
2994 Fixes for obtaining the IP address from reject messages.
2996 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2997 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2998 whitespace changes from 4.64-PH/27
3000 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
3003 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
3004 are found, allow negative numbers in numeric criteria)
3006 JJ/06 exipick.20061117.2, added new $message_body_missing variable
3008 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
3009 to match changes made in 4.64-PH/43
3011 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
3013 PH/47 Put in an explicit test for a DNS lookup of an address record where the
3014 "domain" is actually an IP address, and force a failure. This locks out
3015 those revolvers/nameservers that support "A-for-A" lookups, in
3016 contravention of the specifications.
3018 PH/48 When a host name was looked up from an IP address, and the subsequent
3019 forward lookup of the name timed out, the host name was left in
3020 $sender_host_name, contrary to the specification.
3022 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3023 restricted to single-key lookups, Exim was not diagnosing an error if
3024 * or *@ was used with a query-style lookup.
3026 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3028 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3029 long ago, but noone interested enough thought of it. Let's just say that
3030 the "1.1" means that there are some new functions that weren't there at
3031 some point in the past.
3033 PH/51 Error processing for expansion failure of helo_data from an smtp
3034 transport during callout processing was broken.
3036 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3037 tested/used via the -bh/-bhc/-bs options.
3039 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3040 bug, fixed in subsequent PCRE releases).
3042 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3043 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3045 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3051 SC/01 Use a glob alias rather than an array ref in eximstats generated
3052 parser. This improves both readability and performance.
3054 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3055 Don't display local sender or destination tables in eximstats unless
3056 there is data to show.
3057 Added average volumes into the eximstats top table text output.
3059 SC/03 Collect data on the number of addresses (recipients) as well
3060 as the number of messages in eximstats.
3062 TF/01 Correct an error in the documentation for the redirect router. Exim
3063 does (usually) call initgroups() when daemonizing.
3065 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3066 with consistent privilege compared to when running as a daemon.
3068 TF/03 Note in the spec that $authenticated_id is not set for local
3069 submissions from trusted users.
3071 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3072 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3074 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3075 by adding some example configuration directives to the default
3076 configuration file. A little bit of work is required to uncomment the
3077 directives and define how usernames and passwords are checked, but
3078 there is now a framework to start from.
3080 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3081 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3082 without this. I don't know how relevant this is to other LDAP libraries.
3084 PH/02 Add the verb name to the "unknown ACL verb" error.
3086 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3088 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3090 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3091 directory not expanded when it should be if an expanded home directory
3092 was set for the address (which is overridden by the transport).
3094 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3097 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3098 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3099 because it is too late at that time, and has no effect.
3101 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3102 security issue with \' (bugzilla #107). I could not use the
3103 PQescapeStringConn() function, because it needs a PGconn value as one of
3106 PH/08 When testing addresses using -bt, indicate those final addresses that
3107 are duplicates that would not cause an additional delivery. At least one
3108 person was confused, thinking that -bt output corresponded to deliveries.
3109 (Suppressing duplicates isn't a good idea as you lose the information
3110 about possibly different redirections that led to the duplicates.)
3112 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3113 systems where poll() doesn't work, in particular OS X.
3115 PH/10 Added more information to debugging output for retry time not reached.
3117 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3118 operations in malware.c.
3120 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3123 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3124 syslog_duplication set false, log lines that would normally be written
3125 both the the main log and to the reject log were not written to syslog at
3128 PH/14 In the default configuration, change the use of "message" in ACL warn
3129 statements to "add_header".
3131 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3132 not followed by a command (e.g. "seen endif").
3134 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3135 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3138 PH/17 Added extra conditions to the default value of delay_warning_condition
3142 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3143 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3144 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3147 The Auto-Submitted: and various List- headers are standardised, whereas I
3148 don't think Precedence: ever was.
3150 PH/18 Refactored debugging code in route_finduser() to show more information,
3151 in particular, the error code if getpwnam() issues one.
3153 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3154 This is apparently needed in addition to the PH/07 change above to avoid
3155 any possible encoding problems.
3157 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3158 but not after initializing Perl.
3160 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3161 output them only if debugging. By default they are written stderr,
3162 apparently, which is not desirable.
3164 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3167 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3170 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3172 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3173 authenticated or an ident call has been made. Suppress the default
3174 values for $authenticated_id and $authenticated_sender (but permit -oMai
3175 and -oMas) when testing with -bh.
3177 PH/24 Re-jigged the order of the tests in the default configuration so that the
3178 tests for valid domains and recipients precede the DNS black list and CSA
3179 tests, on the grounds that those ones are more expensive.
3181 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3182 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3183 This bug exists in every version of Exim that I still have, right back to
3186 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3187 However, an attempt to turn on case-sensitivity in a regex key by
3188 including (?-i) didn't work because the subject string was already
3189 lowercased, and the effects were non-intuitive. It turns out that a
3190 one-line patch can be used to allow (?-i) to work as expected.
3196 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3197 other effects) broke the use of negated acl sub-conditions.
3199 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3202 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3203 "Deny" causes Exim to reject the incoming connection with a 554 error.
3204 Unfortunately, if there is a major crisis, such as a disk failure,
3205 tcp-wrappers gives "deny", whereas what one would like would be some
3206 kind of temporary error. A kludge has been added to help with this.
3207 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3208 554 error is used if errno is still zero or contains ENOENT (which occurs
3209 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3212 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3214 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3215 errors. Otherwise a message that provokes a temporary error (when other
3216 messages do not) can cause a whole host to time out.
3218 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3219 addresses were routed directly to files or pipes from a redirect router.
3220 File deliveries just didn't batch; pipe deliveries might have suffered
3223 PH/06 A failure to get a lock for a hints database would erroneously always say
3224 "Failed to get write lock", even when it was really a read lock.
3226 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3227 of 0600. This has been changed to use the value of the lockfile_mode
3228 option (which defaults to 0600).
3230 PH/08 Applied small patch from the Sieve maintainer.
3232 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3233 folder from quota calculations, a direct delivery into this folder messed
3234 up the contents of the maildirsize file. This was because the regex was
3235 used only to exclude .Trash (or whatever) when the size of the mailbox
3236 was calculated. There was no check that a delivery was happening into an
3237 excluded directory. This bug has been fixed by ignoring all quota
3238 processing for deliveries into excluded directories.
3240 PH/10 Added the maildirfolder_create_regex option to appendfile.
3246 PH/01 The code for finding all the local interface addresses on a FreeBSD
3247 system running IPv6 was broken. This may well have applied to all BSD
3248 systems, as well as to others that have similar system calls. The broken
3249 code found IPv4 interfaces correctly, but gave incorrect values for the
3250 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3251 that it would not match correctly against @[] and not recognize the IPv6
3254 PH/02 The ipliteral router was not recognizing addresses of the form user@
3255 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3257 PH/03 Added disable_ipv6.
3259 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3260 decoded form, because it is most often used to construct To: headers
3261 lines in autoreplies, and the decoded form may well be syntactically
3262 invalid. However, $reply_address has leading white space removed, and all
3263 newlines turned into spaces so that the autoreply transport does not
3266 PH/05 If group was specified without a user on a router, and no group or user
3267 was specified on a transport, the group from the router was ignored.
3269 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3270 for visible compile-time settings that can be used to change these
3271 numbers, for those that want even more. Backwards compatibility with old
3272 spool files has been maintained. However, going back to a previous Exim
3273 release will lost any variables that are in spool files.
3275 PH/07 Two small changes when running in the test harness: increase delay when
3276 passing a TCP/IP connection to a new process, in case the original
3277 process has to generate a bounce, and remove special handling of
3278 127.0.0.2 (sic), which is no longer necessary.
3280 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3281 be the same on different OS.
3283 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3286 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3287 whether --show-vars was specified or not
3289 JJ/02 exipick: Added support for new ACL variable spool format introduced
3292 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3293 syntactically invalid From: or Reply-to: line, and a filter used this to
3294 generate an autoreply, and therefore failed to obtain an address for the
3295 autoreply, Exim could try to deliver to a non-existent relative file
3296 name, causing unrelated and misleading errors. What now happens is that
3297 it logs this as a hard delivery error, but does not attempt to create a
3300 PH/11 The exinext utility has a -C option for testing purposes, but although
3301 the given file was scanned by exinext itself; it wasn't being passed on
3302 when Exim was called.
3304 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3305 an end-of-file indication when reading a command response.
3307 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3308 compiled. In many other places in Exim, IPv6 addresses are always
3309 recognized, so I have changed this. It also means that IPv4 domain
3310 literals of the form [IPV4:n.n.n.n] are now always recognized.
3312 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3313 used if the router is not running as root, for example, when verifying at
3314 ACL time, or when using -bh. The debugging output from this situation was
3315 non-existent - all you got was a failure to exec. I have made two
3318 (a) Failures to set uid/gid, the current directory, or a process leader
3319 in a subprocess such as that created by queryprogram now generate
3320 suitable debugging output when -d is set.
3322 (b) The queryprogram router detects when it is not running as root,
3323 outputs suitable debugging information if -d is set, and then runs
3324 the subprocess without attempting to change uid/gid.
3326 PH/15 Minor change to Makefile for building test_host (undocumented testing
3329 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3330 additional section of a DNS packet that returns MX or SRV records.
3331 Instead, it always explicitly searches for A/AAAA records. This avoids
3332 major problems that occur when a DNS server includes only records of one
3333 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3334 fixed another bug: if SRV records were looked up and the corresponding
3335 address records were *not* found in the additional section, the port
3336 values from the SRV records were lost.
3338 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3339 using the correct key (the original address) when searching the retry
3340 rules in order to find which one to use for generating the retry hint.
3342 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3343 adding the default one. Similarly, if it contains a Reply-To: header, the
3344 errors_reply_to option, if set, is not used.
3346 PH/19 When calculating a retry time, Exim used to measure the "time since
3347 failure" by looking at the "first failed" field in the retry record. Now
3348 it does not use this if it is later than than the arrival time of the
3349 message. Instead it uses the arrival time. This makes for better
3350 behaviour in cases where some deliveries succeed, thus re-setting the
3351 "first failed" field. An example is a quota failure for a huge message
3352 when small messages continue to be delivered. Without this change, the
3353 "time since failure" will always be short, possible causing more frequent
3354 delivery attempts for the huge message than are intended.
3355 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3357 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3358 $1, $2, $3) because the numerical variables can be reset during some
3359 expansion items (e.g. "match"), thereby losing the authentication data.
3361 PH/21 Make -bV show the size of off_t variables so that the test suite can
3362 decide whether to run tests for quotas > 2G.
3364 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3365 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3366 filecount value is greater than 2G or if a quota value is greater than 2G
3367 on a system where the size of off_t is not greater than 4, a panic error
3370 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3371 never match. The debug and -bh output now contains an explicit error
3372 message indicating a malformed IPv4 address or mask.
3374 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3375 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3376 PH/23 above applies.
3378 PH/25 Do not write to syslog when running in the test harness. The only
3379 occasion when this arises is a failure to open the main or panic logs
3380 (for which there is an explicit test).
3382 PH/26 Added the /no_tell option to "control=freeze".
3384 PH/27 If a host name lookup failed very early in a connection, for example, if
3385 the IP address matched host_lookup and the reverse lookup yielded a name
3386 that did not have a forward lookup, an error message of the form "no IP
3387 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3388 could be logged. Now it outputs the IP address instead of "NULL".
3390 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3391 allows the sender and the authenticated sender to be set when
3392 submitting a message from within Exim. Since child_open_exim() is
3393 documented for local_scan(), the new function should be too.
3395 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3396 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3397 results in an empty string is now treated as unset.
3399 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3401 PH/31 Added sender_verify_fail logging option.
3403 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3404 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3405 tidied the source and removed it altogether.
3407 PH/33 When a queue run was abandoned because the load average was too high, a
3408 log line was always written; now it is written only if the queue_run log
3409 selector is set. In addition, the log line for abandonment now contains
3410 information about the queue run such as the pid. This is always present
3411 in "start" and "stop" lines but was omitted from the "abandon" line.
3413 PH/34 Omit spaces between a header name and the colon in the error message that
3414 is given when verify = headers_syntax fails (if there are lots of them,
3415 the message gets confusing).
3417 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3418 names, as there are now some PTR records that contain slashes. This check
3419 is only to protect against broken name servers that fall over on strange
3420 characters, so the fact that it applies to all lookups doesn't matter.
3422 PH/36 Now that the new test suite is complete, we can remove some of the
3423 special code in Exim that was needed for the old test suite. For example,
3424 sorting DNS records because real resolvers return them in an arbitrary
3425 order. The new test suite's fake resolver always returns records in the
3428 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3429 bounces) except when queue_only is set, to avoid logging races between
3430 the different processes.
3432 PH/38 Panic-die if .include specifies a non-absolute path.
3434 PH/39 A tweak to the "H" retry rule from its user.
3436 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3437 a label. They prevented compilation on older perls.
3439 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3440 a warning to be raised on newish perls.
3442 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3443 on queue. Changes to match documented behaviour of showing count of
3444 messages matching specified criteria.
3446 PH/40 Changed the default ident timeout from 30s to 5s.
3448 PH/41 Added support for the use of login_cap features, on those BSD systems
3449 that have them, for controlling the resources used by pipe deliveries.
3451 PH/42 The content-scanning code uses fopen() to create files in which to put
3452 message data. Previously it was not paying any attention to the mode of
3453 the files. Exim runs with umask(0) because the rest of the code creates
3454 files with open(), and sets the required mode explicitly. Thus, these
3455 files were ending up world-writeable. This was not a big issue, because,
3456 being within the spool directory, they were not world-accessible. I have
3457 created a function called modefopen, which takes an additional mode
3458 argument. It sets umask(777), creates the file, chmods it to the required
3459 mode, then resets the umask. All the relevant calls to fopen() in the
3460 content scanning code have been changed to use this function.
3462 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3463 to 24 hours. This avoids potential overflow problems when processing G
3464 and H retry rules. I suspect nobody ever tinkers with this value.
3466 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3468 PH/45 When the plaintext authenticator is running as a client, the server's
3469 challenges are checked to ensure they are valid base64 strings. By
3470 default, the authentication attempt is cancelled if an invalid string is
3471 received. Setting client_ignore_invalid_base64 true ignores these errors.
3472 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3473 they are received. Thus, the responses can be made to depend on the
3474 challenges. If an invalid string is ignored, an empty string is placed in
3477 PH/46 Messages that are created by the autoreply transport now contains a
3478 References: header, in accordance with RFCs 2822 and 3834.
3480 PH/47 Added authenticated_sender_force to the smtp transport.
3482 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3484 PH/49 Installed latest patch from the Sieve maintainer.
3486 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3487 was also set, the appendfile transport was unnecessarily scanning a
3488 directory of message files (e.g. for maildir delivery) to find the count
3489 of files (along with the size), even though it did not need this
3490 information. It now does the scan only if it needs to find either the
3491 size of the count of files.
3493 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3495 PH/52 Two bugs concerned with error handling when the smtp transport is
3498 (i) Exim was not creating retry information for temporary errors given
3499 for individual recipients after the DATA command when the smtp transport
3500 was used in LMTP mode. This meant that they could be retried too
3501 frequently, and not timed out correctly.
3503 (ii) Exim was setting the flag that allows error details to be returned
3504 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3505 recipients that were returned after the DATA command.
3507 PH/53 This is related to PH/52, but is more general: for any failing address,
3508 when detailed error information was permitted to be returned to the
3509 sender, but the error was temporary, then after the final timeout, only
3510 "retry timeout exceeded" was returned. Now it returns the full error as
3511 well as "retry timeout exceeded".
3513 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3514 do this, and (what is worse) MTAs that accept it.
3516 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3517 will now be deprecated.
3519 PH/56 New os.c-cygwin from the Cygwin maintainer.
3521 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3522 formats (previously only available in exim formats via -bpr, -bpru,
3523 and -bpra. Now also available in native and exiqgrep formats)
3525 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3526 with very large, slow to parse queues
3528 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3530 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3532 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3533 responses to authentication challenges, though it was showing the
3534 challenges; (ii) I've removed the CR characters from the debug output for
3537 PH/58 Allow for the insertion of a newline as well as a space when a string
3538 is turned into more than one encoded-word during RFC 2047 encoding. The
3539 Sieve code now uses this.
3541 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3542 data_4xx, lost_connection, tls_required.
3544 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3545 message was being sent as an SMTP response.
3547 PH/61 Add -l and -k options to exicyclog.
3549 PH/62 When verifying, if an address was redirected to one new address, so that
3550 verification continued, and the new address failed or deferred after
3551 having set something in $address_data, the value of $address_data was not
3552 passed back to the ACL. This was different to the case when no
3553 redirection occurred. The value is now passed back in both cases.
3555 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3556 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3557 use login_cap.h, so on its own it isn't the distinguishing feature. The
3558 new name refers directly to the setclassresources() function.
3560 PH/65 Added configuration files for NetBSD3.
3562 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3564 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3565 is preferred over IPv4.
3567 PH/68 The bounce_return_message and bounce_return_body options were not being
3568 honoured for bounces generated during the reception of non-SMTP messages.
3569 In particular, this applied to messages rejected by the ACL. This bug has
3570 been fixed. However, if bounce_return_message is true and bounce_return_
3571 body is false, the headers that are returned for a non-SMTP message
3572 include only those that have been read before the error was detected.
3573 (In the case of an ACL rejection, they have all been read.)
3575 PH/69 The HTML version of the specification is now built in a directory called
3576 spec_html instead of spec.html, because the latter looks like a path with
3577 a MIME-type, and this confuses some software.
3579 PH/70 Catch two compiler warnings in sieve.c.
3581 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3582 function verify_get_ident() calls ip_connect() to connect a socket, but
3583 if the "connect()" function timed out, ip_connect() used to close the
3584 socket. However, verify_get_ident() also closes the socket later, and in
3585 between Exim writes to the log, which may get opened at this point. When
3586 the socket was closed in ip_connect(), the log could get the same file
3587 descriptor number as the socket. This naturally causes chaos. The fix is
3588 not to close the socket in ip_connect(); the socket should be closed by
3589 the function that creates it. There was only one place in the code where
3590 this was missing, in the iplookup router, which I don't think anybody now
3591 uses, but I've fixed it anyway.
3593 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3594 well as to direct DNS lookups. Otherwise the handling of names in host
3595 lists is inconsistent and therefore confusing.
3601 PH/01 Two changes to the default runtime configuration:
3603 (1) Move the checks for relay_from_hosts and authenticated clients from
3604 after to before the (commented out) DNS black list checks.
3606 (2) Add control=submission to the relay_from_hosts and authenticated
3607 clients checks, on the grounds that messages accepted by these
3608 statements are most likely to be submissions.
3610 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3612 (1) Generate an error if the third argument for the ${prvs expansion is
3615 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3618 (3) Reset the variables that are obtained from the first argument of
3619 ${prvscheck and used in the second argument before leaving the code,
3620 because their memory is reclaimed, so using them afterwards may do
3623 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3624 one (it's much easier than Tom thought :-).
3626 (5) Because of (4), we can now allow for the use of $prvscheck_result
3627 inside the third argument.
3629 PH/03 For some reason, the default setting of PATH when running a command from
3630 a pipe transport was just "/usr/bin". I have changed it to
3633 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3634 anything to be listed in the output from -bV.
3636 PH/05 When a filter generated an autoreply, the entire To: header line was
3637 quoted in the delivery log line, like this:
3639 => >A.N.Other <ano@some.domain> <original@ddress> ...
3641 This has been changed so that it extracts the operative address. There
3642 may be more than one such address. If so, they are comma-separated, like
3645 => >ano@some.domain,ona@other.domain <original@ddress> ...
3647 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3648 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3649 not being looked up in the rDNS to get a host name, Exim was showing the
3650 IP address twice in Received: lines, even though the IP addresses were
3651 identical. For example:
3653 Received: from [1.2.3.4] (helo=[1.2.3.4])
3655 However, if the real host name was known, it was omitting the HELO data
3656 if it matched the actual IP address. This has been tidied up so that it
3657 doesn't show the same IP address twice.
3659 PH/07 When both +timestamp and +memory debugging was on, the value given by
3660 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3661 called by the debug printing, thereby overwriting the timestamp buffer.
3662 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3664 PH/08 When the original message was included in an autoreply transport, it
3665 always said "this is a copy of the message, including all the headers",
3666 even if body_only or headers_only was set. It now gives an appropriate
3669 PH/09 Applied a patch from the Sieve maintainer which:
3671 o fixes some comments
3672 o adds the (disabled) notify extension core
3673 o adds some debug output for the result of if/elsif tests
3674 o points to the current vacation draft in the documentation
3675 and documents the missing references header update
3679 o fixes a bug in processing the envelope test (when testing
3680 multiple envelope elements, the last element determined the
3683 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3684 Electronic Mail") by including:
3686 Auto-submitted: auto-generated
3688 in the messages that it generates (bounce messages and others, such as
3689 warnings). In the case of bounce messages for non-SMTP messages, there was
3690 also a typo: it was using "Auto_submitted" (underscore instead of
3691 hyphen). Since every message generated by Exim is necessarily in response
3692 to another message, thes have all been changed to:
3694 Auto-Submitted: auto-replied
3696 in accordance with these statements in the RFC:
3698 The auto-replied keyword:
3700 - SHOULD be used on messages sent in direct response to another
3701 message by an automatic process,
3703 - MUST NOT be used on manually-generated messages,
3705 - MAY be used on Delivery Status Notifications (DSNs) and Message
3706 Disposition Notifications (MDNs),
3708 - MUST NOT be used on messages generated by automatic or periodic
3709 processes, except for messages which are automatic responses to
3712 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3713 to the default Received: header definition.
3715 PH/12 Added log selector acl_warn_skipped (default on).
3717 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3718 variables because (a) they are in the wrong storage pool and (b) even if
3719 they were copied, it wouldn't work properly because of the caching.
3721 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3722 checking when decoding. Apparently there are clients that generate
3723 overlong encoded strings. Why am I not surprised?
3725 PH/15 If the first argument of "${if match_address" was not empty, but did not
3726 contain an "@" character, Exim crashed. Now it writes a panic log message
3727 and treats the condition as false.
3729 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3731 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3732 extension "envelope-auth". The code is finished and in agreement with
3733 other implementations, but there is no documentation so far and in fact,
3734 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3735 not changing the active code.
3737 Print executed "if" and "elsif" statements when debugging is used. This
3738 helps a great deal to understand what a filter does.
3740 Document more things not specified clearly in RFC3028. I had all this
3741 sorted out, when out of a sudden new issues came to my mind. Oops."
3743 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3746 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3747 canonical form (as documented). However, after a host name lookup from
3748 the IP address, check_host() was doing a simple string comparison with
3749 addresses acquired from the DNS when checking that the found name did
3750 have the original IP as one of its addresses. Since any found IPv6
3751 addresses are likely to be in abbreviated form, the comparison could
3752 fail. Luckily, there already exists a function for doing the comparison
3753 by converting both addresses to binary, so now that is used instead of
3754 the text comparison.
3756 PH/20 There was another similar case to PH/19, when a complete host name was
3757 given in a host list; looking up its IP address could give an abbreviated
3758 form, whereas the current host's name might or might not be abbreviated.
3759 The same fix has been applied.
3765 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3766 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3769 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3770 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3772 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3774 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3775 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3776 Also, for "Auto-Submitted", treat anything other than "no" as
3777 non-personal, in accordance with RFC 3834. (Previously it treated
3778 anything starting "auto-" as non-personal.)
3780 TF/01 The control=submission/name=... option had a problem with syntax
3781 errors if the name included a slash character. The /name= option
3782 now slurps the rest of the string, so it can include any characters
3783 but it must come last in the list of options (after /sender_retain
3786 PH/05 Some modifications to the interface to the fake nameserver for the new
3794 TK/01 Added the "success_on_redirect" address verification option. See
3795 NewStuff for rationale and an example.
3797 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3799 PH/02 Patch to exigrep to allow it to work on syslog lines.
3801 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3802 fread() to skip over the body file's header line, because in Cygwin the
3803 header line is locked and is inaccessible.
3805 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3806 co-exist for some time) to make it clear that it is the Exim ID that is
3807 referenced, not the Message-ID: header line.
3809 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3810 string_format() function, because snprintf() does not exist on all
3813 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3814 ${stat: expansion item.
3816 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3817 protocol synchronization error", to keep the pedants happy.
3819 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3820 well as for IRIX systems, when gcc is being used. See the host.c source
3823 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3825 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3828 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3829 transport and to the smtp transport in LMTP mode.
3831 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3833 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3834 run a filter in a subprocess. This could lead to confusion in subsequent
3835 lookups in the parent process. There should also be a search_tidyup() at
3836 the end of the subprocess.
3838 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3839 only if the host matched helo_try_verify_hosts, which caused the
3840 verification to occur when the EHLO/HELO command was issued. The ACL just
3841 tested the remembered result. Now, if a previous verification attempt has
3842 not happened, "verify = helo" does it there and then.
3844 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3846 TK/03 Fix log output including CR from clamd.
3848 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3849 exist provoked a memory error which could cause a segfault.
3851 PH/15 Installed PCRE 6.2
3853 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3855 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3856 of the problem. Specifically, suggested +O2 rather than +O1 for the
3859 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3861 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3862 record, the port was not correctly logged when the outgoing_port log
3863 selector was set (it logged the transort's default port).
3865 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3866 fallback_hosts, and "hosts" in the smtp transport.
3868 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3869 host errors such as "Connection refused".
3871 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3872 authentication with radiusclient 0.4.9:
3874 - Error returned from rc_read_config was caught wrongly
3875 - Username/password not passed on to radius server due to wrong length.
3877 The presumption is that some radiusclient API changes for 4.51/PH/17
3878 were not taken care of correctly. The code is still untested by me (my
3879 Linux distribution still has 0.3.2 of radiusclient), but it was
3880 contributed by a Radius user.
3882 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3883 expanding the "port" option of the smtp transport.
3885 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3886 while reading a MIME header. Thanks to Tom Hughes for a patch.
3888 PH/24 Include config.h inside local_scan.h so that configuration settings are
3891 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3892 that in an ACL for RCPT (for example), you can examine exactly what was
3895 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3896 commands, but it was not correctly comparing the address with the actual
3897 client host address. Thus, it would show the EHLO address in Received:
3898 header lines when this was not necessary.
3900 PH/27 Added the % operator to ${eval:}.
3902 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3903 it should be ignoring failures (because with -C, for example, it has lost
3904 privilege). It wasn't ignoring creation failures other than "already
3907 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3910 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3911 because an input line was too long, either on its own, or by virtue of
3912 too many continuations, the temporary file was not being removed, and the
3913 return code was incorrect.
3915 PH/31 Missing "BOOL" in function definition in filtertest.c.
3917 PH/32 Applied Sieve patches from the maintainer.
3919 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3921 PH/33 Added "verify = not_blind".
3923 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3924 Local/Makefile (with some defaults set). These are used in built scripts
3925 such as exicyclog, but they have never been used in the exim_install
3926 script (though there are many overriding facilities there). I have
3927 arranged that the exim_install script now takes note of these two
3930 PH/35 Installed configuration files for Dragonfly.
3932 PH/36 When a locally submitted message by a trusted user did not contain a
3933 From: header, and the sender address was obtained from -f or from an SMTP
3934 MAIL command, and the trusted user did not use -F to supply a sender
3935 name, $originator_name was incorrectly used when constructing a From:
3936 header. Furthermore, $originator_name was used for submission mode
3937 messages from external hosts without From: headers in a similar way,
3938 which is clearly wrong.
3940 PH/37 Added control=suppress_local_fixups.
3942 PH/38 When log_selector = +received_sender was set, and the addition of the
3943 sender made the log line's construction buffer exactly full, or one byte
3944 less than full, an overflow happened when the terminating "\n" was
3947 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3948 when the result of a list match is failure because a DNS lookup failed.
3950 PH/40 RM_COMMAND is now used in the building process.
3952 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3953 the "build-* directories that it finds.
3955 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3956 address in a domain literal was a prefix of an interface address.
3958 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3959 when verifying a sender address, unless rewrite_headers is false.
3961 PH/44 Wrote a long comment about why errors_to addresses are verified as
3962 recipients, not senders.
3964 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3965 the ratelimit ACL was added.
3967 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3969 PH/46 Added extra information about PostgreSQL errors to the error string.
3971 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3972 avoiding the need to install special zones in a real server. This is
3973 backwards compatible; if it can't find the fake resolver, it drops back.
3974 Thus, both old and new test suites can be run.
3976 TF/02 Added util/ratelimit.pl
3978 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3979 clock is set back in time.
3981 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3982 Candler <B.Candler@pobox.com>.
3984 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3985 OK for addresses that are the result of redirections.
3987 PH/48 A number of further additions for the benefit of the new test suite,
3988 including a fake gethostbyname() that interfaces to the fake DNS resolver
3991 TF/06 The fix for widen_domains has also been applied to qualify_single and
3992 search_parents which are the other dnslookup options that can cause
3995 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3998 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
4000 TF/07 Exim produced the error message "an SRV record indicated no SMTP
4001 service" if it encountered an MX record with an empty target hostname.
4002 The message is now "an MX or SRV record indicated no SMTP service".
4004 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
4005 if the DNS of the sending site is misconfigured. This is quite a
4006 common situation. This change restores the behaviour of treating a
4007 helo verification defer as a failure.
4009 PH/51 If self=fail was set on a router, the bounce message did not include the
4010 actual error message.
4016 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
4018 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
4019 command itself ended in error, the underlying message about the transport
4020 filter timeout was being overwritten with the pipe command error. Now the
4021 underlying error message should be appended to the second error message.
4023 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4025 PH/02 Reduce the amount of output that "make" produces by default. Full output
4026 can still be requested.
4028 PH/03 The warning log line about a condition test deferring for a "warn" verb
4029 was being output only once per connection, rather than after each
4030 occurrence (because it was using the same function as for successful
4031 "warn" verbs). This seems wrong, so I have changed it.
4033 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4034 it should not have, which might have caused a crash in the right
4035 circumstances, but probably never did.
4037 PH/04 Installed a modified version of Tony Finch's patch to make submission
4038 mode fix the return path as well as the Sender: header line, and to
4039 add a /name= option so that you can make the user's friendly name appear
4042 TF/03 Added the control = fakedefer ACL modifier.
4044 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4045 Mark Lowes for thorough testing.
4047 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4049 TK/03 Merged latest SRS patch from Miles Wilton.
4051 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4052 with the definition in sysexits.h (which is #included earlier).
4053 Fortunately, Exim does not actually use EX_OK. The code used to try to
4054 preserve the sysexits.h value, by assuming that macro definitions were
4055 scanned for macro replacements. I have been disabused of this notion,
4056 so now the code just undefines EX_OK before #including unistd.h.
4058 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4059 in the smtp transport. When a block could not be written in a single
4060 write() function, the timeout was being re-applied to each part-write.
4061 This seems wrong - if the receiver was accepting one byte at a time it
4062 would take for ever. The timeout is now adjusted when this happens. It
4063 doesn't have to be particularly precise.
4065 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4066 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4068 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4069 without a domain if the check to <postmaster@domain> fails.
4071 SC/01 Eximstats: added -xls and the ability to specify output files
4072 (patch written by Frank Heydlauf).
4074 SC/02 Eximstats: use FileHandles for outputting results.
4076 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4078 SC/04 Eximstats: fixed display of large numbers with -nvr option
4080 SC/05 Eximstats: fixed merging of reports with empty tables.
4082 SC/06 Eximstats: added the -include_original_destination flag
4084 SC/07 Eximstats: removed tabs and trailing whitespace.
4086 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4088 TK/06 MBOX spool code: Add real "From " MBOX separator line
4089 so the .eml file is really in mbox format (even though
4090 most programs do not really care). Patch from Alex Miller.
4092 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4093 The latter is generated from $received_to and is only set if the
4094 message has one envelope recipient. SA can use these headers,
4095 obviously out-of-the-box. Patch from Alex Miller.
4097 PH/08 The ${def test on a variable was returning false if the variable's
4098 value was "0", contrary to what the specification has always said!
4099 The result should be true unless the variable is empty.
4101 PH/09 The syntax error of a character other than { following "${if
4102 def:variable_name" (after optional whitespace) was not being diagnosed.
4103 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4104 accidental colon was present, for example, could give incorrect results.
4106 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4107 result is used (not including appendfile, where other changes are about
4110 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4111 This involved changing a lot of size variables from int to off_t. It
4112 should work with maildirs and everything.
4114 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4115 spamd dying while we are connected to it.
4117 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4120 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4121 function for BDB 4.3.
4123 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4125 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4126 the sources for PCRE have been re-arranged and more files are now
4129 PH/15 The code I had for printing potentially long long variables in PH/11
4130 above was not the best (it lost precision). The length of off_t variables
4131 is now inspected at build time, and an appropriate printing format (%ld
4132 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4133 to be "long long int" or "long int". This is needed for the internal
4134 formatting function string_vformat().
4136 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4137 the configuration file to be ":syslog", then the script "guesses" where
4138 the logs files are, rather than using the compiled in default. In our
4139 case the guess is not the same as the compiled default, so the script
4140 suddenly stopped working when I started to use syslog. The patch checks
4141 to see if log_file_path is "". If so, it attempts to read it from exim
4142 with no configuration file to get the compiled in version, before it
4143 falls back to the previous guessing code."
4145 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4146 implementing BATV in an Exim configuration. See NewStuff for the gory
4149 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4150 Makefile that are specific to HP-UX.
4152 PH/18 If the "use_postmaster" option was set for a recipient callout together
4153 with the "random" option, the postmaster address was used as the MAIL
4154 FROM address for the random test, but not for the subsequent recipient
4155 test. It is now used for both.
4157 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4158 patch removes a few documentation additions to RFC 3028, because the
4159 latest draft now contains them. It adds the new en;ascii-case comparator
4160 and a new error check for 8bit text in MIME parts. Comparator and
4161 require names are now matched exactly. I enabled the subaddress
4162 extension, but it is not well tested yet (read: it works for me)."
4164 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4165 rework some of the code of TK/09 above to avoid the hardwired use of
4166 "%lld" and "long long". Replaced the call to snprintf() with a call to
4169 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4170 records point to non-existent hosts", "retry timeout exceeded", and
4171 "retry time not reached for any host after a long failure period".
4173 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4174 experimental DomainKeys support:
4176 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4177 (2) On an error such as an illegally used "control", the wrong name for
4178 the control was given.
4180 These problems did NOT occur unless DomainKeys support was compiled.
4182 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4184 PH/24 Added ${if match_ip condition.
4186 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4187 databases so that it will be absolutely obvious if a crash occurs in the
4188 DB library. This is a regular occurrence (often caused by mis-matched
4191 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4192 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4193 picked up on a user's system that detects such things. There doesn't seem
4194 to be a gcc warning option for this - only an attribute that has to be
4195 put on the function's prototype. It seems that in Fedora Core 4 they have
4196 set this on a number of new functions. No doubt there will be more in due
4199 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4200 specify a transport. However, if an address that was verified by such a
4201 router was the subject of a callout, Exim crashed because it tried to
4202 read the rcpt_include_affixes from the non-existent transport. Now it
4203 just assumes that the setting of that option is false. This bug was
4204 introduced by 4.51/PH/31.
4206 PH/28 Changed -d+all to exclude +memory, because that information is very
4207 rarely of interest, but it makes the output a lot bigger. People tend to
4208 do -d+all out of habit.
4210 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4211 code in os-type was giving problems when libc.so lives in lib64, like on
4214 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4215 aren't the modern standard, and it seems that some systems' include files
4216 don't always have them. Exim was already checking for some of the newer
4217 ones like T_AAAA, and defining it itself. I've added checks for all the
4218 record types that Exim uses.
4220 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4221 not automatically generating a new one, as it is supposed to. This
4222 prevented TLS from working. If the file did exist, but contained invalid
4223 data, a new version was generated, as expected. It was only the case of a
4224 non-existent file that was broken.
4226 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4227 with a change in libdomainkeys > 0.64.
4229 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4230 from DNS. If the selector record carries the flag, it now has
4231 precedence over the domain-wide flag.
4233 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4235 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4236 the use of a port name that isn't defined in /etc/services) occurred, the
4237 message was deferred as in a normal delivery, and thus remained on the
4238 spool, instead of being failed because of the mua_wrapper setting. This
4239 is now fixed, and I tidied up some of the mua_wrapper messages at the
4242 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4243 the messages in a hash of arrays rather than using individual hashes.
4244 This is a bit cleaner and results in dramatic memory savings, albeit
4245 at a slight CPU cost.
4247 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4248 as requested by Marc Sherman.
4250 SC/10 Eximstats: added histograms for user specified patterns as requested
4253 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4255 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4256 fopen() in the content-scanning modules that did not already have it.
4262 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4263 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4265 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4267 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4269 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4270 to test Sieve filters that use "vacation".
4272 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4273 that changes the way the GnuTLS parameters are stored in the cache file.
4274 The new format can be generated externally. For backward compatibility,
4275 if the data in the cache doesn't make sense, Exim assumes it has read an
4276 old-format file, and it generates new data and writes a new file. This
4277 means that you can't go back to an older release without removing the
4280 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4281 work if there are any delivery delays because "one_time" forces the
4282 parent to be marked "delivered", so its unseen clone is never tried
4283 again. For this reason, Exim now forbids the simultaneous setting of
4286 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4287 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4288 entry if you want to know the details. The fix, however, neglected to
4289 consider the case when local delivery batching is involved. The test for
4290 "previously delivered" was not happening when checking to see if an
4291 address could be batched with a previous (undelivered) one; under
4292 certain circumstances this could lead to multiple deliveries to the same
4295 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4296 in its include files, and this causes problems building Exim.
4298 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4299 header_syntax) but Exim was just ignoring anything given after a slash.
4300 In particular, this caused confusion with an attempt to use "verify =
4301 reverse_host_lookup/defer_ok". An error is now given when options are
4302 supplied for verify items that do not have them. (Maybe reverse_host_
4303 lookup should have a defer_ok option, but that's a different point.)
4305 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4306 defined by RFC 821) to 2048, because there were problems with some AUTH
4307 commands, and RFC 1869 says the size should be increased for extended
4308 SMTP commands that take arguments.
4310 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4313 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4314 "unknown" error; now it says that the functionality isn't in the binary.
4316 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4317 an address' error message when a string expansion fails (syntax or
4318 whatever). Otherwise the password may appear in the log. Following change
4319 PH/42 below, there is no longer a chance of it appearing in a bounce
4322 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4324 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4325 of its addresses was ever tried. (Bugzilla bug #2.)
4327 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4328 the result incorrectly in the debug output. (It correctly added a newline
4329 to what was transported.)
4331 TF/01 Added $received_time.
4333 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4334 commented out examples of how to interface to a virus scanner and to
4335 SpamAssassin. Also added commented examples of av_scanner and
4336 spamd_address settings.
4338 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4339 and controls are allowed in which ACLs. There were a couple of minor
4340 errors. Some of the entries in the conditions table (which is a table of
4341 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4342 negation of where the condition IS allowed.
4344 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4346 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4347 header file does not have a version number, so I've had to invent a new
4348 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4349 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4350 radiusclient), but it was contributed by a Radius user.
4352 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4353 files or directories, for OpenSSL.
4355 PH/19 When an Exim process that is running as root has to create an Exim log
4356 file, it does so in a subprocess that runs as exim:exim so as to get the
4357 ownership right at creation (otherwise, other Exim processes might see
4358 the file with the wrong ownership). There was no test for failure of this
4359 fork() call, which would lead to the process getting stuck as it waited
4360 for a non-existent subprocess. Forks do occasionally fail when resources
4361 run out. I reviewed all the other calls to fork(); they all seem to check
4364 PH/20 When checking for unexpected SMTP input at connect time (before writing
4365 the banner), Exim was not dealing correctly with a non-positive return
4366 from the read() function. If the client had disconnected by this time,
4367 the result was a log entry for a synchronization error with an empty
4368 string after "input=" when read() returned zero. If read() returned -1
4369 (an event I could not check), uninitialized data bytes were printed.
4370 There were reports of junk text (parts of files, etc) appearing after
4373 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4375 PH/22 Added support for macro redefinition, and (re)definition in between
4376 driver and ACL definitions.
4378 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4379 forgetting to use the resulting value; it was using the unexpanded value.
4381 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4382 hadn't been configured. The fix is from Juergen Kreileder, who
4383 understands it better than I do:
4385 "Here's what I see happening with three configured cyrus_sasl
4386 authenticators configured (plain, login, cram-md5):
4388 On startup auth_cyrus_sasl_init() gets called for each of these.
4389 This means three calls to sasl_listmech() without a specified mech_list.
4390 => SASL tests which mechs of all available mechs actually work
4391 => three warnings about OTP not working
4392 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4394 With the patch, sasl_listmech() also gets called three times. But now
4395 SASL's mech_list option is set to the server_mech specified in the the
4396 authenticator. Or in other words, the answer from sasl_listmech()
4397 gets limited to just the mech you're testing for (which is different
4399 => the return list contains just 'plain' or 'login', 'cram-md5' or
4400 nothing depending on the value of ob->server_mech.
4402 I've just tested the patch: Authentication still works fine,
4403 unavailable mechs specified in the exim configuration are still
4404 caught, and the auth.log warnings about OTP are gone."
4406 PH/25 When debugging is enabled, the contents of the command line are added
4407 to the debugging output, even when log_selector=+arguments is not
4410 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4411 answer is "GNU", and only if the return is "GNU/something" is the answer
4414 PH/27 $acl_verify_message is now set immediately after the failure of a
4415 verification in an ACL, and so is available in subsequent modifiers. In
4416 particular, the message can be preserved by coding like this:
4418 warn !verify = sender
4419 set acl_m0 = $acl_verify_message
4421 Previously, $acl_verify_message was set only while expanding "message"
4422 and "log_message" when a very denied access.
4424 PH/28 Modified OS/os.c-Linux with
4426 -#ifndef OS_LOAD_AVERAGE
4427 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4429 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4430 nomenclature these days.)
4432 PH/29 Installed patch from the Sieve maintainer that adds the options
4433 sieve_useraddress and sieve_subaddress to the redirect router.
4435 PH/30 In these circumstances:
4436 . Two addresses routed to the same list of hosts;
4437 . First host does not offer TLS;
4438 . First host accepts first address;
4439 . First host gives temporary error to second address;
4440 . Second host offers TLS and a TLS session is established;
4441 . Second host accepts second address.
4442 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4443 and peerdn, if requested) that were in fact used only for the second
4446 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4447 attention to any local part prefix or suffix that was matched by the
4448 router that accepted the address. It now behaves in the same way as it
4449 does for delivery: the affixes are removed from the local part unless
4450 rcpt_include_affixes is set on the transport.
4452 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4453 timeout during the DATA phase of an incoming message.
4455 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4456 applied a patch sanctioned by the Sieve maintainer.
4458 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4459 the uid or gid is negative. A case of a negative gid caused this to be
4460 noticed. The fix allows for either to be negative.
4462 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4463 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4466 PH/36 The change PH/12 above was broken. Fixed it.
4468 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4469 the grounds that routing the same address twice would always produce the
4470 same answer. This might have been true once, but it is certainly no
4471 longer true now. Routing a child address may depend on the previous
4472 routing that produced that child. Some complicated redirection strategies
4473 went wrong when messages had multiple recipients, and made Exim's
4474 behaviour dependent on the order in which the addresses were given.
4476 I have moved the duplicate checking until after the routing is complete.
4477 Exim scans the addresses that are assigned to local and remote
4478 transports, and removes any duplicates. This means that more work will be
4479 done, as duplicates will always all be routed, but duplicates are
4480 presumably rare, so I don't expect this is of any significance.
4482 For deliveries to pipes, files, and autoreplies, the duplicate checking
4483 still happens during the routing process, since they are not going to be
4486 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4487 It corrects a timeout issue with spamd. This is Ian's comment: "The
4488 background is that sometimes spamd either never reads data from a
4489 connection it has accepted, or it never writes response data. The exiscan
4490 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4491 blindly assumes that writes won't block so it may never time out."
4493 PH/39 Allow G after quota size as well as K and M.
4495 PH/40 The value set for $authenticated_id in an authenticator may not contain
4496 binary zeroes or newlines because the value is written to log lines and
4497 to spool files. There was no check on this. Now the value is run through
4498 the string_printing() function so that such characters are converted to
4499 printable escape sequences.
4501 PH/41 $message_linecount is a new variable that contains the total number of
4502 lines in the message. Compare $body_linecount, which is the count for the
4505 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4506 bounce and delay warning messages, except in certain special cases, which
4509 (a) An SMTP error message from a remote host;
4510 (b) A message specified in a :fail: redirection;
4511 (c) A message specified in a "fail" command in a system filter;
4512 (d) A message specified in a FAIL return from the queryprogram router;
4513 (e) A message specified by the cannot_route_message router option.
4515 In these cases only, Exim does include the error details in bounce and
4516 warning messages. There are also a few cases where bland messages such
4517 as "unrouteable address" or "local delivery error" are given.
4519 PH/43 $value is now also set for the "else" part of a ${run expansion.
4521 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4522 being worked on, but at least Exim now implements the latest version to
4525 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4526 process to complete was treated as a delivery failure, a timeout while
4527 writing the message to the pipe was logged, but erroneously treated as a
4528 successful delivery. Such timeouts include transport filter timeouts. For
4529 consistency with the overall process timeout, these timeouts are now
4530 treated as errors, giving rise to delivery failures by default. However,
4531 there is now a new Boolean option for the pipe transport called
4532 timeout_defer, which, if set TRUE, converts the failures into defers for
4533 both kinds of timeout. A transport filter timeout is now identified in
4536 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4537 systems where "make" and "gmake" are different, calling "gmake" at top
4538 level broke things. I've arranged for the value of $(MAKE) to be passed
4539 from the Makefile to this script so that it can call the same version of
4543 A note about Exim versions 4.44 and 4.50
4544 ----------------------------------------
4546 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4547 changes of various kinds. As a consequence, a big documentation update was
4548 needed. This delayed the release for rather longer than seemed good, especially
4549 in the light of a couple of (minor) security issues. Therefore, the changes
4550 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4551 release. So 4.44 and 4.50 are in effect two different branches that both start
4554 I have left the 4.50 change log unchanged; it contains all the changes since
4555 4.43. The change log for 4.44 is below; many of its items are identical to
4556 those for 4.50. This seems to be the most sensible way to preserve the
4557 historical information.
4563 1. Minor wording change to the doc/README.SIEVE file.
4565 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4566 computation of the current number of files was incorrect.
4568 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4569 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4572 4. Give more explanation in the error message when the command for a transport
4573 filter fails to execute.
4575 5. There are several places where Exim runs a non-Exim command in a
4576 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4577 was being done only for the command run by the queryprogram router. It is
4578 now done for all such subprocesses. The other cases are: ${run, transport
4579 filters, and the commands run by the lmtp and pipe transports.
4581 6. Added CONFIGURE_GROUP build-time option.
4583 7. Some older OS have a limit of 256 on the maximum number of file
4584 descriptors. Exim was using setrlimit() to set 1000 as a large value
4585 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4586 systems. I've change it so that if it can't get 1000, it tries for 256.
4588 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4589 was an oversight, and furthermore, ever since the addition of extra
4590 controls (e.g. 4.43/32), the checks on when to allow different forms of
4591 "control" were broken. There should now be diagnostics for all cases when a
4592 control that does not make sense is encountered.
4594 9. Added the /retain_sender option to "control=submission".
4596 10. $recipients is now available in the predata ACL (oversight).
4598 11. Tidy the search cache before the fork to do a delivery from a message
4599 received from the command line. Otherwise the child will trigger a lookup
4600 failure and thereby defer the delivery if it tries to use (for example) a
4601 cached ldap connection that the parent has called unbind on.
4603 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4604 of $address_data from the recipient verification was clobbered by the
4605 sender verification.
4607 13. The value of address_data from a sender verification is now available in
4608 $sender_address_data in subsequent conditions in the ACL statement.
4610 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4612 15. Added a new option "connect=<time>" to callout options, to set a different
4615 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4616 was its contents. (It was OK if the option was not defined at all.)
4618 17. A "Completed" log line is now written for messages that are removed from
4619 the spool by the -Mrm option.
4621 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4622 information about exactly what failed.
4624 19. Added -dd to debug only the daemon process.
4626 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4627 handles timeouts, both on the server side and network timeouts. Renamed the
4628 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4630 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4631 It is now set to "smtps".
4633 22. $host_address is now set to the target address during the checking of
4634 ignore_target_hosts.
4636 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4637 being passed; this would have caused $sender_host_name to have been used if
4638 matching the list had actually called for a host name (not very likely,
4639 since this list is usually IP addresses). A host name is now passed as
4642 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4643 code that specifies a non-restarting handler (typically sigaction() in
4644 modern systems) in an attempt to fix a rare and obscure crash bug.
4646 25. Narrowed the window for a race in the daemon that could cause it to ignore
4647 SIGCHLD signals. This is not a major problem, because they are used only to
4648 wake it up if nothing else does.
4650 26. A malformed maildirsize file could cause Exim to calculate negative values
4651 for the mailbox size or file count. Odd effects could occur as a result.
4652 The maildirsize information is now recalculated if the size or filecount
4655 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4656 support for a long time. Removed HAVE_SYS_VFS_H.
4658 28. Installed the latest version of exipick from John Jetmore.
4660 29. In an address list, if the pattern was not a regular expression, an empty
4661 subject address (from a bounce message) matched only if the pattern was an
4662 empty string. Non-empty patterns were not even tested. This was the wrong
4663 because it is perfectly reasonable to use an empty address as part of a
4664 database query. An empty address is now tested by patterns that are
4665 lookups. However, all the other forms of pattern expect the subject to
4666 contain a local part and a domain, and therefore, for them, an empty
4667 address still always fails if the pattern is not itself empty.
4669 30. Exim went into a mad DNS loop when attempting to do a callout where the
4670 host was specified on an smtp transport, and looking it up yielded more
4671 than one IP address.
4673 31. Re-factored the code for checking spool and log partition space into a
4674 function that finds that data and another that does the check. The former
4675 is then used to implement four new variables: $spool_space, $log_space,
4676 $spool_inodes, and $log_inodes.
4678 32. The RFC2047 encoding function was originally intended for short strings
4679 such as real names; it was not keeping to the 75-character limit for
4680 encoded words that the RFC imposes. It now respects the limit, and
4681 generates multiple encoded words if necessary. To be on the safe side, I
4682 have increased the buffer size for the ${rfc2047: expansion operator from
4685 33. It is now permitted to omit both strings after an "if" condition; if the
4686 condition is true, the result is "true". As before, when the second string
4687 is omitted, a false condition yields an empty string. This makes it less
4688 cumbersome to write custom ACL and router conditions.
4690 34. Failure to deliver a bounce message always caused it to be frozen, even if
4691 there was an errors_to setting on the router. The errors_to setting is now
4694 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4695 canonical form (fully expanded) before being placed in
4696 $sender_host_address.
4698 36. The table in the code that translates DNS record types into text (T_A to
4699 "A" for instance) was missing entries for NS and CNAME. It is just possible
4700 that this could have caused confusion if both these types were looked up
4701 for the same domain, because the text type is used as part of Exim's
4702 per-process caching. But the chance of anyone hitting this buglet seems
4705 37. The dnsdb lookup has been extended in a number of ways.
4707 (1) There is a new type, "zns", which walks up the domain tree until it
4708 finds some nameserver records. It should be used with care.
4710 (2) There is a new type, "mxh", which is like "mx" except that it returns
4711 just the host names, not the priorities.
4713 (3) It is now possible to give a list of domains (or IP addresses) to be
4714 looked up. The behaviour when one of the lookups defers can be
4715 controlled by a keyword.
4717 (4) It is now possible to specify the separator character for use when
4718 multiple records are returned.
4720 38. The dnslists ACL condition has been extended: it is now possible to supply
4721 a list of IP addresses and/or domains to be looked up in a particular DNS
4724 39. Added log_selector=+queue_time_overall.
4726 40. When running the queue in the test harness, wait just a tad after forking a
4727 delivery process, to get repeatability of debugging output.
4729 41. Include certificate and key file names in error message when GnuTLS fails
4730 to set them up, because the GnuTLS error message doesn't include the name
4731 of the failing file when there is a problem reading it.
4733 42. Allow both -bf and -bF in the same test run.
4735 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4737 44. The "Exiscan patch" is now merged into the mainline Exim source.
4739 45. Sometimes the final signoff response after QUIT could fail to get
4740 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4741 < 0 before doing a fflush(). This bug looks as though it goes back to the
4742 introduction of TLS in release 3.20, but "sometimes" must have been rare
4743 because the tests only now provoked it.
4745 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4746 (this can affect the format of dates).
4748 47. exim_tidydb, when checking for the continued existence of a message for
4749 which it has found a message-specific retry record, was not finding
4750 messages that were in split spool directories. Consequently, it was
4751 deleting retry records that should have stayed in existence.
4753 48. Steve fixed some bugs in eximstats.
4755 49. The SPA authentication driver was not abandoning authentication and moving
4756 on to the next authenticator when an expansion was forced to fail,
4757 contradicting the general specification for all authenticators. Instead it
4758 was generating a temporary error. It now behaves as specified.
4760 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4761 (the order specifies the preference for clients). The order is now AES256,
4762 AES128, 3DES, ARCFOUR128.
4764 51. Small patch to Sieve code - explicitly set From: when generating an
4767 52. Exim crashed if a remote delivery caused a very long error message to be
4768 recorded - for instance if somebody sent an entire SpamAssassin report back
4769 as a large number of 550 error lines. This bug was coincidentally fixed by
4770 increasing the size of one of Exim's internal buffers (big_buffer) that
4771 happened as part of the Exiscan merge. However, to be on the safe side, I
4772 have made the code more robust (and fixed the comments that describe what
4775 53. Now that there can be additional text after "Completed" in log lines (if
4776 the queue_time_overall log selector is set), a one-byte patch to exigrep
4777 was needed to allow it to recognize "Completed" as not the last thing in
4780 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4781 patch that reportedly fixes this has been added. I am not expert enough to
4782 create a test for it. This is what the patch creator wrote:
4784 "I found a little strange behaviour of ldap code when working with
4785 Windows 2003 AD Domain, where users was placed in more than one
4786 Organization Units. When I tried to give exim partial DN, the exit code
4787 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4788 But simultaneously result of request was absolutely normal ldap result,
4789 so I produce this patch..."
4791 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4792 REFERENCE, so I have modified the code to exclude the patch when that macro
4795 55. Some experimental protocols are using DNS PTR records for new purposes. The
4796 keys for these records are domain names, not reversed IP addresses. The
4797 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4798 leaves it alone. Component reversal etc. now happens only for IP addresses.
4801 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4803 57. Double the size of the debug message buffer (to 2048) so that more of very
4804 long debug lines gets shown.
4806 58. The exicyclog utility now does better if the number of log files to keep
4807 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4809 59. Two changes related to the smtp_active_hostname option:
4811 (1) $smtp_active_hostname is now available as a variable.
4812 (2) The default for smtp_banner uses $smtp_active_hostname instead
4813 of $primary_hostname.
4815 60. The host_aton() function is supposed to be passed a string that is known
4816 to be a valid IP address. However, in the case of IPv6 addresses, it was
4817 not checking this. This is a hostage to fortune. Exim now panics and dies
4818 if the condition is not met. A case was found where this could be provoked
4819 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4820 components; fortuitously, this particular loophole had already been fixed
4821 by change 4.50/55 above.
4823 If there are any other similar loopholes, the new check in host_aton()
4824 itself should stop them being exploited. The report I received stated that
4825 data on the command line could provoke the exploit when Exim was running as
4826 exim, but did not say which command line option was involved. All I could
4827 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4828 running as the user.
4831 61. There was a buffer overflow vulnerability in the SPA authentication code
4832 (which came originally from the Samba project). I have added a test to the
4833 spa_base64_to_bits() function which I hope fixes it.
4836 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4837 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4839 63. The daemon start-up calls getloadavg() while still root for those OS that
4840 need the first call to be done as root, but it missed one case: when
4841 deliver_queue_load_max is set with deliver_drop_privilege. This is
4842 necessary for the benefit of the queue runner, because there is no re-exec
4843 when deliver_drop_privilege is set.
4845 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4846 This has been fixed.
4848 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4849 was in use, was not putting the data itself into the right store pool;
4850 consequently, it could be overwritten for a subsequent message in the same
4851 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4854 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4856 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4857 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4858 the function were treating the return as a boolean value, which happened to
4859 work because 0=false and not-0=true, but is not correct code.
4861 68. The host_aton() function was not handling scoped IPv6 addresses (those
4862 with, for example, "%eth0" on the end) correctly.
4864 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4865 negated items (that is, ~something) in unsigned ints. Some compilers
4866 apparently mutter when there is no cast.
4868 70. If an address verification called from an ACL failed, and did not produce a
4869 user-specific message (i.e. there was only a "system" message), nothing was
4870 put in $acl_verify_message. In this situation, it now puts the system
4873 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4874 session; change 4.31/43 added the unwanted input to the log line - except
4875 that it did not do this in the start of session case. It now does.
4877 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4878 This is wrong and can cause the other end to generate a synchronization
4879 error if it is another Exim or anything else that does the synchronization
4880 check. A QUIT command is no longer sent after a timeout.
4882 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4883 during host lookups.
4885 74. The defer_ok option of callout verification was not working if it was used
4886 when verifying addresses in header lines, that is, for this case:
4888 verify = header_sender/callout=defer_ok
4890 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4891 those file descriptors could be used for SMTP connections. If anything
4892 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4893 could be sent to the SMTP client, causing chaos. The daemon now opens
4894 stdin, stdout, and stderr to /dev/null when it puts itself into the
4897 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4898 log by default. The user can override this with suitable Perl magic.
4900 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4901 the log message when discard triggers, was not working for the DATA ACL or
4902 for the non-SMTP ACL.
4904 78. Error message wording change in sieve.c.
4906 79. If smtp_accept_max_per_host was set, the number of connections could be
4907 restricted to fewer than expected, because the daemon was trying to set up
4908 a new connection before checking whether the processes handling previous
4909 connections had finished. The check for completed processes is now done
4910 earlier. On busy systems, this bug wouldn't be noticed because something
4911 else would have woken the daemon, and it would have reaped the completed
4914 80. If a message was submitted locally by a user whose login name contained one
4915 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4916 It caused a spool format error. I have fixed the spool reading code. A
4917 related problem was that the "from" clause in the Received: line became
4918 illegal because of the space(s). It is now covered by ${quote_local_part.
4920 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4923 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4925 83. Patch from Sieve maintainer to fix unterminated string problem in
4926 "vacation" handling.
4928 84. Some minor changes to the Linux configuration files to help with other
4929 OS variants using glibc.
4931 85. One more patch for Sieve to update vacation handling to latest spec.
4934 ----------------------------------------------------
4935 See the note above about the 4.44 and 4.50 releases.
4936 ----------------------------------------------------
4942 1. Change 4.43/35 introduced a bug that caused file counts to be
4943 incorrectly computed when quota_filecount was set in an appendfile
4946 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4947 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4950 3. Give more explanation in the error message when the command for a transport
4951 filter fails to execute.
4953 4. There are several places where Exim runs a non-Exim command in a
4954 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4955 was being done only for the command run by the queryprogram router. It is
4956 now done for all such subprocesses. The other cases are: ${run, transport
4957 filters, and the commands run by the lmtp and pipe transports.
4959 5. Some older OS have a limit of 256 on the maximum number of file
4960 descriptors. Exim was using setrlimit() to set 1000 as a large value
4961 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4962 systems. I've change it so that if it can't get 1000, it tries for 256.
4964 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4965 was an oversight, and furthermore, ever since the addition of extra
4966 controls (e.g. 4.43/32), the checks on when to allow different forms of
4967 "control" were broken. There should now be diagnostics for all cases when a
4968 control that does not make sense is encountered.
4970 7. $recipients is now available in the predata ACL (oversight).
4972 8. Tidy the search cache before the fork to do a delivery from a message
4973 received from the command line. Otherwise the child will trigger a lookup
4974 failure and thereby defer the delivery if it tries to use (for example) a
4975 cached ldap connection that the parent has called unbind on.
4977 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4978 of $address_data from the recipient verification was clobbered by the
4979 sender verification.
4981 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4982 was its contents. (It was OK if the option was not defined at all.)
4984 11. A "Completed" log line is now written for messages that are removed from
4985 the spool by the -Mrm option.
4987 12. $host_address is now set to the target address during the checking of
4988 ignore_target_hosts.
4990 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4991 being passed; this would have caused $sender_host_name to have been used if
4992 matching the list had actually called for a host name (not very likely,
4993 since this list is usually IP addresses). A host name is now passed as
4996 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4997 code that specifies a non-restarting handler (typically sigaction() in
4998 modern systems) in an attempt to fix a rare and obscure crash bug.
5000 15. Narrowed the window for a race in the daemon that could cause it to ignore
5001 SIGCHLD signals. This is not a major problem, because they are used only to
5002 wake it up if nothing else does.
5004 16. A malformed maildirsize file could cause Exim to calculate negative values
5005 for the mailbox size or file count. Odd effects could occur as a result.
5006 The maildirsize information is now recalculated if the size or filecount
5009 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5010 support for a long time. Removed HAVE_SYS_VFS_H.
5012 18. Updated exipick to current release from John Jetmore.
5014 19. Allow an empty sender to be matched against a lookup in an address list.
5015 Previously the only cases considered were a regular expression, or an
5018 20. Exim went into a mad DNS lookup loop when doing a callout where the
5019 host was specified on the transport, if the DNS lookup yielded more than
5022 21. The RFC2047 encoding function was originally intended for short strings
5023 such as real names; it was not keeping to the 75-character limit for
5024 encoded words that the RFC imposes. It now respects the limit, and
5025 generates multiple encoded words if necessary. To be on the safe side, I
5026 have increased the buffer size for the ${rfc2047: expansion operator from
5029 22. Failure to deliver a bounce message always caused it to be frozen, even if
5030 there was an errors_to setting on the router. The errors_to setting is now
5033 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5034 canonical form (fully expanded) before being placed in
5035 $sender_host_address.
5037 24. Updated eximstats to version 1.33
5039 25. Include certificate and key file names in error message when GnuTLS fails
5040 to set them up, because the GnuTLS error message doesn't include the name
5041 of the failing file when there is a problem reading it.
5043 26. Expand error message when OpenSSL has problems setting up cert/key files.
5046 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5047 (this can affect the format of dates).
5049 28. exim_tidydb, when checking for the continued existence of a message for
5050 which it has found a message-specific retry record, was not finding
5051 messages that were in split spool directories. Consequently, it was
5052 deleting retry records that should have stayed in existence.
5054 29. eximstats updated to version 1.35
5055 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5056 1.35 - bugfix such that pie charts by volume are generated correctly
5058 30. The SPA authentication driver was not abandoning authentication and moving
5059 on to the next authenticator when an expansion was forced to fail,
5060 contradicting the general specification for all authenticators. Instead it
5061 was generating a temporary error. It now behaves as specified.
5063 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5064 (the order specifies the preference for clients). The order is now AES256,
5065 AES128, 3DES, ARCFOUR128.
5067 31. Small patch to Sieve code - explicitly set From: when generating an
5070 32. Exim crashed if a remote delivery caused a very long error message to be
5071 recorded - for instance if somebody sent an entire SpamAssassin report back
5072 as a large number of 550 error lines. This bug was coincidentally fixed by
5073 increasing the size of one of Exim's internal buffers (big_buffer) that
5074 happened as part of the Exiscan merge. However, to be on the safe side, I
5075 have made the code more robust (and fixed the comments that describe what
5078 33. Some experimental protocols are using DNS PTR records for new purposes. The
5079 keys for these records are domain names, not reversed IP addresses. The
5080 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5081 leaves it alone. Component reversal etc. now happens only for IP addresses.
5084 34. The host_aton() function is supposed to be passed a string that is known
5085 to be a valid IP address. However, in the case of IPv6 addresses, it was
5086 not checking this. This is a hostage to fortune. Exim now panics and dies
5087 if the condition is not met. A case was found where this could be provoked
5088 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5089 components; fortuitously, this particular loophole had already been fixed
5090 by change 4.50/55 or 4.44/33 above.
5092 If there are any other similar loopholes, the new check in host_aton()
5093 itself should stop them being exploited. The report I received stated that
5094 data on the command line could provoke the exploit when Exim was running as
5095 exim, but did not say which command line option was involved. All I could
5096 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5097 running as the user.
5100 35. There was a buffer overflow vulnerability in the SPA authentication code
5101 (which came originally from the Samba project). I have added a test to the
5102 spa_base64_to_bits() function which I hope fixes it.
5105 36. The daemon start-up calls getloadavg() while still root for those OS that
5106 need the first call to be done as root, but it missed one case: when
5107 deliver_queue_load_max is set with deliver_drop_privilege. This is
5108 necessary for the benefit of the queue runner, because there is no re-exec
5109 when deliver_drop_privilege is set.
5111 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5112 was in use, was not putting the data itself into the right store pool;
5113 consequently, it could be overwritten for a subsequent message in the same
5114 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5117 38. Sometimes the final signoff response after QUIT could fail to get
5118 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5119 < 0 before doing a fflush(). This bug looks as though it goes back to the
5120 introduction of TLS in release 3.20, but "sometimes" must have been rare
5121 because the tests only now provoked it.
5127 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5128 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5129 Now it returns an integer. A number of calls were still expecting a T/F
5130 return. Fortuitously, in all cases, the tests worked in OK situations,
5131 which is the norm. However, things would have gone wrong on any write
5132 failures on the smtp file descriptor. This function is used when sending
5133 messages over SMTP and also when doing verify callouts.
5135 2. When Exim is called to do synchronous delivery of a locally submitted
5136 message (the -odf or -odi options), it no longer closes stderr before doing
5139 3. Implemented the mua_wrapper option.
5141 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5143 5. Implemented the functions header_remove(), header_testname(),
5144 header_add_at_position(), and receive_remove_recipient(), and exported them
5147 6. If an ACL "warn" statement specified the addition of headers, Exim already
5148 inserted X-ACL-Warn: at the start if there was no header name. However, it
5149 was not making this test for the second and subsequent header lines if
5150 there were newlines in the string. This meant that an invalid header could
5151 be inserted if Exim was badly configured.
5153 7. Allow an ACL "warn" statement to add header lines at the start or after all
5154 the Received: headers, as well as at the end.
5156 8. Added the rcpt_4xx retry error code.
5158 9. Added postmaster_mailfrom=xxx to callout verification option.
5160 10. Added mailfrom=xxxx to the callout verification option, for verify=
5163 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5164 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5166 12. Inserted some casts to stop certain compilers warning when using pointer
5167 differences as field lengths or precisions in printf-type calls (mostly
5168 affecting debugging statements).
5170 13. Added optional readline() support for -be (dynamically loaded).
5172 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5173 same clock tick as a message's arrival, so that its received time was the
5174 same as the "first fail" time on the retry record, and that message
5175 remained on the queue past the ultimate address timeout, every queue runner
5176 would try a delivery (because it was past the ultimate address timeout) but
5177 after another failure, the ultimate address timeout, which should have then
5178 bounced the address, did not kick in. This was a "< instead of <=" error;
5179 in most cases the first failure would have been in the next clock tick
5180 after the received time, and all would be well.
5182 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5183 being recognized when the domain list was tested by the match_domain
5184 condition in an expansion string.
5186 16. Added the ${str2b64: operator.
5188 17. Exim was always calling setrlimit() to set a large limit for the number of
5189 processes, without checking whether the existing limit was already
5190 adequate. (It did check for the limit on file descriptors.) Furthermore,
5191 errors from getrlimit() and setrlimit() were being ignored. Now they are
5192 logged to the main and panic logs, but Exim does carry on, to try to do its
5193 job under whatever limits there are.
5195 18. Imported PCRE 5.0.
5197 19. Trivial typo in log message " temporarily refused connection" (the leading
5200 20. If the log selector return_path_on_delivery was set and an address was
5201 redirected to /dev/null, the delivery process crashed because it assumed
5202 that a return path would always be set for a "successful" delivery. In this
5203 case, the whole delivery is bypassed as an optimization, and therefore no
5206 21. Internal re-arrangement: the function for sending a challenge and reading
5207 a response while authentication was assuming a zero-terminated challenge
5208 string. It's now changed to take a pointer and a length, to allow for
5209 binary data in such strings.
5211 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5213 23. Exim was not respecting finduser_retries when seeking the login of the
5214 uid under which it was called; it was always trying 10 times. (The default
5215 setting of finduser_retries is zero.) Also, it was sleeping after the final
5216 failure, which is pointless.
5218 24. Implemented tls_on_connect_ports.
5220 25. Implemented acl_smtp_predata.
5222 26. If the domain in control=submission is set empty, Exim assumes that the
5223 authenticated id is a complete email address when it generates From: or
5224 Sender: header lines.
5226 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5227 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5228 chown and chgrp in /bin and hostname in /usr/bin.
5230 28. Exim was keeping the "process log" file open after each use, just as it
5231 does for the main log. This opens the possibility of it remaining open for
5232 long periods when the USR1 signal hits a daemon. Occasional processlog
5233 errors were reported, that could have been caused by this. Anyway, it seems
5234 much more sensible not to leave this file open at all, so that is what now
5237 29. The long-running daemon process does not normally write to the log once it
5238 has entered its main loop, and it closes the log before doing so. This is
5239 so that log files can straightforwardly be renamed and moved. However,
5240 there are a couple of unusual error situations where the daemon does write
5241 log entries, and I had neglected to close the log afterwards.
5243 30. The text of an SMTP error response that was received during a remote
5244 delivery was being truncated at 512 bytes. This is too short for some of
5245 the long messages that one sometimes sees. I've increased the limit to
5248 31. It is now possible to make retry rules that apply only when a message has a
5249 specific sender, in particular, an empty sender.
5251 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5252 it possible to be selective about when SMTP synchronization is enforced.
5254 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5256 32. Implemented hosts_connection_nolog.
5258 33. Added an ACL for QUIT.
5260 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5263 35. Added mailbox_size and mailbox_filecount to appendfile.
5265 36. Added control = no_multiline_responses to ACLs.
5267 37. There was a bug in the logic of the code that waits for the clock to tick
5268 in the case where the clock went backwards by a substantial amount such
5269 that the microsecond fraction of "now" was more than the microsecond
5270 fraction of "then" (but the whole seconds number was less).
5272 38. Added support for the libradius Radius client library this is found on
5273 FreeBSD (previously only the radiusclient library was supported).
5279 1. When certain lookups returned multiple values in the form name=value, the
5280 quoting of the values was not always being done properly. Specifically:
5281 (a) If the value started with a double quote, but contained no whitespace,
5283 (b) If the value contained whitespace other than a space character (i.e.
5284 tabs or newlines or carriage returns) it was not quoted.
5285 This fix has been applied to the mysql and pgsql lookups by writing a
5286 separate quoting function and calling it from the lookup code. The fix
5287 should probably also be applied to nisplus, ibase and oracle lookups, but
5288 since I cannot test any of those, I have not disturbed their existing code.
5290 2. A hit in the callout cache for a specific address caused a log line with no
5291 reason for rejecting RCPT. Now it says "Previous (cached) callout
5292 verification failure".
5294 3. There was an off-by-one bug in the queryprogram router. An over-long
5295 return line was truncated at 256 instead of 255 characters, thereby
5296 overflowing its buffer with the terminating zero. As well as fixing this, I
5297 have increased the buffer size to 1024 (and made a note to document this).
5299 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5300 when Exim is waiting for an SMTP response from a remote server, Exim
5301 restarts its select() call on the socket, thereby resetting its timeout.
5302 This is not a problem when such interrupts are rare. Somebody set up a cron
5303 job to run exiwhat every 2 minutes, which is less than the normal select()
5304 timeout (5 or 10 minutes). This meant that the select() timeout never
5305 kicked in because it was always reset. I have fixed this by comparing the
5306 time when an interrupt arrives with the time at the start of the first call
5307 to select(). If more time than the timeout has elapsed, the interrupt is
5308 treated as a timeout.
5310 5. Some internal re-factoring in preparation for the addition of Sieve
5311 extensions (by MH). In particular, the "personal" test is moved to a
5312 separate function, and given an option for scanning Cc: and Bcc: (which is
5313 not set for Exim filters).
5315 6. When Exim created an email address using the login of the caller as the
5316 local part (e.g. when creating a From: or Sender: header line), it was not
5317 quoting the local part when it contained special characters such as @.
5319 7. Installed new OpenBSD configuration files.
5321 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5322 try to make them clearer.
5324 9. Callout options, other than the timeout value, were being ignored when
5325 verifying sender addresses in header lines. For example, when using
5327 verify = header_sender/callout=no_cache
5329 the cache was (incorrectly) being used.
5331 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5332 only the Cygwin environment.
5334 11. When return_path_on_delivery was set as a log selector, if different remote
5335 addresses in the same message used different return paths and parallel
5336 remote delivery occurred, the wrong values would sometimes be logged.
5337 (Whenever a remote delivery process finished, the return path value from
5338 the most recently started remote delivery process was logged.)
5340 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5341 lines when AUTH and/or TLS are in use. This is the "received protocol"
5342 field. Exim used to use "asmtp" for authenticated SMTP, without any
5343 indication (in the protocol name) for TLS use. Now it follows the RFC and
5344 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5345 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5346 names appear in log lines as well as in Received: header lines.
5348 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5349 extensions, and comparison tests, and to fix some bugs.
5351 14. Changes to the "personal" filter test:
5353 (1) The test was buggy in that it was just doing the equivalent of
5354 "contains" tests on header lines. For example, if a user's address was
5355 anne@some.where, the "personal" test would incorrectly be true for
5357 To: susanne@some.where
5359 This test is now done by extracting each address from the header in turn,
5360 and checking the entire address. Other tests that are part of "personal"
5361 are now done using regular expressions (for example, to check local parts
5362 of addresses in From: header lines).
5364 (2) The list of non-personal local parts in From: addresses has been
5365 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5366 taken from the Sieve specification recommendations.
5368 (3) If the message contains any header line starting with "List-" it is
5369 treated as non-personal.
5371 (4) The test for "circular" in the Subject: header line has been removed
5372 because it now seems ill-conceived.
5374 15. Minor typos in src/EDITME comments corrected.
5376 16. Installed latest exipick from John Jetmore.
5378 17. If headers_add on a router specified a text string that was too long for
5379 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5380 of string_sprintf() is now avoided.
5382 18. $message_body_size was not set (it was always zero) when running the DATA
5383 ACL and the local_scan() function.
5385 19. For the "mail" command in an Exim filter, no default was being set for
5386 the once_repeat time, causing a random time value to be used if "once" was
5387 specified. (If the value happened to be <= 0, no repeat happened.) The
5388 default is now 0s, meaning "never repeat". The "vacation" command was OK
5389 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5390 (I found it when inspecting the code).
5392 20. There is now an overall timeout for performing a callout verification. It
5393 defaults to 4 times the callout timeout, which applies to individual SMTP
5394 commands during the callout. The overall timeout applies when there is more
5395 than one host that can be tried. The timeout is checked before trying the
5396 next host. This prevents very long delays if there are a large number of
5397 hosts and all are timing out (e.g. when the network connections are timing
5398 out). The value of the overall timeout can be changed by specifying an
5399 additional sub-option for "callout", called "maxwait". For example:
5401 verify = sender/callout=5s,maxwait=20s
5403 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5404 to the end before writing, but this should make it even safer).
5406 22. Exim was forgetting that it had advertised PIPELINING for the second and
5407 subsequent messages on an SMTP connection. It was also not resetting its
5408 memory on STARTTLS and an internal HELO.
5410 23. When Exim logs an SMTP synchronization error within a session, it now
5411 records whether PIPELINING has been advertised or not.
5413 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5414 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5415 rather than long int.
5417 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5419 26. Added the never_mail option to autoreply.
5425 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5426 crash if the getsockname() call failed; this can happen if a connection is
5427 closed very soon after it is established. The problem was simply in the
5428 order in which certain operations were done, causing Exim to try to write
5429 to the SMTP stream before it had set up the file descriptor. The bug has
5430 been fixed by making things happen in the correct order.
5436 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5437 before the connection was closed, thus losing the rejection response.
5439 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5440 some early Solaris releases, but causes trouble in current releases where
5441 socklen_t is defined.
5443 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5446 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5449 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5450 that releasing the top of it at the end releases what was used for sub-
5451 expansions (unless the block got too big). However, discard this block if
5452 the first thing is a variable or header, so that we can use its block when
5453 it is dynamic (useful for very large $message_headers, for example).
5455 6. Lookups now cache *every* query, not just the most recent. A new, separate
5456 store pool is used for this. It can be recovered when all lookup caches are
5457 flushed. Lookups now release memory at the end of their result strings.
5458 This has involved some general refactoring of the lookup sources.
5460 7. Some code has been added to the store_xxx() functions to reduce the amount
5461 of flapping under certain conditions.
5463 8. log_incoming_interface used to affect only the <= reception log lines. Now
5464 it causes the local interface and port to be added to several more SMTP log
5465 lines, for example "SMTP connection from", and rejection lines.
5467 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5469 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5471 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5472 could be overwritten at the end of the current message (or the start of a
5473 new message if it was set in a HELO ACL). The value is now preserved for
5474 the duration of the SMTP connection.
5476 12. If a transport had a headers_rewrite setting, and a matching header line
5477 contained an unqualified address, that address was qualified, even if it
5478 did not match any rewriting rules. The underlying bug was that the values
5479 of the flags that permit the existence of unqualified sender and recipient
5480 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5481 non-local messages, and by -bnq for local messages) were not being
5482 preserved with the message after it was received.
5484 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5485 "next input=" as part of the text comprising the host identity instead of
5486 the correct text. The code was using the same buffer for two different
5487 strings. However, depending on which order the printing function evaluated
5488 its arguments, the bug did not always show up. Under Linux, for example, my
5489 test suite worked just fine.
5491 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5492 doesn't work with some older versions of Perl. It has been changed to "my",
5493 which in any case is probably the better facility to use.
5495 15. A really picky compiler found some instances of statements for creating
5496 error messages that either had too many or two few arguments for the format
5499 16. The size of the buffer for calls to the DNS resolver has been increased
5500 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5501 for addresses that have a lot of PTR records. This alleviates a problem; it
5502 does not fully solve it.
5504 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5505 buffer now truncates the list and logs the incident, which is the same
5506 action as happens when Exim is looking up a host name and its aliases.
5507 Previously in this situation something unpredictable would happen;
5508 sometimes it was "internal error: store_reset failed".
5510 18. If a server dropped the connection unexpectedly when an Exim client was
5511 using GnuTLS and trying to read a response, the client delivery process
5512 crashed while trying to generate an error log message.
5514 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5515 string, for example:
5517 warn message = H1: something\nH2: something
5519 the text was added as a single header line from Exim's point of view
5520 though it ended up OK in the delivered message. However, searching for the
5521 second and subsequent header lines using $h_h2: did not work. This has been
5522 fixed. Similarly, if a system filter added multiple headers in this way,
5523 the routers could not see them.
5525 20. Expanded the error message when iplsearch is called with an invalid key to
5526 suggest using net-iplsearch in a host list.
5528 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5529 ACLs are no longer actually imposed (and a message to that effect is
5532 22. If a "gecos" field in a passwd entry contained escaped characters, in
5533 particular, if it contained a \" sequence, Exim got it wrong when building
5534 a From: or a Sender: header from that name. A second bug also caused
5535 incorrect handling when an unquoted " was present following a character
5536 that needed quoting.
5538 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5539 was not being matched caselessly.
5541 24. Arranged for all hyphens in the exim.8 source to be escaped with
5544 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5545 properly. Recipient callout cache records were still being keyed to include
5546 the sender, even when use_sender was set false. This led to far more
5547 callouts that were necessary. The sender is no longer included in the key
5548 when use_sender is false.
5550 26. Added "control = submission" modifier to ACLs.
5552 27. Added the ${base62d: operator to decode base 62 numbers.
5554 28. dnsdb lookups can now access SRV records.
5556 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5557 the configuration file.
5559 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5560 (-v) mode. This makes the output for a verbose queue run more intelligible.
5562 31. Added a use_postmaster feature to recipient callouts.
5564 32. Added the $body_zerocount variable, containing the number of binary zero
5565 bytes in the message body.
5567 33. The time of last modification of the "new" subdirectory is now used as the
5568 "mailbox time last read" when there is a quota error for a maildir
5571 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5573 35. Added +ignore_unknown as a special item in host lists.
5575 36. Code for decoding IPv6 addresses in host lists is now included, even if
5576 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5577 address was recognized as an IP address, but was then not correctly decoded
5578 into binary, causing unexpected and incorrect effects when compared with
5585 1. Very minor rewording of debugging text in manualroute to say "list of
5586 hosts" instead of "hostlist".
5588 2. If verify=header_syntax was set, and a header line with an unqualified
5589 address (no domain) and a large number of spaces between the end of the
5590 name and the colon was received, the reception process suffered a buffer
5591 overflow, and (when I tested it) crashed. This was caused by some obsolete
5592 code that should have been removed. The fix is to remove it!
5594 3. When running in the test harness, delay a bit after writing a bounce
5595 message to get a bit more predictability in the log output.
5597 4. Added a call to search_tidyup() just before forking a reception process. In
5598 theory, someone could use a lookup in the expansion of smtp_accept_max_
5599 per_host which, without the tidyup, could leave open a database connection.
5601 5. Added the variables $recipient_data and $sender_data which get set from a
5602 lookup success in an ACL "recipients" or "senders" condition, or a router
5603 "senders" option, similar to $domain_data and $local_part_data.
5605 6. Moved the writing of debug_print from before to after the "senders" test
5608 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5609 problems for message scanning, either using a data ACL, or using
5610 local_scan() because the Received: header was not generated till after they
5611 were called (in order to set the time as the time of reception completion).
5612 I have revised the way this works. The header is now generated after the
5613 body is received, but before the ACL or local_scan() are called. After they
5614 are run, the timestamp in the header is updated.
5620 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5621 before starting a queue runner without re-exec. This happened only when
5622 deliver_drop_privilege was set or when the Exim user was set to root. The
5623 effect of the bug was that timeouts during subsequent deliveries caused
5624 crashes instead of being properly handled. The handler is now left at its
5625 default (and expected) setting.
5627 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5628 message, again when deliver_drop_privilege is set or Exim is run as root.
5629 The bug described in (1) was not present in this case, but the tidying up
5630 of the other signals was missing. I have made the two cases consistent.
5632 3. The ignore_target_hosts setting on a manualroute router was being ignored
5633 for hosts that were looked up using the /MX notation.
5635 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5638 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5639 operated on the sender address. After changing the $sender_address to <>
5640 for the sender address verify, Exim was re-instated it as the original
5641 (before rewriting) address, but remembering that it had rewritten it, so it
5642 wasn't rewriting it again. This bug also had the effect of breaking the
5643 sender address verification caching when the sender address was rewritten.
5645 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5646 This has been changed so that if the ip literal address matches
5647 ignore_target_hosts, the router declines.
5649 7. Added expansion conditions match_domain, match_address, and match_local_
5650 part (NOT match_host).
5652 8. The placeholder for the Received: header didn't have a length field set.
5654 9. Added code to Exim itself and to exim_lock to test for a specific race
5655 condition that could lead to file corruption when using MBX delivery. The
5656 issue is with the lockfile that is created in /tmp. If this file is removed
5657 after a process has opened it but before that process has acquired a lock,
5658 there is the potential for a second process to recreate the file and also
5659 acquire a lock. This could lead to two Exim processes writing to the file
5660 at the same time. The added code performs the same test as UW imapd; it
5661 checks after acquiring the lock that its file descriptor still refers to
5662 the same named file.
5664 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5665 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5668 11. Added the smtp_active_hostname option. If used, this will typically be made
5669 to depend on the incoming interface address. Because $interface_address is
5670 not set up until the daemon has forked a reception process, error responses
5671 that can happen earlier (such as "too many connections") no longer contain
5674 12. If an expansion in a condition on a "warn" statement fails because a lookup
5675 defers, the "warn" statement is abandoned, and the next ACL statement is
5676 processed. Previously this caused the whole ACL to be aborted.
5678 13. Added the iplsearch lookup type.
5680 14. Added ident_timeout as a log selector.
5682 15. Added tls_certificate_verified as a log selector.
5684 16. Added a global option tls_require_ciphers (compare the smtp transport
5685 option of the same name). This controls incoming TLS connections.
5687 17. I finally figured out how to make tls_require_ciphers do a similar thing
5688 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5689 before starting the TLS session.
5691 18. Tabs are now shown as \t in -bP output.
5693 19. If the log selector return_path_on_delivery was set, Exim crashed when
5694 bouncing a message because it had too many Received: header lines.
5696 20. If two routers both had headers_remove settings, and the first one included
5697 a superfluous trailing colon, the final name in the first list and the
5698 first name in the second list were incorrectly joined into one item (with a
5699 colon in the middle).
5705 1. Added -C and -D options to the exinext utility, mainly to make it easier
5706 to include in the automated testing, but these could be helpful when
5707 multiple configurations are in use.
5709 2. The exinext utility was not formatting the output nicely when there was
5710 an alternate port involved in the retry record key, nor when there was a
5711 message id as well (for retries that were specific to a specific message
5712 and a specific host). It was also confused by IPv6 addresses, because of
5713 the additional colons they contain. I have fixed the IPv4 problem, and
5714 patched it up to do a reasonable job for IPv6.
5716 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5717 delivery, the log line now contains "pipelined" if PIPELINING was used.
5719 4. An SMTP transport process used to panic and die if the bind() call to set
5720 an explicit outgoing interface failed. This has been changed; it is now
5721 treated in the same way as a connect() failure.
5723 5. A reference to $sender_host_name in the part of a conditional expansion
5724 that was being skipped was still causing a DNS lookup. This no longer
5727 6. The def: expansion condition was not recognizing references to header lines
5728 that used bh_ and bheader_.
5730 7. Added the _cache feature to named lists.
5732 8. The code for checking quota_filecount in the appendfile transport was
5733 allowing one more file than it should have been.
5735 9. For compatibility with Sendmail, the command line option
5743 and sets the incoming protocol and host name (for trusted callers). The
5744 host name and its colon can be omitted when only the protocol is to be set.
5745 Note the Exim already has two private options, -pd and -ps, that refer to
5746 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5747 "s", but I don't think that's a major issue.
5749 10. A number of refactoring changes to the code, none of which should affect
5752 (a) The number of logging options was getting close to filling up the
5753 32-bit word that was used as a bit map. I have split them into two classes:
5754 those that are passed in the argument to log_write(), and those that are
5755 only ever tested independently outside of that function. These are now in
5756 separate 32-bit words, so there is plenty of room for expansion again.
5757 There is no change in the user interface or the logging behaviour.
5759 (b) When building, for example, log lines, the code previously used a
5760 macro that called string_cat() twice, in order to add two strings. This is
5761 not really sufficiently general. Furthermore, there was one instance where
5762 it was actually wrong because one of the argument was used twice, and in
5763 one call a function was used. (As it happened, calling the function twice
5764 did not affect the overall behaviour.) The macro has been replaced by a
5765 function that can join an arbitrary number of extra strings onto a growing
5768 (c) The code for expansion conditions now uses a table and a binary chop
5769 instead of a serial search (which was left over from when there were very
5770 few conditions). Also, it now recognizes conditions like "pam" even when
5771 the relevant support is not compiled in: a suitably worded error message is
5772 given if an attempt is made to use such a condition.
5774 11. Added ${time_interval:xxxxx}.
5776 12. A bug was causing one of the ddress fields not to be passed back correctly
5777 from remote delivery subprocesses. The field in question was not being
5778 subsequently used, so this caused to problems in practice.
5780 13. Added new log selectors queue_time and deliver_time.
5782 14. Might have fixed a bug in maildirsizefile handling that threw up
5783 "unexpected character" debug warnings, and recalculated the data
5784 unnecessarily. In any case, I expanded the warning message to give more
5787 15. Added the message "Restricted characters in address" to the statements in
5788 the default ACL that block characters like @ and % in local parts.
5790 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5791 Three changes have been made:
5793 (a) There was a serious bug; a negative response to MAIL caused the whole
5794 recipient domain to be cached as invalid, thereby blocking all messages
5795 to all local parts at the same domain, from all senders. This bug has
5796 been fixed. The domain is no longer cached after a negative response to
5797 MAIL if the sender used is not empty.
5799 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5802 (c) A new callout option, "use_sender" has been added for people who want
5803 the modified behaviour.
5809 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5812 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5813 indeed breaks things for older releases.
5815 3. Added additional logging to the case where there is a problem reading data
5816 from a filter that is running in a subprocess using a pipe, in order to
5817 try to track down a specific problem.
5819 4. Testing facility fudge: when running in the test harness and attempting
5820 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5821 getting "No route to host". Convert this to a timeout.
5823 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5826 6. Some OS don't have socklen_t but use size_t instead. This affects the
5827 fifth argument of getsockopt() amongst other things. This is now
5828 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5829 can be set for individual OS. I have set it for SunOS5, OSF1, and
5830 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5831 some earlier ones do not.
5833 7. Change 4.30/15 was not doing the test caselessly.
5835 8. The standard form for an IPv6 address literal was being rejected by address
5836 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5837 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5838 this, as well as the form without the "IPv6" on the front (but only when
5839 address literals are enabled, of course).
5841 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5843 10. Exim crashed if a message with an empty sender address specified by -f
5844 encountered a router with an errors_to setting. This could be provoked only
5845 by a command such as
5849 where an empty string was supplied; "<>" did not hit this bug.
5851 11. Installed PCRE release 4.5.
5853 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5854 remained set. It is now erased.
5856 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5857 times from message ids (which are base 36 rather than the normal 62).
5859 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5860 were being counted as actual protocol errors, and logged if the log
5861 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5862 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5863 and DATA following a set of rejected RCPTs do not count as protocol errors.
5864 In other words, Exim assumes they were pipelined, though this may not
5865 actually be the case. Of course, in all cases the client gets an
5866 appropriate error code.
5868 15. If a lookup fails in an ACL condition, a message about the failure may
5869 be available; it is used if testing the ACL cannot continue, because most
5870 such messages specify what the cause of the deferral is. However, some
5871 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5872 that caused an old message to be retained and used if a later statement
5873 caused a defer, replacing the real cause of the deferral.
5875 16. If an IP address had so many PTR records that the DNS lookup buffer
5876 was not large enough to hold them, Exim could crash while trying to process
5877 the truncated data. It now detects and logs this case.
5879 17. Further to 4.21/58, another change has been made: if (and only if) the
5880 first line of a message (the first header line) ends with CRLF, a bare LF
5881 in a subsequent header line has a space inserted after it, so as not to
5882 terminate the header.
5884 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5885 unnecessarily, used atoi() instead of strtol(), and didn't check the
5886 termination when getting file sizes from file names by regex.
5888 19. Completely re-implemented the support for maildirsize files, in the light
5889 of a number of problems with the previous contributed implementation
5890 (4.30/29). In particular:
5892 . If the quota is zero, the maildirsize file is maintained, but no quota is
5895 . If the maildir directory does not exist, it is created before any attempt
5896 to write a maildirsize file.
5898 . The quota value in the file is just a cache; if the quota is changed in
5899 the transport, the new value overrides.
5901 . A regular expression is available for excluding directories from the
5904 20. The autoreply transport checks the characters in options that define the
5905 message's headers; it allows continued headers, but it was checking with
5906 isspace() after an embedded newline instead of explicitly looking for a
5909 21. If all the "regular" hosts to which an address was routed had passed their
5910 expiry times, and had not reached their retry times, the address was
5911 bounced, even if fallback hosts were defined. Now Exim should go on to try
5914 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5915 equivalent code in the SMTP transport. Some hosts send humungous responses
5916 to HELO/EHLO, more than 1024 it seems.
5918 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5919 gives compiler warnings in some environments. I've now done it "properly",
5922 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5923 (because of problems with the built-in one) was declared to return uschar *
5924 instead of char *, causing compiler failure.
5926 25. Fixed a file descriptor leak when processing alias/forward files.
5928 26. Fixed a minor format string issue in dbfn.c.
5930 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5932 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5933 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5936 29. When a sender address is verified, it is cached, to save repeating the test
5937 when there is more than one recipient in a message. However, when the
5938 verification involves a callout, it is possible for different callout
5939 options to be set for different recipients. It is too complicated to keep
5940 track of this in the cache, so now Exim always runs a verification when a
5941 callout is required, relying on the callout cache for the optimization.
5942 The overhead is duplication of the address routing, but this should not be
5945 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5946 to be verified with callout=postmaster, and the main callout worked but the
5947 postmaster check failed, the verification correctly failed. However, if a
5948 subsequent RCPT command asked for sender verification *without* the
5949 postmaster check, incorrect caching caused this verification also to fail,
5952 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5953 it was not caching the DNS options (qualify_single, search_parents) that
5954 were used when the lookup failed. A subsequent lookup with different
5955 options therefore always gave the same answer, though there were cases
5956 where it should not have. (Example: a "domains = !$mx_any" option on a
5957 dnslookup router: the "domains" option is always processed without any
5958 widening, but the router might have qualify_single set.) Now Exim uses the
5959 cached value only when the same options are set.
5961 32. Added John Jetmore's "exipick" utility to the distribution.
5963 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5964 than a timeout (e.g. a certificate is required, and is not provided), an
5965 Exim server now closes the connection immediately. Previously it waited for
5966 the client to close - but if the client is SSL, it seems that they each
5967 wait for each other, leading to a delay before one of them times out.
5969 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5970 maintained 0.8.x compatibility because I don't think many are using it, and
5971 it is clearly obsolete.
5973 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5976 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5977 client certificate was expired. A simple patch fixes this, though I don't
5978 understand the full logic of why the verify callback is called multiple
5981 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5982 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5983 which causes problems with some clients (such as the Certicom SSL Plus
5984 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5985 disables the coutermeasure allowing Eudora to connect."
5987 38. Exim was not checking that a write() to a log file succeeded. This could
5988 lead to Bad Things if a log got too big, in particular if it hit a file
5989 size limit. Exim now panics and dies if it cannot write to a log file, just
5990 as it does if it cannot open a log file.
5992 39. Modified OS/Makefile-Linux so that it now contains
5994 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5996 The two -D definitions ensure that Exim is compiled with large file
5997 support, which makes it possible to handle log files that are bigger than
6000 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
6001 instance) a domain was checked against a named list that involved a lookup,
6002 causing $domain_data to be set, then another domain was checked against the
6003 same list, then the first domain was re-checked, the value of $domain_data
6004 after the final check could be wrong. In particular, if the second check
6005 failed, it could be set empty. This bug probably also applied to
6008 41. The strip_trailing_dot option was not being applied to the address given
6009 with the -f command-line option.
6011 42. The code for reading a message's header from the spool was incrementing
6012 $received_count, but never initializing it. This meant that the value was
6013 incorrect (doubled) while delivering a message in the same process in which
6014 it was received. In the most common configuration of Exim, this never
6015 happens - a fresh exec is done - but it can happen when
6016 deliver_drop_privilege is set.
6018 43. When Exim logs an SMTP synchronization error - client data sent too soon -
6019 it now includes up to 150 characters of the unexpected data in the log
6022 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6023 and building data strings. The size of both of these buffers was 10 000
6024 bytes - far larger than anybody would *ever* want, thought I. Needless to
6025 say, somebody hit the limit. I have increased the maximum line length to
6026 20 000 and the maximum data length of concatenated lines to 100 000. I have
6027 also fixed two bugs, because there was no checking on these buffers. Tsk,
6028 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6029 buffer is too small.
6031 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6032 lsearch lookups. Now it does.
6034 46. When parsing a route_list item in a manualroute router, a fixed-length
6035 buffer was used for the list of hosts. I made this 1024 bytes long,
6036 thinking that nobody would ever have a list of hosts that long. Wrong.
6037 Somebody had a whole pile of complicated expansion conditions, and the
6038 string was silently truncated, leading to an expansion error. It turns out
6039 that it is easier to change to an unlimited length (owing to other changes
6040 that have happened since this code was originally written) than to build
6041 structure for giving a limitation error. The length of the item that
6042 expands into the list of hosts is now unlimited.
6044 47. The lsearch lookup could not handle data where the length of text line was
6045 more than 4095 characters. Such lines were truncated, leading to shortened
6046 data being returned. It should now handle lines of any length.
6048 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6049 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6052 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6053 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6054 no longer happens in comment lines. A list of replacements is now placed
6055 at the head of all of the source files, except those whose only change is
6056 to replace PERL_COMMAND in the very first #! line.
6058 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6059 messages on the queue, with a bottom-up merge sort, using code contributed
6060 by Michael Haardt. This should make operations like -bp somewhat faster on
6061 large queues. It won't affect queue runners, except when queue_run_in_order
6064 51. Installed eximstats 1.31 in the distribution.
6066 52. Added support for SRV lookups to the dnslookup router.
6068 53. If an ACL referred to $message_body or $message_body_end, the value was not
6069 reset for any messages that followed in the same SMTP session.
6071 54. The store-handling optimization for building very long strings was not
6072 differentiating between the different store pools. I don't think this
6073 actually made any difference in practice, but I've tidied it.
6075 55. While running the routers to verify a sender address, $sender_address
6076 was still set to the sender address. This is wrong, because when routing to
6077 send a bounce to the sender, it would be empty. Therefore, I have changed
6078 it so that, while verifying a sender address, $sender_address is set to <>.
6079 (There is no change to what happens when verifying a recipient address.)
6081 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6082 target A or AAAA records (if not already returned) without resetting the
6083 qualify_single or search_parents options of the DNS resolver. These are
6084 inappropriate in this case because the targets of MX and SRV records must
6085 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6086 target that, when qualified, matched something in the local domain. These
6087 two options are now turned off when doing these lookups.
6089 57. It seems that at least some releases of Reiserfs (which does not have the
6090 concept of a fixed number of inodes) returns zero and not -1 for the
6091 number of available inodes. This interacted badly with check_spool_inodes,
6092 which assumed that -1 was the "no such thing" setting. What I have done is
6093 to check that the total number of inodes is greater than zero before doing
6094 the test of how many are available.
6096 58. When a "warn" ACL statement has a log_message modifier, the message is
6097 remembered, and not repeated. This is to avoid a lot of repetition when a
6098 message has many recipients that cause the same warning to be written.
6099 However, Exim was preserving the list of already written lines for an
6100 entire SMTP session, which doesn't seem right. The memory is now reset if a
6101 new message is started.
6103 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6104 debugging output unless log_rewrite was also set.
6106 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6107 of (int)(handle) when we know that handle contains (void *)(-1).
6109 61. The Exim daemon panic-logs an error return when it closes the incoming
6110 connection. However "connection reset by peer" seems to be common, and
6111 isn't really an error worthy of noting specially, so that particular error
6114 62. When Exim is trying to find all the local interfaces, it used to panic and
6115 die if the ioctl to get the interface flags failed. However, it seems that
6116 on at least one OS (Solaris 9) it is possible to have an interface that is
6117 included in the list of interfaces, but for which you get a failure error
6118 for this call. This happens when the interface is not "plumbed" into a
6119 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6120 failure of the "get flags" call assumes that the interface is down.
6122 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6123 makes life easier for people who are doing arithmetic on fields extracted
6124 from dates, where you often get leading zeros that should not be
6125 interpreted as octal.
6127 64. Added qualify_domain to the redirect router, to override the global
6130 65. If a pathologically long header line contained very many addresses (the
6131 report of this problem mentioned 10 000) and each of them was rewritten,
6132 Exim could use up a very large amount of memory. (It kept on making new
6133 copies of the header line as it rewrote, and never released the old ones.)
6134 At the expense of a bit more processing, the header rewriting function has
6135 been changed so that it no longer eats memory in this way.
6137 66. The generation of the Received: header has been moved from the time that a
6138 message starts to be received, to the time that it finishes. The timestamp
6139 in the Received: header should now be very close to that of the <= log
6140 line. There are two side-effects of this change:
6142 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6143 logged header lines no longer include the local Received: line, because
6144 it has not yet been created. The same applies to a copy of the message
6145 that is returned to a non-SMTP sender when a message is rejected.
6147 (b) When a filter file is tested using -bf, no additional Received: header
6148 is added to the test message. After some thought, I decided that this
6151 This change does not affect the value of $received_for. It is still set
6152 after address rewriting, but before local_scan() is called.
6154 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6156 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6157 gave an unhelpful panic error message, and a defer error. I have managed to
6158 change this behaviour so that it now rejects any supplied certificate,
6159 which seems right, as the list of acceptable certificates is empty.
6161 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6162 gave an unhelpful defer error. I have not managed to make this reject any
6163 supplied certificates, but the error message it gives is "no certificate
6164 supplied", which is not helpful.
6166 70. exigrep's output now also includes lines that are not associated with any
6167 message, but which match the given pattern. Implemented by a patch from
6168 Martin Sluka, which also tidied up the Perl a bit.
6170 71. Recipient callout verification, like sender verification, was using <> in
6171 the MAIL FROM command. This isn't really the right thing, since the actual
6172 sender may affect whether the remote host accepts the recipient or not. I
6173 have changed it to use the actual sender in the callout; this means that
6174 the cache record is now keyed on a recipient/sender pair, not just the
6175 recipient address. There doesn't seem to be a real danger of callout loops,
6176 since a callout by the remote host to check the sender would use <>.
6177 [SEE ABOVE: changed after hitting problems.]
6179 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6180 temporary errors. However, in the case of such a code being given after
6181 the end of a data transmission (i.e. after ".") Exim was failing to write
6182 a retry record for the message. (Yes, there was some broken host that was
6183 actually sending 8xx at this point.)
6185 73. An unknown lookup type in a host list could cause Exim to panic-die when
6186 the list was checked. (An example that provoked this was putting <; in the
6187 middle of a list instead of at the start.) If this happened during a DATA
6188 ACL check, a -D file could be left lying around. This kind of configuration
6189 error no longer causes Exim to die; instead it causes a defer error. The
6190 incident is still logged to the main and panic logs.
6192 74. Buglet left over from Exim 3 conversion. The message "too many messages
6193 in one connection" was written to the rejectlog but not the mainlog, except
6194 when address rewriting (yes!) was being logged.
6196 75. Added write_rejectlog option.
6198 76. When a system filter was run not as root (that is, when system_filter_user
6199 was set), the values of the $n variables were not being returned to the
6200 main process; thus, they were not subsequently available in the $sn
6203 77. Added +return_path_on_delivery log selector.
6205 78. A connection timeout was being treated differently from recipients deferred
6206 when testing hosts_max_try with a message that was older than the host's
6207 retry timeout. (The host should not be counted, thus allowing all hosts to
6208 be tried at least once before bouncing.) This may have been the cause of an
6209 occasionally reported bug whereby a message would remain on the queue
6210 longer than the retry timeout, but would be bounced if a delivery was
6211 forced. I say "may" because I never totally pinned down the problem;
6212 setting up timeout/retry tests is difficult. See also the next item.
6214 79. The ultimate address timeout was not being applied to errors that involved
6215 a combination of host plus message (for example, a timeout on a MAIL
6216 command). When an address resolved to a number of possible hosts, and they
6217 were not all tried for each delivery (e.g. because of hosts_max_try), a
6218 message could remain on the queue longer than the retry timeout.
6220 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6223 81. Fixed an obscure SMTP outgoing bug which required at least the following
6224 conditions: (a) there was another message waiting for the same server;
6225 (b) the server returned 5xx to all RCPT commands in the first message so
6226 that the message was not completed; (c) the server dropped the connection
6227 or gave a negative response to the RSET that Exim sends to abort the
6228 transaction. The observed case was a dropped connection after DATA that had
6229 been sent in pipelining mode. That is, the server had advertised PIPELINING
6230 but was not implementing it correctly. The effect of the bug was incorrect
6231 behaviour, such as trying another host, and this could lead to a crash.
6237 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6238 and daemon.c were passed as pointers to ints; they should have been
6239 pointers to socklen_t variables (which are typically unsigned ints).
6241 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6244 3. Fixed a really odd bug that affected only the testing scheme; patching a
6245 certain fixed string in the binary changed the value of another string that
6246 happened to be identical to the end of the original first string.
6248 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6249 name", it returns that address as the IP address. On some operating
6250 systems (e.g. Solaris), it also passes back the IP address string as the
6251 "host name". However, on others (e.g. Linux), it passes back an empty
6252 string. Exim wasn't checking for this, and was changing the host name to an
6253 empty string, assuming it had been canonicalized.
6255 5. Although rare, it is permitted to have more than one PTR record for a given
6256 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6257 all the names associated with an address, because they do in Solaris.
6258 However, it seems that they do not in Linux for data that comes from the
6259 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6260 I found this out when I moved to a new Linux workstation and tried to run
6261 the Exim test suite.
6263 To get round this problem I have changed the code so that it now does its
6264 own call to the DNS to look up PTR records when searching for a host name.
6265 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6266 addresses that are only in /etc/hosts are still found.
6268 This behaviour is, however, controlled by an option called host_lookup_
6269 order, which defaults to "bydns:byaddr". If people want to use the other
6270 order, or indeed, just use one or the other means of lookup, they can
6271 specify it in this variable.
6273 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6274 some operating systems, this comes back from gethostbyaddr() as an empty
6275 string, and this is what Exim used to test for. However, it seems that in
6276 other systems, "." is yielded. Exim now tests for this case too.
6278 7. The values of check_spool_space and check_log_space are now held internally
6279 as a number of kilobytes instead of an absolute number of bytes. If a
6280 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6281 kilobyte. This means that much larger values can be stored.
6283 8. Exim monitor: an attempt to get the action menu when not actually pointing
6284 at a message produces an empty menu entitled "No message selected". This
6285 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6286 no entries in it ("Shell widget menu has zero width and/or height"). So I
6287 have added a single, blank menu entry in this case.
6289 9. Added ${quote_local_part.
6291 10. MIME decoding is now applied to the contents of Subject: header lines when
6294 11. Now that a reference to $sender_host_address automatically causes a reverse
6295 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6296 host lookup before query-style lookups in lists that might use this
6297 variable. This has therefore been abolished, and the "net-" prefix is no
6298 longer necessary for query-style lookups.
6300 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6301 have been a typo for LFLAGS, so it has been changed.
6303 13. The install script calls Exim with "-C /dev/null" in order to find the
6304 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6305 to be output. However, since Exim outputs its version number before the
6306 error, it didn't break the script. It just looked ugly. I fixed this by
6307 always allowing "-C /dev/null" if the caller is root.
6309 14. Ignore overlarge ACL variable number when reading spool file - insurance
6310 against a later release with more variables having written the file.
6312 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6313 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6314 this, as well as the form without the "IPv6" on the front.
6316 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6317 OS/Makefile-Darwin file.
6319 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6320 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6322 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6323 variable. However, some OS use ints for inodes. I've added cast to long int
6324 to get rid of the compiler warning.
6326 19. I had forgotten to lock out "/../" in configuration file names when
6327 ALT_CONFIG_PREFIX was set.
6329 20. Routers used for verification do not need to specify transports. However,
6330 if such a router generated a host list, and callout was configured, Exim
6331 crashed, because it could not find a port number from the (non-existent)
6332 transport. It now assumes port 25 in this circumstance.
6334 21. Added the -t option to exigrep.
6336 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6337 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6338 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6339 them is set, LOOKUP_LSEARCH is forced.
6341 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6343 24. Added sender and host information to the "rejected by local_scan()" log
6344 line; previously there was no indication of these.
6346 25. Added .include_if_exists.
6348 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6349 Linux. It turns out that not all file systems support this. Apparently some
6350 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6351 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6352 not supported on the file descriptor, is now ignored when Exim is trying to
6353 sync a directory. This applies only to Linux.
6355 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6357 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6358 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6361 29. Added support for maildirsize files from supplied patch (modified a bit).
6363 30. The use of :fail: followed by an empty string could lead Exim to respond to
6364 sender verification failures with (e.g.):
6366 550 Verification failed for <xxx>
6367 550 Sender verify failed
6369 where the first response line was missing the '-' that indicates it is not
6370 the final line of the response.
6372 31. The loop for finding the name of the user that called Exim had a hardwired
6373 limit of 10; it now uses the value of finduser_retries, which is used for
6374 all other user lookups.
6376 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6379 33. Exim was neglecting to zero errno before one call of strtol() when
6380 expanding a string and expecting an integer value. On some systems this
6381 resulted in spurious "integer overflow" errors. Also, it was casting the
6382 result into an int without checking.
6384 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6385 did not work. The code looks as if it has *never* worked, though it appears
6386 to have been documented since at least release 1.62. I have made it work.
6388 35. The "timeout_DNS" error in retry rules, also documented since at least
6389 1.62, also never worked. As it isn't clear exactly what this means, and
6390 clearly it isn't a major issue, I have abolished the feature by treating it
6391 as "timeout", and writing a warning to the main and panic logs.
6393 36. The display of retry rules for -brt wasn't always showing the error code
6396 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6397 timeout_connect_A, timeout_connect_MX.
6399 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6400 to the empty sender.
6402 39. The daemon was not analysing the content of -oX till after it had closed
6403 stderr and disconnected from the controlling terminal. This meant that any
6404 syntax errors were only noted on the panic log, and the return code from
6405 the command was 0. By re-arranging the code a little, I've made the
6406 decoding happen first, so such errors now appear on stderr, and the return
6407 code is 1. However, the actual setting up of the sockets still happens in
6408 the disconnected process, so errors there are still only recorded on the
6411 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6412 connections (as happens on some IP stacks) was logged at start up time as
6413 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6414 it from "IPv6 and IPv4", which means that two separate sockets are being
6417 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6418 says whether AF_INET or AF_INET6 was passed as an argument.
6420 42. Exiwhat output was messed up when time zones were included in log
6423 43. Exiwhat now gives more information about the daemon's listening ports,
6424 and whether -tls-on-connect was used.
6426 44. The "port" option of the smtp transport is now expanded.
6428 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6429 silently ignored. Now an error message is written to the main and panic
6432 46. There's a new ACL modifier called "logwrite" which writes to a log file
6433 as soon as it is encountered.
6435 47. Added $local_user_uid and $local_user_gid at routing time.
6437 48. Exim crashed when trying to verify a sender address that was being
6440 49. Exim was recognizing only a space character after ".include". It now also
6441 recognizes a tab character.
6443 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6444 extracting the relevant information from the specification. The man page no
6445 longer contains scrambled data for the -d option, and I've added a section
6446 at the front about calling Exim under different names.
6448 51. Added "extra_headers" argument to the "mail" command in filter files.
6450 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6453 53. Installed eximstats 1.29.
6455 54. Added transport_filter_timeout as a generic transport option.
6457 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6458 Cc: header lines. This was required by RFC 822, but it not required by RFC
6461 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6462 incoming messages that did not have them. Now it does so only if the
6463 message originates locally, that is, if there is no associated remote host
6464 address. When Resent- header lines are present, this applies to the Resent-
6465 lines rather than the non-Resent- lines.
6467 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6468 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6470 58. Messages for configuration errors now include the name of the main
6471 configuration file - useful now that there may be more than one file in a
6472 list (.included file names were always shown).
6474 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6475 for those rare installations that do not start the daemon as root or run it
6476 setuid root. I've cut out the call to initgroups() if the daemon is not
6479 60. The Exim user and group can now be bound into the binary as text strings
6480 that are looked up at the start of Exim's processing.
6482 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6484 62. Added $mailstore_basename variable.
6486 63. Installed patch to sieve.c from Michael Haardt.
6488 64. When Exim failed to open the panic log after failing to open the main log,
6489 the original message it was trying to log was written to stderr and debug
6490 output, but if they were not available (the usual case in production), it
6491 was lost. Now it is written to syslog before the two lines that record the
6492 failures to open the logs.
6494 65. Users' Exim filters run in subprocesses under the user's uid. It is
6495 possible for a "deliver" command or an alias in a "personal" command to
6496 provoke an address rewrite. If logging of address rewriting is configured,
6497 this fails because the process is not running as root or exim. There may be
6498 a better way of dealing with this, but for the moment (because 4.30 needs
6499 to be released), I have disabled address rewrite logging when running a
6500 filter in a non-root, non-exim process.
6506 1. The buildconfig auxiliary program wasn't quoting the value set for
6507 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6508 not defined. This bug was masked in 4.22 by the effect that was fixed in
6511 2. Some messages that were rejected after a message id was allocated were
6512 shown as "incomplete" by exigrep. It no longer does this for messages that
6513 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6515 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6516 have allow_domain_literals set, the ID did not get logged in the <= line.
6517 Domain literals are now always recognized in Message-ID: header lines.
6519 4. The first argument for a ${extract expansion item is the key name or field
6520 number. Leading and trailing spaces in this item were not being ignored,
6521 causing some misleading effects.
6523 5. When deliver_drop_privilege was set, single queue runner processes started
6524 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6525 same command in the process it spins off) were not dropping privilege.
6527 6. When the daemon running as "exim" started a queue runner, it always
6528 re-executed Exim in the spun-off process. This is a waste of effort when
6529 deliver_drop_privilege is set. The new process now just calls the
6530 queue-runner function directly.
6536 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6539 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6540 ignored. Though the use of -oP was forcing the writing of a pid file, it
6541 was always written to the default place.
6543 3. If the message "no IP address found for host xxxx" is generated during
6544 incoming verification, it is now followed by identification of the incoming
6545 connection (so you can more easily find what provoked it).
6547 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6549 5. Added some features to "harden" Exim a bit more against certain attacks:
6551 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6552 be put in Local/Makefile. This is like the never_users runtime option,
6553 but it cannot be overridden. The default setting is "root".
6555 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6556 prefix string with which any file named in a -C command line option
6559 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6560 is retained for -C and -D only if the caller of Exim is root. Without
6561 it, the exim user may also use -C and -D and retain privilege.
6563 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6564 command line option is disabled.
6566 6. Macro names set by the -D option must start with an upper case letter, just
6567 like macro names defined in the configuration file.
6569 7. Added "dereference=" facility to LDAP.
6571 8. Two instances of the typo "uknown" in the source files are fixed.
6573 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6574 the Configure-Makefile script screwed up while processing it.
6576 10. Incorporated PCRE 4.4.
6578 11. The SMTP synchronization check was not operating right at the start of an
6579 SMTP session. For example, it could not catch a HELO sent before the client
6580 waited for the greeting. There is now a check for outstanding input at the
6581 point when the greeting is written. Because of the duplex, asynchronous
6582 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6583 way, but not yet received, when the check is performed.
6585 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6586 on TCP/IP sockets, because this apparently causes some broken clients to
6589 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6590 unchanged) from the Cygwin maintainer.
6592 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6593 was supported instead of testing for mbx. Effectively a typo.
6595 15. The spa authenticator server code was not checking that the input it
6596 received was valid base64.
6598 16. The debug output line for the "set" modifier in ACLs was not showing the
6599 name of the variable that was being set.
6601 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6603 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6604 lookup on its own. Something else was needed to trigger the lookup. For
6605 example, a match in host_lookup or the need for a host name in a host list.
6606 Now, if $sender_host_name is referenced and the host name has not yet been
6607 looked up, a lookup is performed. If the lookup fails, the variable remains
6608 empty, and $host_lookup_failed is set to "1".
6610 19. Added "eqi" as a case-independent comparison operator.
6612 20. The saslauthd authentication condition could segfault if neither service
6613 nor realm was specified.
6615 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6616 error message that was logged was misleading, and incoming SMTP
6617 connections were dropped. The message is now more accurate, and temporary
6618 errors are given to SMTP connections.
6620 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6621 (except RSET) with a 421 error, until QUIT is received. However, it was
6622 failing to send a response to QUIT.
6624 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6625 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6626 ACL is now run after the helo_try_verify_hosts code.
6628 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6629 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6630 case-independent, so other case variants are also recognized). Apparently
6631 some systems use these upper case variants.
6633 25. If more than two messages were waiting for the same host, and a transport
6634 filter was specified for the transport, Exim sent two messages over the
6635 same TCP/IP connection, and then failed with "socket operation on non-
6636 socket" when it tried to send the third.
6638 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6640 27. The extern definition of crypt16() in expand.c was not being excluded when
6641 the OS had its own crypt16() function.
6643 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6644 as a preferred synonym for return_size_limit, both as an option and as an
6647 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6649 30. Changed the default configuration ACL to relax the local part checking rule
6650 for addresses that are not in any local domains. For these addresses,
6651 slashes and pipe symbols are allowed within local parts, but the sequence
6652 /../ is explicitly forbidden.
6654 31. SPA server authentication was not clearing the challenge buffer before
6657 32. log_message in a "warn" ACL statement was writing to the reject log as
6658 well as to the main log, which contradicts the documentation and doesn't
6659 seem right (because no rejection is happening). So I have stopped it.
6661 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6662 However, I am unable to do any testing of this.
6664 34. Fixed an infelicity in the appendfile transport. When checking directories
6665 for a mailbox, to see if any needed to be created, it was accidentally
6666 using path names with one or more superfluous leading slashes; tracing
6667 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6669 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6670 contents are added to the log line that is written for every discarded
6671 recipient. (Previously a log_message setting was ignored.)
6673 36. The ${quote: operator now quotes the string if it is empty.
6675 37. The install script runs exim in order to find its version number. If for
6676 some reason other than non-existence or emptiness, which it checks, it
6677 could not run './exim', it was installing it with an empty version number,
6678 i.e. as "exim-". This error state is now caught, and the installation is
6681 38. An argument was missing from the function that creates an error message
6682 when Exim fails to connect to the socket for saslauthd authentication.
6683 This could cause Exim to crash, or give a corrupted message.
6685 39. Added isip, isip4, and isip6 to ${if conditions.
6687 40. The ACL variables $acl_xx are now saved with the message, and can be
6688 accessed later in routers, transports, and filters.
6690 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6691 strings in the file are not string-expanded.
6693 42. If a MAIL command specified a SIZE value that was too large to fit into an
6694 int variable, the check against message_size_limit failed. Such values are
6695 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6696 day this will have to be increased, but I don't think I want to be around
6697 when emails are that large.
6704 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6705 iconv() is not standard in FreeBSD.
6707 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6708 IPv6 enabled. The observed symptom was a segmentation fault on return from
6709 the function os_common_find_running_interfaces() in src/os.c.
6711 3. In the check_special_case() function in daemon.c I had used "errno" as an
6712 argument name, which causes warnings on some systems. This was basically a
6713 typo, since it was named "eno" in the comments!
6715 4. The code that waits for the clock to tick (at a resolution of some fraction
6716 of a second) so as to ensure message-id uniqueness was always waiting for
6717 at least one whole tick, when it could have waited for less. [This is
6718 almost certainly not relevant at current processor speeds, where it is
6719 unlikely to ever wait at all. But we try to future-proof.]
6721 5. The function that sleeps for a time interval that includes fractions of a
6722 second contained a race. It did not block SIGALRM between setting the
6723 timer, and suspending (a couple of lines later). If the interval was short
6724 and the sigsuspend() was delayed until after it had expired, the suspension
6725 never ended. On busy systems this could lead to processes getting stuck for
6728 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6729 process, before it forks any delivery processes. The open lookup caching
6730 mechanism meant that the open file or database connection was passed into
6731 the delivery process. The problem was that delivery processes always tidy
6732 up cached lookup data. This could cause a problem for the next delivery
6733 process started by the queue runner, because the external queue runner
6734 process does not know about the closure. So the next delivery process
6735 still has data in the lookup cache. In the case of a file lookup, there was
6736 no problem because closing a file descriptor in a subprocess doesn't affect
6737 the parent. However, if the lookup was caching a connection to a database,
6738 the connection was closed, and the second delivery process was likely to
6739 see errors such as "PGSQL: query failed: server closed the connection
6740 unexpectedly". The problem has been fixed by closing all cached lookups
6741 in a queue runner before running a delivery process.
6743 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6744 seem to have the "const" qualifier which it has on other OS. I've
6747 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6748 *of the same type* (client or server) with the same public name that an
6749 error should be diagnosed.
6751 9. When Exim looked up a host name for an IP address, but failed to find the
6752 original IP address when looking up the host name (a safety check), it
6753 output the message "<ip address> does not match any IP for NULL", which was
6754 confusing, to say the least. The bug was that the host name should have
6755 appeared instead of "NULL".
6757 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6758 user that is built into the binary, and the -C or -D options is used, root
6759 privilege is dropped before the configuration file is read. In addition,
6760 logging is switched to stderr instead of the normal log files. If the
6761 configuration then re-defines the Exim user, the unprivileged environment
6762 is probably not what is expected, so Exim logs a panic warning message (but
6765 However, if deliver_drop_privilege is set, the unprivileged state may well
6766 be exactly what is intended, so the warning has been cut out in that case,
6767 and Exim is allowed to try to write to its normal log files.
6773 1. smtp_return_error_details was not giving details for temporary sender
6774 or receiver verification errors.
6776 2. Diagnose a configuration error if two authenticators have the same public
6779 3. Exim used not to create the message log file for a message until the first
6780 delivery attempt. This could be confusing when incoming messages were held
6781 for policy or load reasons. The message log file is now created at the time
6782 the message is received, and an initial "Received" line is written to it.
6784 4. The automatically generated man page for command line options had a minor
6785 bug that caused no ill effects; however, a more serious problem was that
6786 the procedure for building the man page automatically didn't always
6787 operate. Consequently, release 4.20 contains an out-of-date version. This
6788 shouldn't happen again.
6790 5. When building Exim with embedded Perl support, the script that builds the
6791 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6792 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6794 6. The freeze_tell option was not being used for messages that were frozen on
6795 arrival, either by an ACL or by local_scan().
6797 7. Added the smtp_incomplete_transaction log selector.
6799 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6800 was accepting AUTH without a new EHLO.
6802 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6803 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6806 10. Logging of TCP/IP connections (when configured) now happens in the main
6807 daemon process instead of the child process, so that the TCP/IP connection
6808 count is more accurate (but it can never be perfect).
6810 11. The use of "drop" in a nested ACL was not being handled correctly in the
6811 outer ACL. Now, if condition failure induced by the nested "drop" causes
6812 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6813 or "require"), the connection is dropped.
6815 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6816 that yield "discard" can now be used with an "accept" or a "discard" verb,
6817 but an error is generated for any others (because I can't see a useful way
6818 to define what should happen).
6820 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6821 are now processed in the same way as lines in the Exim configuration file.
6822 In particular, continuation lines are supported.
6824 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6826 15. Added -ti meaning -t -i.
6828 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6829 domains, and warn by logging if others are found.
6831 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6832 returned from GIFCONF when Exim is trying to find the list of interfaces on
6833 a host. The code in os.c has been modified to copy each ifreq to an aligned
6834 structure in all cases.
6836 Also, in some cases, the returned ifreq's were being copied to a 'struct
6837 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6838 means the last couple of bytes of an IPv6 address could be chopped if the
6839 ifreq contained only a normal sockaddr (14 bytes storage).
6841 18. Named domain lists were not supported in the hosts_treat_as_local option.
6842 An entry such as +xxxx was not recognized, and was treated as a literal
6845 19. Ensure that header lines added by a DATA ACL are included in the reject log
6846 if the ACL subsequently rejects the message.
6848 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6849 MD5 (which is deprecated).
6851 21. When testing a filter file using -bf, Exim was writing a message when it
6852 took the sender from a "From " line in the message, but it was not doing so
6853 when it took $return_path from a Return-Path: header line. It now does.
6855 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6856 with a valid header line field name (a series of printing characters
6857 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6859 23. Changed "disc" in the source to "disk" to conform to the documentation and
6860 the book and for uniformity.
6862 24. Ignore Sendmail's -Ooption=value command line item.
6864 25. When execve() failed while trying to run a command in a pipe transport,
6865 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6866 could be confused with a return value of 69 from the command itself. This
6867 has been changed to 127, the value the shell returns if it is asked to run
6868 a non-existent command. The wording for the related log line suggests a
6869 non-existent command as the problem.
6871 26. If received_header_text expands to an empty string, do not add a Received:
6872 header line to the message. (Well, it adds a token one on the spool, but
6873 marks it "old" so that it doesn't get used or transmitted.)
6875 27. Installed eximstats 1.28 (addition of -nt option).
6877 28. There was no check for failure on the call to getsockname() in the daemon
6878 code. This can fail if there is a shortage of resources on the system, with
6879 ENOMEM, for example. A temporary error is now given on failure.
6881 29. Contrary to the C standard, it seems that in some environments, the
6882 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6883 program. Exim now does this explicitly; it affects the formatting of
6884 timestamps using strftime().
6886 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6887 complaints. I've now initialized all the variables, to avoid this.
6889 32. Header lines added by a system filter were not being "seen" during
6890 transport-time rewrites.
6892 33. The info_callback() function passed to OpenSSL is set up with type void
6893 (*)(SSL *, int, int), as described somewhere. However, when calling the
6894 function (actually a macro) that sets it up, the type void(*)() is
6895 expected. I've put in a cast to prevent warnings from picky compilers.
6897 34. If a DNS black list lookup found a CNAME record, but there were no A
6898 records associated with the domain it pointed at, Exim crashed.
6900 35. If a DNS black list lookup returned more than one A record, Exim ignored
6901 all but the first. It now scans all returned addresses if a particular IP
6902 value is being sought. In this situation, the contents of the
6903 $dnslist_value variable are a list of all the addresses, separated by a
6906 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6907 to accept a host name and all its aliases if the forward lookup for any of
6908 them yielded the IP address of the incoming connection. Now it accepts only
6909 those names whose forward lookup yields the correct IP address. Any other
6910 names are discarded. This closes a loophole whereby a rogue DNS
6911 administrator could create reverse DNS records to break through a
6912 wildcarded host restriction in an ACL.
6914 37. If a user filter or a system filter that ran in a subprocess used any of
6915 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6916 the wrong values were passed to the pipe command ($thisaddress had the
6917 value of $0, $0 had the value of $1, etc). This bug was introduced by
6918 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6920 38. Improved the line breaking for long SMTP error messages from ACLs.
6921 Previously, if there was no break point between 40 and 75 characters, Exim
6922 left the rest of the message alone. Two changes have been made: (a) I've
6923 reduced the minimum length to 35 characters; (b) if it can't find a break
6924 point between 35 and 75 characters, it looks ahead and uses the first one
6925 that it finds. This may give the occasional overlong line, but at least the
6926 remaining text gets split now.
6928 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6929 file descriptors might be low, and that setting 1000 would always raise it.
6930 It turns out that in some environments, the limit is already over 1000 and
6931 that lowering it causes trouble. So now Exim takes care not to decrease it.
6933 40. When delivering a message, the value of $return_path is set to $sender_
6934 address at the start of routing (routers may change the value). By an
6935 oversight, this default was not being set up when an address was tested by
6936 -bt or -bv, which affected the outcome if any router or filter referred to
6939 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6940 log only when "message" or "log_message" are set. However, if one of the
6941 conditions was an address verification, or a call to a nested ACL, the
6942 messages generated by the underlying test were being passed through. This
6943 no longer happens. The underlying message is available in $acl_verify_
6944 message for both "message" and "log_message" expansions, so it can be
6945 passed through if needed.
6947 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6948 new expansion $bh_ to give the encoded byte string without charset
6949 translation. Translation happens only if iconv() is available; HAVE_ICONV
6950 indicates this at build time. HEADERS_CHARSET gives the charset to
6951 translate to; headers_charset can change it in the configuration, and
6952 "headers charset" can change it in an individual filter file.
6954 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6955 that creates RFC 2047 encoded "words" labels them as that charset instead
6956 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6957 expansion operator; (ii) when Exim creates a From: line for a local
6958 message; (iii) when a header line is rewritten to include a "phrase" part.
6960 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6961 buggy, causing it to skip the first lines of messages whose message ID
6962 ended in 'D'. This would not have bitten before Exim release 4.14, because
6963 message IDs were unlikely to end in 'D' before then. The effect was to have
6964 incorrect size information for certain domains.
6966 45. #include "config.h" was missing at the start of the crypt16.c module. This
6967 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6970 46. If there was a timeout during a "random" callout check, Exim treated it as
6971 a failure of the random address, and carried on sending RSET and the real
6972 address. If the delay was just some slowness somewhere, the response to the
6973 original RCPT would be taken as a response to RSET and so on, causing
6974 mayhem of various kinds.
6976 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6977 when I implemented it. It didn't allow for the fact that some option values
6978 may legitimately be negative (e.g. size_addition), and it didn't even do
6979 the right test for positive values.
6981 48. Domain names in DNS records are case-independent. Exim always looks them up
6982 in lower case. Some resolvers return domain names in exactly the case they
6983 appear in the zone file, that is, they may contain uppercase letters. Not
6984 all resolvers do this - some return always lower case. Exim was treating a
6985 change of case by a resolver as a change of domain, similar to a widening
6986 of a domain abbreviation. This triggered its re-routing code and so it was
6987 trying to route what was effectively the same domain again. This normally
6988 caused routing to fail (because the router wouldn't handle the domain
6989 twice). Now Exim checks for this case specially, and just changes the
6990 casing of the domain that it ultimately uses when it transmits the message
6993 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6996 50. If a filter generated a file delivery with a non-absolute name (possible if
6997 no home directory exists for the router), the forbid_file option was not
7000 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
7001 the existing equality matching.
7003 52. Exim was using ints instead of ino_t variables in some places where it was
7004 dealing with inode numbers.
7006 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
7007 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
7008 called TMPDIR, and if it finds it is different, it changes its value.
7010 54. The smtp_printf() function is now made available to local_scan() so
7011 additional output lines can be written before returning. There is also an
7012 smtp_fflush() function to enable the detection of a dropped connection.
7013 The variables smtp_input and smtp_batched_input are exported to
7016 55. Changed the default runtime configuration: the message "Unknown user"
7017 has been removed from the ACL, and instead placed on the localuser router,
7018 using the cannot_route_message feature. This means that any verification
7019 failures that generate their own messages won't get overridden. Similarly,
7020 the "Unrouteable address" message that was in the ACL for unverifiable
7021 relay addresses has also been removed.
7023 56. Added hosts_avoid_esmtp to the smtp transport.
7025 57. The exicyclog script was not checking for the esoteric option
7026 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7027 will work only if exicyclog is run under the appropriate euid.
7029 58. Following a discussion on the list, the rules by which Exim recognises line
7030 endings on incoming messages have been changed. The -dropcr and drop_cr
7031 options are now no-ops, retained only for backwards compatibility. The
7032 following line terminators are recognized: LF CRLF CR. However, special
7033 processing applies to CR:
7035 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7036 nor a local message in the state where . is a terminator.
7038 (ii) If a bare CR is encountered in a header line, an extra space is added
7039 after the line terminator so as not to end the header. The reasoning
7040 behind this is that bare CRs in header lines are most likely either
7041 to be mistakes, or people trying to play silly games.
7043 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7044 was being incorrectly given as 18 bytes larger than it should have been.
7045 This is a VOB (very old bug).
7047 60. This may never have affected anything current, but just in case it has:
7048 When the local host is found other than at the start of a list of hosts,
7049 the local host, those with the same MX, and any that follow, are discarded.
7050 When the list in question was part of a longer list of hosts, the following
7051 hosts (not currently being processed) were also being discarded. This no
7052 longer happens. I'm not sure if this situation could ever has previously
7055 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7058 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7059 header. This is something that is recommended in a new Internet Draft, and
7060 is something that is documented as being done by Sendmail. There are two
7061 possible values. For messages generated by the autoreply transport, Exim
7064 Auto-Submitted: auto-replied
7066 whereas for all other generated messages (e.g. bounces) it adds
7068 Auto-Submitted: auto-generated
7070 63. The "personal" condition in filters now includes a test for the
7071 Auto-Submitted: header. If it contains the string "auto-" the message it
7072 not considered personal.
7074 64. Added rcpt_include_affixes as a generic transport option.
7076 65. Added queue_only_override (default true).
7078 66. Added the syslog_duplication option.
7080 67. If what should have been the first header line of a message consisted of
7081 a space followed by a colon, Exim was mis-interpreting it as a header line.
7082 It isn't of course - it is syntactically invalid and should therefore be
7083 treated as the start of the message body. The misbehaviour could have
7084 caused a number of strange effects, including loss of data in subsequent
7085 header lines, and spool format errors.
7087 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7088 client host had authenticated. This control can now be exercised by an ACL
7089 for more flexibility.
7091 69. By default, callouts do not happen when testing with -bh. There is now a
7092 variant, -bhc, which does actually run the callout code, including
7093 consulting and updating the callout cache.
7095 70. Added support for saslauthd authentication, courtesy of Alexander
7098 71. If statvfs() failed on the spool or log directories while checking their
7099 size for availability, Exim confusingly gave the error "space shortage".
7100 Furthermore, in debugging mode it crashed with a floating point exception.
7101 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7102 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7103 serious problem, Exim now writes to the main and panic logs when this
7104 happens, with details of the failure. It then refuses to accept the
7105 incoming message, giving the message "spool directory problem" or "log
7106 directory problem" with a 421 code for SMTP messages.
7108 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7109 0, 1, and 2 exist, because some OS complain for execs without them (see
7110 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7111 descriptors. However, the code omitted to check that the open succeeded,
7112 causing mysterious errors if for some reason the permissions on /dev/null
7113 got screwed. Now Exim writes a message to the main and panic logs, and
7114 bombs out if it can't open /dev/null.
7116 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7117 interact so that it is all more flexible. It is supposed to remain
7118 backwards compatible. Also added extra_local_interfaces.
7120 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7121 to bomb out with an assertion failure - to the client this appears as a
7122 connection drop. This problem occurs in the part of the code that was taken
7123 from the Samba project. Fortunately, the assertion is in a very simple
7124 function, so I have fixed this by reproducing the function inline in the
7125 one place where it is called, and arranging for authentication to fail
7126 instead of killing the process with assert().
7128 75. The SPA client code was not working when the server requested OEM rather
7129 than Unicode encoding.
7131 76. Added code to make require_files with a specific uid setting more usable in
7132 the case where statting the file as root fails - usually a non-root-mounted
7133 NFS file system. When this happens and the failure is EACCES, Exim now
7134 forks a subprocess and does the per-uid checking as the relevant uid.
7136 77. Added process_log_path.
7138 78. If log_file_path was not explicitly set, a setting of check_log_space or
7139 check_log_inodes was ignored.
7141 79. If a space check for the spool or log partitions fails, the incident is now
7142 logged. Of course, in the latter case the data may get lost...
7144 80. Added the %p formatting code to string_format() so that it can be used to
7145 print addresses in debug_print(). Adjusted all the address printing in the
7146 debugging in store.c to use %p rather than %d.
7148 81. There was a concern that a line of code in smtp_in.c could overflow a
7149 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7150 initially expressed, the concern was not well-founded, because trailing
7151 spaces are removed early. However, if the trailing spaces were followed by
7152 a NULL, they did not get removed, so the overflow was possible. Two fixes
7155 (a) I re-wrote the offending code in a cleaner fashion.
7156 (b) If an incoming SMTP command contains a NULL character, it is rejected
7159 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7160 runs initgroups(), so that if the Exim user is in any additional groups,
7161 they will be used during message reception.
7167 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob