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
209 JH/38 DKIM: accept Ed25519 pubkeys in SubjectPublicKeyInfo-wrapped form,
210 since the IETF WG has not yet settled on that versus the original
211 "bare" representation.
217 JH/01 Rework error string handling in TLS interface so that the caller in
218 more cases is responsible for logging. This permits library-sourced
219 string to be attached to addresses during delivery, and collapses
220 pairs of long lines into single ones.
222 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
223 during configuration. Wildcards are allowed and expanded.
225 JH/02 Rework error string handling in DKIM to pass more info back to callers.
226 This permits better logging.
228 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
229 do not close it down and have the child transport start it up again on
230 the passed-on TCP connection. Instead, proxy the child (and any
231 subsequent ones) for TLS via a unix-domain socket channel. Logging is
232 affected: the continued delivery log lines do not have any DNSSEC, TLS
233 Certificate or OCSP information. TLS cipher information is still logged.
235 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
236 identical IP addresses on different listening ports. Will also affect
239 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
240 add noisy ifdef guards to special-case this sillyness.
241 Patch from Bernd Kuhls.
243 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
244 than 255 are no longer allowed.
246 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
247 Disable the session-cache too, which might reduce our load. Since we
248 currrectly use a new context for every connection, both as server and
249 client, there is no benefit for these.
250 GnuTLS appears to not support tickets server-side by default (we don't
251 call gnutls_session_ticket_enable_server()) but client side is enabled
252 by default on recent versions (3.1.3 +) unless the PFS priority string
255 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
256 <https://reproducible-builds.org/specs/source-date-epoch/>.
258 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
259 the check for any unsuccessful recipients did not notice the limit, and
260 erroneously found still-pending ones.
262 JH/08 Pipeline CHUNKING command and data together, on kernels that support
263 MSG_MORE. Only in-clear (not on TLS connections).
265 JH/09 Avoid using a temporary file during transport using dkim. Unless a
266 transport-filter is involved we can buffer the headers in memory for
267 creating the signature, and read the spool data file once for the
268 signature and again for transmission.
270 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
271 4.77 as the kernel support then wasn't solid, having issues in 64bit
272 mode. Now, it's been long enough. Add support for FreeBSD also.
274 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
275 case where the routing stage had gathered several addresses to send to
276 a host before calling the transport for the first, we previously failed
277 to close down TLS in the old transport process before passing the TCP
278 connection to the new process. The new one sent a STARTTLS command
279 which naturally failed, giving a failed delivery and bloating the retry
280 database. Investigation and fix prototype from Wolfgang Breyha.
282 JH/12 Fix check on SMTP command input synchronisation. Previously there were
283 false-negatives in the check that the sender had not preempted a response
284 or prompt from Exim (running as a server), due to that code's lack of
285 awareness of the SMTP input buffering.
287 PP/04 Add commandline_checks_require_admin option.
288 Exim drops privileges sanely, various checks such as -be aren't a
289 security problem, as long as you trust local users with access to their
290 own account. When invoked by services which pass untrusted data to
291 Exim, this might be an issue. Set this option in main configuration
292 AND make fixes to the calling application, such as using `--` to stop
295 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
296 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
297 responses to those, into a single TLS record each way (this usually means
298 a single packet). As a side issue, smtp_enforce_sync now works on TLS
301 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
302 affects you only if you're dancing at the edge of the param size limits.
303 If you are, and this message makes sense to you, then: raise the
304 configured limit or use OpenSSL 1.1. Nothing we can do for older
307 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
308 cmdline element to get the documented default one. Previously it was
311 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
312 are unsuable for non-message-specific reasons - eg. connection timeout,
313 banner-time rejection.
315 JH/16 Fix logging of delivery remote port, when specified by router, under
318 PP/06 Repair manualroute's ability to take options in any order, even if one
319 is the name of a transport.
322 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
324 JH/17 Change the list-building routines interface to use the expanding-string
325 triplet model, for better allocation and copying behaviour.
327 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
328 Previously it was constructed the first time a possibly-matching string
329 was met in the configuration file input during startup; now it is done
332 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
333 compatible one, to avoid the (poorly documented) possibility of a config
334 file in the working directory redirecting the DB files, possibly correpting
335 some existing file. CVE-2017-10140 assigned for BDB.
337 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
338 cache-hot. Previously, although the result was properly cached, the
339 initial verify call returned a defer.
341 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
342 the main verify for receipient in uncached-mode.
344 JH/22 Retire historical build files to an "unsupported" subdir. These are
345 defined as "ones for which we have no current evidence of testing".
347 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
348 if present. Previously it was ignored.
350 JH/24 Start using specified-initialisers in C structure init coding. This is
351 a C99 feature (it's 2017, so now considered safe).
353 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
354 if was a fixed-sized field and bitmask ops via macros; it is now more
357 PP/07 GitHub PR 56: Apply MariaDB build fix.
358 Patch provided by Jaroslav Škarvada.
360 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
361 during Coverity cleanups [4.87 JH/47]
362 Diagnosis and fix provided by Michael Fischer v. Mollard.
364 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
365 the right size to place the terminating semicolon on its own folded
366 line, the header hash was calculated to an incorrect value thanks to
367 the (relaxed) space the fold became.
369 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
370 and confused the parent.
372 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
373 which could crash as a result. This could lead to undeliverable messages.
375 JH/28 Logging: "next input sent too soon" now shows where input was truncated
378 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
379 matters on fast-turnover and PID-randomising systems, which were getting
380 out-of-order delivery.
382 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
383 a possibly-overlapping copy. The symptom was that "Remote host closed
384 connection in response to HELO" was logged instead of the actual 4xx
387 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
388 Previously only that bufferd was discarded, resulting in SYMTP command
391 JH/32 DKIM: when a message has multiple signatures matching an identity given
392 in dkim_verify_signers, run the dkim acl once for each. Previously only
393 one run was done. Bug 2189.
395 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
396 "panic the current process" to "deliberately defer". The panic log is
397 still written with the problem list name; the mail and reject logs now
398 get a temp-reject line for the message that was being handled, saying
399 something like "domains check lookup or other defer". The SMTP 451
400 message is still "Temporary local problem".
402 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
403 A crafted sequence of BDAT commands could result in in-use memory beeing
404 freed. CVE-2017-16943.
406 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
407 from SMTP input. Previously it was always done; now only done for DATA
408 and not BDAT commands. CVE-2017-16944.
410 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
411 to the message (such as an overlong header line). Previously this was
412 not done and we did not exit BDAT mode. Followon from the previous item
413 though a different problem.
419 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
420 than -2003 did; needs libidn2 in addition to libidn.
422 JH/02 The path option on a pipe transport is now expanded before use.
424 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
425 Patch provided by "Björn", documentation fix added too.
427 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
428 missing a wire-to-host endian conversion.
430 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
431 close after a BDAT command line could be taken as a following command,
432 giving a synch failure. Fix by only checking for synch immediately
433 before acknowledging the chunk.
435 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
436 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
437 macro. Patches provided by Josh Soref.
439 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
440 Previously we did not; the RFC seems ambiguous and VRFY is not listed
441 by IANA as a service extension. However, John Klensin suggests that we
444 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
445 the dkim code may be unix-mode line endings rather than smtp wire-format
446 CRLF, so prepend a CR to any bare LF.
448 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
449 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
451 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
452 the first were themselves being wrongly included in the feed into dkim
453 processing; with most chunk sizes in use this resulted in an incorrect
454 body hash calculated value.
456 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
457 DKIM signature block, for verification. Although advised against by
458 standards it is specifically not ruled illegal.
460 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
462 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
463 missing a body hash (the bh= tag).
465 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
466 It seems that HAProxy sends the Proxy Protocol information in clear and
467 only then does a TLS startup, so do the same.
469 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
470 TCP connections (such as for Spamd) unless the daemon successfully set
471 Fast Open mode on its listening sockets. This fixes breakage seen on
472 too-old kernels or those not configured for Fast Open, at the cost of
473 requiring both directions being enabled for TFO, and TFO never being used
474 by non-daemon-related Exim processes.
476 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
477 endings, at least on the first header line. Try to canonify any that get
478 past that check, despite the cost.
480 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
481 now limited to an arbitrary five deep, while parsing addresses with the
482 strip_excess_angle_brackets option enabled.
484 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
485 instead leave the unprompted TLS handshake in socket buffer for the
486 TLS library to consume.
488 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
490 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
492 JH/16 Drop variables when they go out of scope. Memory management drops a whole
493 region in one operation, for speed, and this leaves assigned pointers
494 dangling. Add checks run only under the testsuite which checks all
495 variables at a store-reset and panics on a dangling pointer; add code
496 explicitly nulling out all the variables discovered. Fixes one known
497 bug: a transport crash, where a dangling pointer for $sending_ip_address
498 originally assigned in a verify callout, is re-used.
500 PP/06 Drop '.' from @INC in various Perl scripts.
502 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
504 PP/08 Reduce a number of compilation warnings under clang; building with
505 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
506 should be warning-free.
508 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
510 HS/01 Fix portability problems introduced by PP/08 for platforms where
511 realloc(NULL) is not equivalent to malloc() [SunOS et al].
513 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
514 chunk. This allows us to accept broken chunked messages. We need a more
515 general solution here.
517 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
518 already-broken messages in the queue.
520 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
522 JH/19 Fix reference counting bug in routing-generated-address tracking.
528 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
529 supports it and a size is available (ie. the sending peer gave us one).
531 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
532 years of being deprecated). The replacements are the ACLs
533 acl_smtp_mime and acl_not_smtp_mime.
535 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
536 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
537 or even an in-clear connection were permitted. Now, if the host lookup
538 was dnssec and dane was requested then the host is only used if the
539 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
540 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
541 if one fails this test.
542 This means that a poorly-configured remote DNS will make it incommunicado;
543 but it protects against a DNS-interception attack on it.
545 JH/04 Bug 1810: make continued-use of an open smtp transport connection
546 non-noisy when a race steals the message being considered.
548 JH/05 If main configuration option tls_certificate is unset, generate a
549 self-signed certificate for inbound TLS connections.
551 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
552 in rewrites and routers.
554 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
555 and logged a warning sing 4.83; now they are a configuration file error.
557 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
558 (lacking @domain). Apply the same qualification processing as RCPT.
560 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
562 JH/10 Support ${sha256:} applied to a string (as well as the previous
565 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
566 a cutthrough deliver is pending, as we always want to make a connection.
567 This also avoids re-routing the message when later placing the cutthrough
568 connection after a verify cache hit.
569 Do not update it with the verify result either.
571 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
572 when routing results in more than one destination address.
574 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
575 signing (which inhibits the cutthrough capability). Previously only
576 the presence of an option was tested; now an expansion evaluating as
577 empty is permissible (obviously it should depend only on data available
578 when the cutthrough connection is made).
580 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
581 the relevant preceding SMTP command did not note the pipelining mode.
583 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
584 Previously they were not counted.
586 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
587 as one having no matching records. Previously we deferred the message
588 that needed the lookup.
590 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
591 distinguished as "(=".
593 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
594 for missing MX records. Previously it only worked for missing A records.
596 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
598 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
599 after the data-go-ahead and data-ack. Patch from Jason Betts.
601 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
602 even for a "none" policy. Patch from Tony Meyer.
604 JH/22 Fix continued use of a connection for further deliveries. If a port was
605 specified by a router, it must also match for the delivery to be
608 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
609 When one of the recipients of a message was unsuitable for the connection
610 (has no matching addresses), we lost track of needing to mark it
611 deferred. As a result mail would be lost.
613 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
615 JH/25 Decoding ACL controls is now done using a binary search; the source code
616 takes up less space and should be simpler to maintain. Merge the ACL
617 condition decode tables also, with similar effect.
619 JH/26 Fix problem with one_time used on a redirect router which returned the
620 parent address unchanged. A retry would see the parent address marked as
621 delivered, so not attempt the (identical) child. As a result mail would
624 JH/27 Fix a possible security hole, wherein a process operating with the Exim
625 UID can gain a root shell. Credit to http://www.halfdog.net/ for
626 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
629 JH/28 Enable {spool,log} filesystem space and inode checks as default.
630 Main config options check_{log,spool}_{inodes,space} are now
631 100 inodes, 10MB unless set otherwise in the configuration.
633 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
634 Previously it only applied to the main-section connection policy
637 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
639 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
640 by me. Added RFC7919 DH primes as an alternative.
642 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
643 are not in the system include path.
645 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
646 GnuTLS, when a session startup failed (eg because the client disconnected)
647 Exim did stdio operations after fclose. This was exposed by a recent
648 change which nulled out the file handle after the fclose.
650 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
651 signed directly by the cert-signing cert, rather than an intermediate
652 OCSP-signing cert. This is the model used by LetsEncrypt.
654 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
656 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
657 an incoming connection.
659 HS/02 Bug 1802: Do not half-close the connection after sending a request
662 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
663 fallback to "prime256v1".
665 JH/34 SECURITY: Use proper copy of DATA command in error message.
666 Could leak key material. Remotely exploitable. CVE-2016-9963.
672 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
673 and 3.4.4 - once the server is enabled to respond to an OCSP request
674 it does even when not requested, resulting in a stapling non-aware
675 client dropping the TLS connection.
677 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
678 support variable-length bit vectors. No functional change.
680 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
681 The I= interface field on outgoing lines is now after the H= remote
682 host field, same as incoming lines. There is a separate
683 outgoing_interface log selector which allows you to disable the
686 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
687 If not running log_selector +smtp_connection the mainlog would be held
688 open indefinitely after a "too many connections" event, including to a
689 deleted file after a log rotate. Leave the per net connection logging
690 leaving it open for efficiency as that will be quickly detected by the
691 check on the next write.
693 HS/01 Bug 1671: Fix post transport crash.
694 Processing the wait-<transport> messages could crash the delivery
695 process if the message IDs didn't exist for some reason. When
696 using 'split_spool_directory=yes' the construction of the spool
697 file name failed already, exposing the same netto behaviour.
699 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
700 mime_regex ACL conditions.
702 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
703 to DSN fail messages (bounces): remote IP, remote greeting, remote response
704 to HELO, local diagnostic string.
706 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
707 log line to debug. Even when configured with a tls authenticator many
708 client connections are expected to not authenticate in this way, so
709 an authenticate fail is not an error.
711 HS/02 Add the Exim version string to the process info. This way exiwhat
712 gives some more detail about the running daemon.
714 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
715 matter for fast-change records such as DNSBLs.
717 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
718 retry record, even if constant. There may be multiple transports with
719 different interface settings and the retry behaviour needs to be kept
722 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
724 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
726 JH/10 Bug 840: fix log_defer_output option of pipe transport
728 JH/11 Bug 830: use same host for all RCPTS of a message, even under
729 hosts_randomize. This matters a lot when combined with mua_wrapper.
731 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
732 ${quote_pgsql:<string>} operator.
734 JH/13 Bug 1708: avoid misaligned access in cached lookup.
736 JH/14 Change header file name for freeradius-client. Relevant if compiling
737 with Radius support; from the Gentoo tree and checked under Fedora.
739 JH/15 Bug 1712: Introduce $prdr_requested flag variable
741 JH/16 Bug 1714: Permit an empty string as expansion result for transport
742 option transport_filter, meaning no filtering.
744 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
746 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
747 defaults to "*" (all hosts). The variable is now available when not built
748 with TLS, default unset, mainly to enable keeping the testsuite sane.
749 If a server certificate is not supplied (via tls_certificate) an error is
750 logged, and clients will find TLS connections fail on startup. Presumably
751 they will retry in-clear.
752 Packagers of Exim are strongly encouraged to create a server certificate
753 at installation time.
755 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
756 with the $config_file variable.
758 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
759 in transport context, after the attempt, and per-recipient. The latter type
760 is per host attempted. The event data is the error message, and the errno
761 information encodes the lookup type (A vs. MX) used for the (first) host,
762 and the trailing two digits of the smtp 4xx response.
764 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
765 to write to mainlog (or rejectlog, paniclog) in the window between file
766 creation and permissions/ownership being changed. Particularly affects
767 installations where exicyclog is run as root, rather than exim user;
768 result is that the running daemon panics and dies.
770 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
772 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
773 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
774 "pri" and "weight". Note that the previous implicit priority given by the
775 list order is no longer honoured.
777 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
780 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
781 by defining SUPPORT_SOCKS.
783 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
784 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
785 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
786 variables are renamed to proxy_{local,external}_{address,port}.
788 JH/27 Move Internationalisation support from Experimental to mainline, enabled
789 for a build by defining SUPPORT_I18N
791 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
792 of the query string, and make ${quote_redis:} do that quoting.
794 JH/29 Move Events support from Experimental to mainline, enabled by default
795 and removable for a build by defining DISABLE_EVENT.
797 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
799 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
800 cached by the daemon.
802 JH/32 Move Redis support from Experimental to mainline, enabled for a build
803 by defining LOOKUP_REDIS. The libhiredis library is required.
805 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
806 keys are given for lookup.
808 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
809 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
810 only supported when built with TLS support. The PolarSSL SHA routines
811 are still used when the TLS library is too old for convenient support.
813 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
814 openssl_options), for security. OpenSSL forces this from version 1.1.0
815 server-side so match that on older versions.
817 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
818 allocation for $value could be released as the expansion processing
819 concluded, but leaving the global pointer active for it.
821 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
822 and to use the domains and local_parts ACL conditions.
824 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
825 incorrectly not doubled on cutthrough transmission, hence seen as a
826 body-termination at the receiving system - resulting in truncated mails.
827 Commonly the sender saw a TCP-level error, and retransmitted the message
828 via the normal store-and-forward channel. This could result in duplicates
829 received - but deduplicating mailstores were liable to retain only the
830 initial truncated version.
832 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
834 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
836 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
837 we're in there, support oversigning also; bug 1309.
839 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
841 HS/04 Add support for keep_environment and add_environment options.
843 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
844 either intentional arithmetic overflow during PRNG, or testing config-
847 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
848 delivery resulted in actual delivery. Cancel cutthrough before DATA
851 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
852 rejects a recipient: pass the reject to the originator.
854 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
855 Many were false-positives and ignorable, but it's worth fixing the
858 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
859 for the new environment-manipulation done at startup. Move the routines
860 from being local to tls.c to being global via the os.c file.
862 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
863 an extract embedded as result-arg for a map, the first arg for extract
864 is unavailable so we cannot tell if this is a numbered or keyed
865 extraction. Accept either.
871 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
874 JH/02 The smtp transport option "multi_domain" is now expanded.
876 JH/03 The smtp transport now requests PRDR by default, if the server offers
879 JH/04 Certificate name checking on server certificates, when exim is a client,
880 is now done by default. The transport option tls_verify_cert_hostnames
881 can be used to disable this per-host. The build option
882 EXPERIMENTAL_CERTNAMES is withdrawn.
884 JH/05 The value of the tls_verify_certificates smtp transport and main options
885 default to the word "system" to access the system default CA bundle.
886 For GnuTLS, only version 3.0.20 or later.
888 JH/06 Verification of the server certificate for a TLS connection is now tried
889 (but not required) by default. The verification status is now logged by
890 default, for both outbound TLS and client-certificate supplying inbound
893 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
896 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
897 Status Notification (bounce) messages are now MIME format per RFC 3464.
898 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
899 under the control of the dsn_advertise_hosts option, and routers may
900 have a dsn_lasthop option.
902 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
903 default, modifiable by a malware= option. The list separator for
904 the options can now be changed in the usual way. Bug 68.
906 JH/10 The smtp_receive_timeout main option is now expanded before use.
908 JH/11 The incoming_interface log option now also enables logging of the
909 local interface on delivery outgoing connections.
911 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
912 if the interface and destination host and port all match.
914 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
917 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
918 Patch from Andrew Lewis.
920 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
921 now supports optional time-restrictions, weighting, and priority
922 modifiers per server. Patch originally by <rommer@active.by>.
924 JH/16 The spamd_address main option now supports a mixed list of local
925 and remote servers. Remote servers can be IPv6 addresses, and
926 specify a port-range.
928 JH/17 Bug 68: The spamd_address main option now supports an optional
929 timeout value per server.
931 JH/18 Bug 1581: Router and transport options headers_add/remove can
932 now have the list separator specified.
934 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
937 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
940 JH/21 Support for the A6 type of dns record is withdrawn.
942 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
943 rather than the verbs used.
945 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
946 from 255 to 1024 chars.
948 JH/24 Verification callouts now attempt to use TLS by default.
950 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
951 are generic router options now. The defaults didn't change.
953 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
954 Original patch from Alexander Shikoff, worked over by JH.
956 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
957 files if ZCAT_COMMAND is not executable.
959 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
961 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
963 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
964 Normally benign, it bites when the pair was led to by a CNAME;
965 modern usage is to not canonicalize the domain to a CNAME target
966 (and we were inconsistent anyway for A-only vs AAAA+A).
968 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
970 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
971 when evaluating $sender_host_dnssec.
973 JH/31 Check the HELO verification lookup for DNSSEC, adding new
974 $sender_helo_dnssec variable.
976 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
978 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
980 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
982 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
983 documented as working, but never had. Support all but $spam_report.
985 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
986 added for tls authenticator.
988 HS/03 Add perl_taintmode main config option
994 TL/01 When running the test suite, the README says that variables such as
995 no_msglog_check are global and can be placed anywhere in a specific
996 test's script, however it was observed that placement needed to be near
997 the beginning for it to behave that way. Changed the runtest perl
998 script to read through the entire script once to detect and set these
999 variables, reset to the beginning of the script, and then run through
1000 the script parsing/test process like normal.
1002 TL/02 The BSD's have an arc4random API. One of the functions to induce
1003 adding randomness was arc4random_stir(), but it has been removed in
1004 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
1005 function when detected.
1007 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
1008 cause callback expansion.
1010 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
1011 syntax errors in an expansion can be treated as a string instead of
1012 logging or causing an error, due to the internal use of bool_lax
1013 instead of bool when processing it.
1015 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
1016 server certificates when making smtp deliveries.
1018 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
1020 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
1022 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
1024 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
1025 Merged patch from Sebastian Wiedenroth.
1027 JH/05 Fix results-pipe from transport process. Several recipients, combined
1028 with certificate use, exposed issues where response data items split
1029 over buffer boundaries were not parsed properly. This eventually
1030 resulted in duplicates being sent. This issue only became common enough
1031 to notice due to the introduction of connection certificate information,
1032 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1034 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1035 size buffer was used, resulting in syntax errors when an expansion
1038 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1039 version 3.3.6 or later.
1041 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1042 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1043 both become "event_action", the variables become $event_name, $event_data
1044 and $event_defer_errno. There is a new variable $verify_mode, usable in
1045 routers, transports and related events. The tls:cert event is now also
1046 raised for inbound connections, if the main configuration event_action
1049 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1050 early OCSP support, but no stapling (appears to be less than 1.0.0).
1052 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1053 server certificate names available under the smtp transport option
1054 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1057 JH/10 Time-related extraction expansions from certificates now use the main
1058 option "timezone" setting for output formatting, and are consistent
1059 between OpenSSL and GnuTLS compilations. Bug 1541.
1061 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1062 encoded parameter in the incoming message. Bug 1558.
1064 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1065 include certificate info, eximon was claiming there were spoolfile
1068 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1070 JH/14 Log delivery-related information more consistently, using the sequence
1071 "H=<name> [<ip>]" wherever possible.
1073 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1074 are problematic for Debian distribution, omit them from the release
1077 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1079 JH/16 Fix string representation of time values on 64bit time_t architectures.
1082 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1083 output list separator was used.
1088 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1089 checkers that were complaining about end of non-void function with no
1092 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1093 This was a regression introduced in 4.83 by another bugfix.
1095 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1097 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1098 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1104 TF/01 Correctly close the server side of TLS when forking for delivery.
1106 When a message was received over SMTP with TLS, Exim failed to clear up
1107 the incoming connection properly after forking off the child process to
1108 deliver the message. In some situations the subsequent outgoing
1109 delivery connection happened to have the same fd number as the incoming
1110 connection previously had. Exim would try to use TLS and fail, logging
1111 a "Bad file descriptor" error.
1113 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1114 utilities have not been installed.
1116 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1117 temporary space as the ACL may create new global variables.
1119 TL/01 LDAP support uses per connection or global context settings, depending
1120 upon the detected version of the libraries at build time.
1122 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1123 to extract and use the src ip:port in logging and expansions as if it
1124 were a direct connection from the outside internet. PPv2 support was
1125 updated based on HAProxy spec change in May 2014.
1127 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1129 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1130 Properly escape header and check for NULL return.
1132 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1135 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1137 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1138 characters in header names, implemented as a verify condition.
1139 Contributed by Michael Fischer v. Mollard.
1141 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1142 results permerror and temperror. Previous values are deprecated but
1143 still accepted. In a future release, err_perm and err_temp will be
1144 completely removed, which will be a backward incompatibility if the
1145 ACL tests for either of these two old results. Patch contributed by
1146 user bes-internal on the mailing list.
1148 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1150 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1151 selectors, in both main and reject logs.
1153 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1156 JH/07 Add malware type "sock" for talking to simple daemon.
1158 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1160 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1161 routers/transports under cutthrough routing.
1163 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1164 numbers. Touch up "bool" conditional to keep the same definition.
1166 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1168 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1170 JH/12 Expand items in router/transport headers_add or headers_remove lists
1171 individually rather than the list as a whole. Bug 1452.
1173 Required for reasonable handling of multiple headers_ options when
1174 they may be empty; requires that headers_remove items with embedded
1175 colons must have them doubled (or the list-separator changed).
1177 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1178 view the policy declared in the DMARC record. Currently, $dmarc_status
1179 is a combined value of both the record presence and the result of the
1182 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1184 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1185 dnslookup router and the smtp transport (applying to the forward
1188 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1189 of ldap servers used for a specific lookup. Patch provided by Heiko
1192 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1193 New variable $lookup_dnssec_authenticated for observability.
1195 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1196 Patch submitted by Lars Timman.
1198 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1200 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1201 Requires trusted mode and valid format message id, aborts otherwise.
1202 Patch contributed by Heiko Schlichting.
1204 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1205 certextract with support for various fields. Bug 1358.
1207 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1208 is requested by default, modifiable by smtp transport option
1211 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1212 operate on certificate variables to give certificate fingerprints
1213 Also new ${sha256:cert_variable}.
1215 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1217 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1218 Christian Aistleitner.
1220 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1222 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1223 file. Patch from Wolfgang Breyha.
1225 JH/25 Expand the coverage of the delivery $host and $host_address to
1226 client authenticators run in verify callout. Bug 1476.
1228 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1229 align with daemon_smtp_ports. Bug 72.
1231 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1232 support and error reporting did not work properly.
1234 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1235 and is readable. Patch from Andrew Colin Kissa.
1237 TL/14 Enhance documentation of ${run expansion and how it parses the
1238 commandline after expansion, particularly in the case when an
1239 unquoted variable expansion results in an empty value.
1241 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1243 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1244 by renaming away from it. Addresses GH issue 15, reported by
1247 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1249 TL/15 SECURITY: prevent double expansion in math comparison functions
1250 (can expand unsanitized data). Not remotely exploitable.
1257 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1259 PP/02 Make -n do something, by making it not do something.
1260 When combined with -bP, the name of an option is not output.
1262 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1265 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1266 $sender_host_name and config options to manage this, and basic check
1269 PP/05 DSCP support for outbound connections and control modifier for inbound.
1271 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1272 (Only plugin which currently uses this is kerberos4, which nobody should
1273 be using, but we should make it available and other future plugins might
1274 conceivably use it, even though it would break NAT; stuff *should* be
1275 using channel bindings instead).
1277 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1278 name; added for Sendmail compatibility; requires admin caller.
1279 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1280 just ignore it); requires trusted caller.
1281 Also parse but ignore: -Ac -Am -X<logfile>
1284 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1286 TL/02 Add +smtp_confirmation as a default logging option.
1288 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1289 Patch by Magnus Holmgren from 2007-02-20.
1291 TL/04 Bugzilla 1281 - Spec typo.
1292 Bugzilla 1283 - Spec typo.
1293 Bugzilla 1290 - Spec grammar fixes.
1295 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1297 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1299 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1300 mailing list by by Dmitry Isaikin.
1302 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1304 JH/02 Support "G" suffix to numbers in ${if comparisons.
1306 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1308 NM/01 Bugzilla 1197 - Spec typo
1309 Bugzilla 1196 - Spec examples corrections
1311 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1313 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1314 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1317 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1318 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1320 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1321 "acl {{name}{arg}...}", and optional args on acl condition
1324 JH/05 Permit multiple router/transport headers_add/remove lines.
1326 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1328 JH/07 Avoid using a waiting database for a single-message-only transport.
1329 Performance patch from Paul Fisher. Bugzilla 1262.
1331 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1334 JH/09 Add $headers_added variable, with content from use of ACL modifier
1335 add_header (but not yet added to the message). Bugzilla 199.
1337 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1338 Pulled from Bugzilla 817 by Wolfgang Breyha.
1340 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1342 (nb: this is the same fix as in Exim 4.80.1)
1344 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1347 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1350 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1352 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1353 advertise SMTP AUTH mechanism to us, instead of a generic
1354 protocol violation error. Also, make Exim more robust to bad
1355 data from the Dovecot auth socket.
1357 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1359 When a queue runner is handling a message, Exim first routes the
1360 recipient addresses, during which it prunes them based on the retry
1361 hints database. After that it attempts to deliver the message to
1362 any remaining recipients. It then updates the hints database using
1365 So if a recipient address works intermittently, it can get repeatedly
1366 deferred at routing time. The retry hints record remains fresh so the
1367 address never reaches the final cutoff time.
1369 This is a fairly common occurrence when a user is bumping up against
1370 their storage quota. Exim had some logic in its local delivery code
1371 to deal with this. However it did not apply to per-recipient defers
1372 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1374 This change adds a proper retry rule check during routing so that the
1375 final cutoff time is checked against the message's age. We only do
1376 this check if there is an address retry record and there is not a
1377 domain retry record; this implies that previous attempts to handle
1378 the address had the retry_use_local_parts option turned on. We use
1379 this as an approximation for the destination being like a local
1380 delivery, as in LMTP.
1382 I suspect this new check makes the old local delivery cutoff check
1383 redundant, but I have not verified this so I left the code in place.
1385 TF/02 Correct gecos expansion when From: is a prefix of the username.
1387 Test 0254 submits a message to Exim with the header
1391 When I ran the test suite under the user fanf2, Exim expanded
1392 the header to contain my full name, whereas it should have added
1393 a Resent-Sender: header. It erroneously treats any prefix of the
1394 username as equal to the username.
1396 This change corrects that bug.
1398 GF/01 DCC debug and logging tidyup
1399 Error conditions log to paniclog rather than rejectlog.
1400 Debug lines prefixed by "DCC: " to remove any ambiguity.
1402 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1404 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1405 Bug spotted by Jeremy Harris; was flawed since initial commit.
1406 Would have resulted in OCSP responses post-SNI triggering an Exim
1407 NULL dereference and crash.
1409 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1411 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1412 Bug detection, analysis and fix by Samuel Thibault.
1413 Bugzilla 1331, Debian bug #698092.
1415 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1417 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1418 Server implementation by Todd Lyons, client by JH.
1419 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1420 config variable "prdr_enable" controls whether the server
1421 advertises the facility. If the client requests PRDR a new
1422 acl_data_smtp_prdr ACL is called once for each recipient, after
1423 the body content is received and before the acl_smtp_data ACL.
1424 The client is controlled by both of: a hosts_try_prdr option
1425 on the smtp transport, and the server advertisement.
1426 Default client logging of deliveries and rejections involving
1427 PRDR are flagged with the string "PRDR".
1429 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1430 fclose(). Diagnosis by Todd Lyons.
1432 PP/17 Update configure.default to handle IPv6 localhost better.
1433 Patch by Alain Williams (plus minor tweaks).
1436 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1437 This is now consistent with GnuTLS, and is now documented: the
1438 previous undocumented portable approach to treating the option as
1439 unset was to force an expansion failure. That still works, and
1440 an empty string is now equivalent.
1442 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1443 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1444 not performing validation itself.
1446 PP/20 Added force_command boolean option to pipe transport.
1447 Patch from Nick Koston, of cPanel Inc.
1449 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1452 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1454 PP/21 Fix eximon continuous updating with timestamped log-files.
1455 Broken in a format-string cleanup in 4.80, missed when I repaired the
1456 other false fix of the same issue.
1457 Report and fix from Heiko Schlichting.
1460 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1461 Report from Prashanth Katuri.
1463 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1464 It's SecureTransport, so affects any MacOS clients which use the
1465 system-integrated TLS libraries, including email clients.
1467 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1468 using a MIME ACL for non-SMTP local injection.
1469 Report and assistance in diagnosis by Warren Baker.
1471 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1473 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1475 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1476 last id that failed so it may be referenced in subsequent ACL's.
1478 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1481 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1482 ldap library initialization, allowing self-signed CA's to be
1483 used. Also properly sets require_cert option later in code by
1484 using NULL (global ldap config) instead of ldap handle (per
1485 session). Bug diagnosis and testing by alxgomz.
1487 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1488 the src/util/ subdirectory.
1490 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1491 renamed to Transport Post Delivery Action by Jeremy Harris, as
1494 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1495 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1496 redis_servers = needs to be configured which will be used by the redis
1497 lookup. Patch from Warren Baker, of The Packet Hub.
1499 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1501 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1502 hostname or reverse DNS when processing a host list. Used suggestions
1503 from multiple comments on this bug.
1505 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1507 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1508 Missed a few lines, added it to make the runtest require no keyboard
1511 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1512 contains upper case chars. Make router use caseful_local_part.
1514 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1515 support when GnuTLS has been built with p11-kit.
1521 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1523 This, or similar/improved, will also be change PP/11 of 4.82.
1529 PP/01 Handle short writes when writing local log-files.
1530 In practice, only affects FreeBSD (8 onwards).
1531 Bugzilla 1053, with thanks to Dmitry Isaikin.
1533 NM/01 Bugzilla 949 - Documentation tweak
1535 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1538 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1540 PP/02 Implemented gsasl authenticator.
1542 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1544 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1545 `pkg-config foo` for cflags/libs.
1547 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1548 with rest of GSASL and with heimdal_gssapi.
1550 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1551 `pkg-config foo` for cflags/libs for the TLS implementation.
1553 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1554 properties get this fed in as external SSF. A number of robustness
1555 and debugging improvements to the cyrus_sasl authenticator.
1557 PP/08 cyrus_sasl server now expands the server_realm option.
1559 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1560 Patch by Jeremy Harris.
1562 PP/10 Added dbmjz lookup type.
1564 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1566 PP/12 MAIL args handles TAB as well as SP, for better interop with
1567 non-compliant senders.
1568 Analysis and variant patch by Todd Lyons.
1570 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1571 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1572 Patch from Dirk Mueller <dmueller@suse.com>
1574 PP/13 tls_peerdn now print-escaped for spool files.
1575 Observed some $tls_peerdn in wild which contained \n, which resulted
1576 in spool file corruption.
1578 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1579 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1580 or write after TLS renegotiation, which otherwise led to messages
1583 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1584 as a tracking header (ie: a signed header comes before the signature).
1585 Patch from Wolfgang Breyha.
1587 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1588 comma-sep list; embedded commas doubled.
1590 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1592 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1594 Report and patch from Dmitry Banschikov.
1596 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1597 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1598 failures. We appear to now support TLS1.1+ with Exim.
1600 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1601 lets Exim select keys and certificates based upon TLS SNI from client.
1602 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1603 before an outbound SMTP session. New log_selector, +tls_sni.
1605 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1606 NULL dereference. Report and patch from Alun Jones.
1608 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1609 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1610 Not seeing resolver debug output on NetBSD, but suspect this is a
1611 resolver implementation change.
1613 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1614 Left warnings. Added "eximon gdb" invocation mode.
1616 PP/21 Defaulting "accept_8bitmime" to true, not false.
1618 PP/22 Added -bw for inetd wait mode support.
1620 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1621 locate the relevant includes and libraries. Made this the default.
1623 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1624 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1626 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1627 This may cause build issues on older platforms.
1629 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1630 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1631 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1632 Added SNI support via GnuTLS too.
1633 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1635 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1637 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1638 Applied second patch from Janne, implementing suggestion to default
1639 multiple-strings-in-record handling to match SPF spec.
1641 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1643 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1644 read-only, out of scope).
1645 Patch from Wolfgang Breyha, report from Stuart Northfield.
1647 PP/29 Fix three issues highlighted by clang analyser static analysis.
1648 Only crash-plausible issue would require the Cambridge-specific
1649 iplookup router and a misconfiguration.
1650 Report from Marcin Mirosław.
1652 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1654 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1655 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1656 As part of this, removing so much warning spew let me fix some minor
1657 real issues in debug logging.
1659 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1660 assignment on my part. Fixed.
1662 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1663 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1664 Janne Snabb (who went above and beyond: thank you).
1666 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1667 string otherwise requires a connection and a bunch more work and it's
1668 relatively easy to get wrong. Should also expose TLS library linkage
1671 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1672 64-bit ${eval} (JH/03).
1674 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1675 GNU libc to support some of the 64-bit stuff, should not lead to
1676 conflicts. Defined before os.h is pulled in, so if a given platform
1677 needs to override this, it can.
1679 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1680 protection layer was required, which is not implemented.
1681 Bugzilla 1254, patch from Wolfgang Breyha.
1683 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1684 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1685 tls_dhparam take prime identifiers. Also unbreak combination of
1686 OpenSSL+DH_params+TLSSNI.
1688 PP/39 Disable SSLv2 by default in OpenSSL support.
1694 PP/01 Solaris build fix for Oracle's LDAP libraries.
1695 Bugzilla 1109, patch from Stephen Usher.
1697 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1699 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1702 TF/02 Fix a couple more cases where we did not log the error message
1703 when unlink() failed. See also change 4.74-TF/03.
1705 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1706 lock up or crash if it happened to be inside a call to libc when it
1707 got a SIGUSR1 from exiwhat.
1709 The SIGUSR1 handler appends the current process status to the process
1710 log which is later printed by exiwhat. It used to use the general
1711 purpose logging code to do this, but several functions it calls are
1712 not safe for signals.
1714 The new output code in the SIGUSR1 handler is specific to the process
1715 log, and simple enough that it's easy to inspect for signal safety.
1716 Removing some special cases also simplifies the general logging code.
1717 Removing the spurious timestamps from the process log simplifies
1720 TF/04 Improved ratelimit ACL condition.
1722 The /noupdate option has been deprecated in favour of /readonly which
1723 has clearer semantics. The /leaky, /strict, and /readonly update modes
1724 are mutually exclusive. The update mode is no longer included in the
1725 database key; it just determines when the database is updated. (This
1726 means that when you upgrade Exim will forget old rate measurements.)
1728 Exim now checks that the per_* options are used with an update mode that
1729 makes sense for the current ACL. For example, when Exim is processing a
1730 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1731 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1732 must specify per_mail/readonly. If you omit the update mode it defaults to
1733 /leaky where that makes sense (as before) or /readonly where required.
1735 The /noupdate option is now undocumented but still supported for
1736 backwards compatibility. It is equivalent to /readonly except that in
1737 ACLs where /readonly is required you may specify /leaky/noupdate or
1738 /strict/noupdate which are treated the same as /readonly.
1740 A useful new feature is the /count= option. This is a generalization
1741 of the per_byte option, so that you can measure the throughput of other
1742 aggregate values. For example, the per_byte option is now equivalent
1743 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1745 The per_rcpt option has been generalized using the /count= mechanism
1746 (though it's more complicated than the per_byte equivalence). When it is
1747 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1748 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1749 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1750 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1751 Note that using per_rcpt with a non-readonly update mode in more than
1752 one ACL will cause the recipients to be double-counted. (The per_mail
1753 and per_byte options don't have this problem.)
1755 The handling of very low rates has changed slightly. If the computed rate
1756 is less than the event's count (usually one) then this event is the first
1757 after a long gap. In this case the rate is set to the same as this event's
1758 count, so that the first message of a spam run is counted properly.
1760 The major new feature is a mechanism for counting the rate of unique
1761 events. The new per_addr option counts the number of different
1762 recipients that someone has sent messages to in the last time period. It
1763 behaves like per_rcpt if all the recipient addresses are different, but
1764 duplicate recipient addresses do not increase the measured rate. Like
1765 the /count= option this is a general mechanism, so the per_addr option
1766 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1767 example, measure the rate that a client uses different sender addresses
1768 with the options per_mail/unique=$sender_address. There are further
1769 details in the main documentation.
1771 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1773 TF/06 Removed a few PCRE remnants.
1775 TF/07 Automatically extract Exim's version number from tags in the git
1776 repository when doing development or release builds.
1778 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1779 Bugzilla 879. Patch from Paul Fisher.
1781 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1782 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1785 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1787 PP/05 Variable $av_failed, true if the AV scanner deferred.
1788 Bugzilla 1078. Patch from John Horne.
1790 PP/06 Stop make process more reliably on build failure.
1791 Bugzilla 1087. Patch from Heiko Schlittermann.
1793 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1794 Bugzilla 1089. Patch from Heiko Schlittermann.
1796 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1797 Bugzilla 1131. Patch from Holger Weiß.
1799 PP/09 Handle IPv6 addresses with SPF.
1800 Bugzilla 860. Patch from Wolfgang Breyha.
1802 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1804 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1807 PP/11 match_* no longer expand right-hand-side by default.
1808 New compile-time build option, EXPAND_LISTMATCH_RHS.
1809 New expansion conditions, "inlist", "inlisti".
1811 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1813 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1815 PP/14 fix log_write() format string regression from TF/03.
1816 Bugzilla 1152. Patch from Dmitry Isaikin.
1822 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1824 PP/02 Harmonised TLS library version reporting; only show if debugging.
1825 Layout now matches that introduced for other libraries in 4.74 PP/03.
1827 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1829 PP/04 New "dns_use_edns0" global option.
1831 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1834 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1835 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1837 TK/01 Updated PolarSSL code to 0.14.2.
1838 Bugzilla 1097. Patch from Andreas Metzler.
1840 PP/07 Catch divide-by-zero in ${eval:...}.
1841 Fixes bugzilla 1102.
1843 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1846 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1847 format-string attack -- SECURITY: remote arbitrary code execution.
1849 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1850 time unintentionally subject to list matching rules, letting the header
1851 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1852 arbitrary string expansion). This allowed for information disclosure.
1854 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1855 INT_MIN/-1 -- value coerced to INT_MAX.
1861 NM/01 Workaround for PCRE version dependency in version reporting
1864 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1865 This fixes portability to compilers other than gcc, notably
1866 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1868 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1869 makefiles for portability to HP-UX and POSIX correctness.
1871 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1872 Also via indented variable definition in the Makefile.
1873 (Debugging by Oliver Heesakkers).
1875 PP/02 Restore caching of spamd results with expanded spamd_address.
1876 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1878 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1879 Improves build reliability. Fix from: Frank Elsner
1881 NM/02 Fix wide character breakage in the rfc2047 coding
1882 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1884 NM/03 Allow underscore in dnslist lookups
1885 Fixes bug 1026. Patch from Graeme Fowler
1887 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1888 Code patches from Adam Ciarcinski of NetBSD.
1890 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1893 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1894 is logged, to avoid truncation. Patch from John Horne.
1896 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1897 Patch from Jakob Hirsch.
1899 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1900 SQL string expansion failure details.
1901 Patch from Andrey Oktyabrski.
1903 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1904 Patch from Simon Arlott.
1906 PP/09 New lookups functionality failed to compile on old gcc which rejects
1907 extern declarations in function scope.
1908 Patch from Oliver Fleischmann
1910 PP/10 Use sig_atomic_t for flags set from signal handlers.
1911 Check getgroups() return and improve debugging.
1912 Fixed developed for diagnosis in bug 927 (which turned out to be
1915 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1916 Patch from Mark Zealey.
1918 PP/12 Bugzilla 1056: Improved spamd server selection.
1919 Patch from Mark Zealey.
1921 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1922 Based on patch from Heiko Schlittermann.
1924 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1925 Patch from Uwe Doering, sign-off by Michael Haardt.
1927 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1928 variable declaration deep within a block. Bug and patch from
1931 PP/15 lookups-Makefile IRIX compatibility coercion.
1933 PP/16 Make DISABLE_DKIM build knob functional.
1935 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1936 Patch by Simon Arlott
1938 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1939 variable argument macros. Our copy now differs from upstream.
1945 TF/01 Failure to get a lock on a hints database can have serious
1946 consequences so log it to the panic log.
1948 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1949 controlled using the smtp_confirmation log selector.
1951 TF/03 Include the error message when we fail to unlink a spool file.
1953 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1954 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1955 for maintaining out-of-tree patches for some time.
1957 PP/01 Bugzilla 139: Documentation and portability issues.
1958 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1959 Handle per-OS dynamic-module compilation flags.
1961 PP/02 Let /dev/null have normal permissions.
1962 The 4.73 fixes were a little too stringent and complained about the
1963 permissions on /dev/null. Exempt it from some checks.
1964 Reported by Andreas M. Kirchwitz.
1966 PP/03 Report version information for many libraries, including
1967 Exim version information for dynamically loaded libraries. Created
1968 version.h, now support a version extension string for distributors
1969 who patch heavily. Dynamic module ABI change.
1971 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1972 privilege escalation vulnerability whereby the Exim run-time user
1973 can cause root to append content of the attacker's choosing to
1976 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1979 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1980 If dropping privileges for untrusted macros, we disabled normal logging
1981 on the basis that it would fail; for the Exim run-time user, this is not
1982 the case, and it resulted in successful deliveries going unlogged.
1983 Fixed. Reported by Andreas Metzler.
1989 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1990 only prepend for the Resent-* case. Fixes regression introduced in
1991 Exim 4.70 by NM/22 for Bugzilla 607.
1993 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1994 increasing numbers of administrators be bitten by this.
1996 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1998 PP/03 Bugzilla 994: added openssl_options main configuration option.
2000 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
2002 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
2004 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
2006 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
2007 without a peer certificate, leading to a segfault because of an
2008 assumption that peers always have certificates. Be a little more
2009 paranoid. Problem reported by Martin Tscholak.
2011 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
2012 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
2013 NB: ClamAV planning to remove STREAM in "middle of 2010".
2014 CL also introduces -bmalware, various -d+acl logging additions and
2015 more caution in buffer sizes.
2017 PP/09 Implemented reverse_ip expansion operator.
2019 PP/10 Bugzilla 937: provide a "debug" ACL control.
2021 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
2023 PP/12 Bugzilla 973: Implement --version.
2025 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2027 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2029 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2031 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2032 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2033 ignore trailing whitespace.
2035 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2037 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2040 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2041 Notification from Dr Andrew Aitchison.
2043 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2044 ExtendedDetectionInfo response format.
2045 Notification from John Horne.
2047 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2050 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2051 XSL and documented dependency on system catalogs, with examples of how
2054 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2057 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2058 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2059 configuration file which is writeable by the Exim user or group.
2061 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2062 of configuration files to cover files specified with the -C option if
2063 they are going to be used with root privileges, not just the default
2066 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2067 option (effectively making it always true).
2069 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2070 files to be used while preserving root privileges.
2072 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2073 that rogue child processes cannot use them.
2075 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2076 run-time user, instead of root.
2078 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2079 Exim run-time user without dropping privileges.
2081 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2082 result string, instead of calling string_vformat() twice with the same
2085 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2086 for other users. Others should always drop root privileges if they use
2087 -C on the command line, even for a whitelisted configure file.
2089 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2091 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2097 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2098 $data_path, and $header_path variables; fixed documentation bugs and
2101 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2102 exipick to access non-standard spools, including the "frozen" queue
2105 NM/01 Bugzilla 965: Support mysql stored procedures.
2106 Patch from Alain Williams
2108 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2110 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2111 Patch from Andreas Metzler
2113 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2114 Patch from Kirill Miazine
2116 NM/05 Bugzilla 671: Added umask to procmail example.
2118 JJ/03 installed exipick 20100323.0, fixing doc bug
2120 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2121 directory. Notification and patch from Dan Rosenberg.
2123 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2125 TK/02 Improve log output when DKIM signing operation fails.
2127 MH/01 Treat the transport option dkim_domain as a colon separated
2128 list, not as a single string, and sign the message with each element,
2129 omitting multiple occurences of the same signer.
2131 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2132 Bugzilla 985, 986. Patch by Simon Arlott
2134 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2135 Patch by Simon Arlott
2137 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2138 MBX locking. Notification from Dan Rosenberg.
2144 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2146 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2148 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2150 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2152 NM/04 Bugzilla 915: Flag broken perl installation during build.
2158 TK/01 Added patch by Johannes Berg that expands the main option
2159 "spamd_address" if it starts with a dollar sign.
2161 TK/02 Write list of recipients to X-Envelope-Sender header when building
2162 the mbox-format spool file for content scanning (suggested by Jakob
2165 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2166 (http://www.dcc-servers.net/) support via dccifd. Activated by
2167 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2169 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2170 by Mark Daniel Reidel <mr@df.eu>.
2172 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2173 When building exim an external PCRE library is now needed -
2174 PCRE is a system library on the majority of modern systems.
2175 See entry on PCRE_LIBS in EDITME file.
2177 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2178 conversation. Added nologin parameter to request.
2179 Patch contributed by Kirill Miazine.
2181 TF/01 Do not log submission mode rewrites if they do not change the address.
2183 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2185 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2186 log files in place. Contributed by Roberto Lima.
2188 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2190 TF/03 Bugzilla 615: When checking the local_parts router precondition
2191 after a local_part_suffix or local_part_prefix option, Exim now
2192 does not use the address's named list lookup cache, since this
2193 contains cached lookups for the whole local part.
2195 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2196 Robert Millan. Documentation is in experimental-spec.txt.
2198 TF/04 Bugzilla 668: Fix parallel build (make -j).
2200 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2202 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2203 Patch provided by Jan Srzednicki.
2205 TF/05 Leading white space used to be stripped from $spam_report which
2206 wrecked the formatting. Now it is preserved.
2208 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2209 that they are available at delivery time.
2211 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2213 TF/08 TLS error reporting now respects the incoming_interface and
2214 incoming_port log selectors.
2216 TF/09 Produce a more useful error message if an SMTP transport's hosts
2217 setting expands to an empty string.
2219 NM/06 Bugzilla 744: EXPN did not work under TLS.
2220 Patch provided by Phil Pennock.
2222 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2223 Patch provided by Richard Godbee.
2225 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2226 acl_smtp_notquit, added index entry.
2228 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2229 Patch provided by Eugene Bujak.
2231 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2232 accept(). Patch provided by Maxim Dounin.
2234 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2235 Patch provided by Phil Pennock.
2237 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2239 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2240 Patch provided by Brad "anomie" Jorsch.
2242 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2243 Patch provided by Dean Brooks.
2245 TK/05 Add native DKIM support (does not depend on external libraries).
2247 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2248 Patch provided by Graeme Fowler.
2250 NM/16 Bugzilla 851: Documentation example syntax fix.
2252 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2254 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2257 NM/19 Bugzilla 745: TLS version reporting.
2258 Patch provided by Phil Pennock.
2260 NM/20 Bugzilla 167: bool: condition support.
2261 Patch provided by Phil Pennock.
2263 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2264 clients. Patch provided by Phil Pennock.
2266 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2267 Patch provided by Brad "anomie" Jorsch.
2269 NM/23 Bugzilla 687: Fix misparses in eximstats.
2270 Patch provided by Heiko Schlittermann.
2272 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2273 Patch provided by Heiko Schlittermann.
2275 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2276 plus update to original patch.
2278 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2280 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2281 Patch provided by David Brownlee.
2283 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2285 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2287 NM/30 Bugzilla 888: TLS documentation bugfixes.
2289 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2291 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2292 Unlike the original bugzilla I have changed all shell scripts in src tree.
2294 NM/33 Bugzilla 898: Transport filter timeout fix.
2295 Patch by Todd Rinaldo.
2297 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2298 Patch by Serge Demonchaux.
2300 NM/35 Bugzilla 39: Base64 decode bug fixes.
2301 Patch by Jakob Hirsch.
2303 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2305 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2307 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2309 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2315 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2316 ALT-N's libdkim that I have put here:
2317 http://duncanthrax.net/exim-experimental/
2319 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2320 were called 'true' and it seems that C99 defines that as a reserved
2321 keyword to be used with 'bool' variable types. That means you could
2322 not include C99-style headers which use bools without triggering
2323 build errors in sieve.c.
2325 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2326 as mailq or other aliases. Changed the --help handling significantly
2327 to do whats expected. exim_usage() emits usage/help information.
2329 SC/01 Added the -bylocaldomain option to eximstats.
2331 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2333 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2335 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2341 PH/01 Another patch from the Sieve maintainer.
2343 PH/02 When an IPv6 address is converted to a string for single-key lookup
2344 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2345 dots are used instead of colons so that keys in lsearch files need not
2346 contain colons. This was done some time before quoting was made available
2347 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2348 (notated using the quote facility) so as to distinguish them from IPv4
2349 keys. This meant that lookups for IP addresses in host lists did not work
2350 for iplsearch lookups.
2352 This has been fixed by arranging for IPv6 addresses to be expressed with
2353 colons if the lookup type is iplsearch. This is not incompatible, because
2354 previously such lookups could never work.
2356 The situation is now rather anomalous, since one *can* have colons in
2357 ordinary lsearch keys. However, making the change in all cases is
2358 incompatible and would probably break a number of configurations.
2360 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2363 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2364 conversion specification without a maximum field width, thereby enabling
2365 a rogue spamd server to cause a buffer overflow. While nobody in their
2366 right mind would setup Exim to query an untrusted spamd server, an
2367 attacker that gains access to a server running spamd could potentially
2368 exploit this vulnerability to run arbitrary code as the Exim user.
2370 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2371 $primary_hostname instead of what libspf2 thinks the hosts name is.
2373 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2374 a directory entry by the name of the lookup key. Previously, if a
2375 symlink pointed to a non-existing file or a file in a directory that
2376 Exim lacked permissions to read, a lookup for a key matching that
2377 symlink would fail. Now it is enough that a matching directory entry
2378 exists, symlink or not. (Bugzilla 503.)
2380 PH/03 The body_linecount and body_zerocount variables are now exported in the
2383 PH/04 Added the $dnslist_matched variable.
2385 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2386 This means they are set thereafter only if the connection becomes
2389 PH/06 Added the client_condition to authenticators so that some can be skipped
2390 by clients under certain conditions.
2392 PH/07 The error message for a badly-placed control=no_multiline_responses left
2393 "_responses" off the end of the name.
2395 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2397 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2398 (without spaces) instead of just copying the configuration text.
2400 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2402 PH/11 Added $max_received_linelength.
2404 PH/12 Added +ignore_defer and +include_defer to host lists.
2406 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2407 way in which PCRE > 7.0 is built.
2409 PH/14 Implemented queue_only_load_latch.
2411 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2412 MAIL command. The effect was to mangle the value on 64-bit systems.
2414 PH/16 Another patch from the Sieve maintainer.
2416 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2418 PH/18 If a system quota error occurred while trying to create the file for
2419 a maildir delivery, the message "Mailbox is full" was not appended to the
2420 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2421 only to a quota excession during the actual writing of the file.
2423 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2424 characters?) which causes problems in log lines. The DN values are now
2425 passed through string_printing() before being added to log lines.
2427 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2428 and InterBase are left for another time.)
2430 PH/21 Added message_body_newlines option.
2432 PH/22 Guard against possible overflow in moan_check_errorcopy().
2434 PH/23 POSIX allows open() to be a macro; guard against that.
2436 PH/24 If the recipient of an error message contained an @ in the local part
2437 (suitably quoted, of course), incorrect values were put in $domain and
2438 $local_part during the evaluation of errors_copy.
2444 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2445 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2448 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2449 issue a MAIL command.
2451 PH/02 In an ACL statement such as
2453 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2455 if a client was not listed at all, or was listed with a value other than
2456 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2457 the condition was not true (as it should be), so access was not denied.
2458 The bug was that the ! inversion was incorrectly passed on to the second
2459 item. This has been fixed.
2461 PH/03 Added additional dnslists conditions == and =& which are different from
2462 = and & when the dns lookup returns more than one IP address.
2464 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2465 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2467 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2468 FSYNC, which compiles an option called disable_fsync that allows for
2469 bypassing fsync(). The documentation is heavily laced with warnings.
2471 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2473 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2474 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2475 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2476 including adding "make clean"; (3) Added -fPIC when compiling the test
2477 dynamically loaded module, to get rid of a warning.
2479 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2480 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2481 = 0s. The bug is otherwise harmless.
2483 PH/07 There was a bug in the dovecot authenticator such that the value of
2484 $auth1 could be overwritten, and so not correctly preserved, after a
2485 successful authentication. This usually meant that the value preserved by
2486 the server_setid option was incorrect.
2488 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2490 PH/09 Installed PCRE release 7.0.
2492 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2493 run for batched SMTP input. It is now run at the start of every message
2494 in the batch. While fixing this I discovered that the process information
2495 (output by running exiwhat) was not always getting set for -bs and -bS
2496 input. This is fixed, and it now also says "batched" for BSMTP.
2498 PH/11 Added control=no_pipelining.
2500 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2501 patch, slightly modified), and move the expansion of helo_data till after
2502 the connection is made in the smtp transport (so it can use these
2505 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2507 PH/14 Added log_selector = +pid.
2509 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2511 PH/16 Add ${if forany and ${if forall.
2513 PH/17 Added dsn_from option to vary the From: line in DSNs.
2515 PH/18 Flush SMTP output before performing a callout, unless control =
2516 no_callout_flush is set.
2518 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2519 was true (the default) a successful delivery failed to delete the retry
2520 item, thus causing premature timeout of the address. The bug is now
2523 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2525 PH/21 Long custom messages for fakedefer and fakereject are now split up
2526 into multiline responses in the same way that messages for "deny" and
2527 other ACL rejections are.
2529 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2530 with slight modification.
2532 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2533 draft, changing the syntax and factoring some duplicate code".
2535 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2536 for deliveries of the second and subsequent messages over the same SMTP
2539 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2540 ${reduce, with only minor "tidies".
2542 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2544 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2545 expansion side effects.
2547 PH/27 When a message times out after an over-quota error from an Exim-imposed
2548 quota, the bounce message says "mailbox is full". This message was not
2549 being given when it was a system quota that was exceeded. It now should
2552 MH/03 Made $recipients available in local_scan(). local_scan() already has
2553 better access to the recipient list through recipients_list[], but
2554 $recipients can be useful in postmaster-provided expansion strings.
2556 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2557 in the case of a MAIL command with additional options following the
2558 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2559 were accidentally chopped off.
2561 PH/29 SMTP synchronization checks are implemented when a command is read -
2562 there is a check that no more input is waiting when there shouldn't be
2563 any. However, for some commands, a delay in an ACL can mean that it is
2564 some time before the response is written. In this time, more input might
2565 arrive, invalidly. So now there are extra checks after an ACL has run for
2566 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2567 pipelining has not been advertised.
2569 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2571 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2572 returned in the SMTP response when smtp_return_error_details was false.
2573 This has been fixed.
2575 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2576 instead of the C I/O that was originally supplied, because problems were
2577 reported on Solaris.
2579 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2580 Exim which did not show up earlier: it was assuming that a call to
2581 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2582 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2583 was a macro that became an assignment, so it seemed to work. This has
2584 changed to a proper function call with a void return, hence the compile
2585 error. Exim's code has been fixed.
2587 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2590 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2592 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2594 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2595 because it turns out to be too big or there is a timeout) there is no
2596 "Completed" line in the log. When some messages of this type were
2597 selected by exigrep, they were listed as "not completed". Others were
2598 picked up by some special patterns. I have improved the selection
2599 criteria to be more general.
2601 PH/38 The host_find_failed option in the manualroute router can now be set
2602 to "ignore", to completely ignore a host whose IP address cannot be
2603 found. If all hosts are ignored, the behaviour is controlled by the new
2604 host_all_ignored option.
2606 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2607 homing or because of multiple MX records with /mx) generated more than
2608 one IP address, and the following item turned out to be the local host,
2609 all the secondary addresses of the first item were incorrectly removed
2610 from the list, along with the local host and any following hosts (which
2611 is what is supposed to happen).
2613 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2614 whoever called Exim into the -H file. In the case of the daemon it was
2615 behaving confusingly. When first started, it used values for whoever
2616 started the daemon, but after a SIGHUP it used the Exim user (because it
2617 calls itself on a restart). I have changed the code so that it now always
2620 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2621 message are rejected with the same error (e.g. no authentication or bad
2622 sender address), and a DATA command is nevertheless sent (as can happen
2623 with PIPELINING or a stupid MUA), the error message that was given to the
2624 RCPT commands is included in the rejection of the DATA command. This is
2625 intended to be helpful for MUAs that show only the final error to their
2628 PH/42 Another patch from the Sieve maintainer.
2630 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2631 Eximstats - Fixed some broken HTML links and added missing column headers
2633 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2634 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2636 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2638 PH/43 Yet another patch from the Sieve maintainer.
2640 PH/44 I found a way to check for a TCP/IP connection going away before sending
2641 the response to the final '.' that terminates a message, but only in the
2642 case where the client has not sent further data following the '.'
2643 (unfortunately, this is allowed). However, in many cases there won't be
2644 any further data because there won't be any more messages to send. A call
2645 to select() can be used: if it shows that the input is "ready", there is
2646 either input waiting, or the socket has been closed. An attempt to read
2647 the next input character can distinguish the two cases. Previously, Exim
2648 would have sent an OK response which the client would never have see.
2649 This could lead to message repetition. This fix should cure that, at
2650 least in a lot of common cases.
2652 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2653 advertised in response to EHLO.
2659 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2660 fixed by 4.65/MH/01 (is this a record?) are fixed:
2662 (i) An empty string was always treated as zero by the numeric comparison
2663 operators. This behaviour has been restored.
2665 (ii) It is documented that the numeric comparison operators always treat
2666 their arguments as decimal numbers. This was broken in that numbers
2667 starting with 0 were being interpreted as octal.
2669 While fixing these problems I realized that there was another issue that
2670 hadn't been noticed. Values of message_size_limit (both the global option
2671 and the transport option) were treated as octal if they started with 0.
2672 The documentation was vague. These values are now always treated as
2673 decimal, and I will make that clear in the documentation.
2679 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2680 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2683 MH/01 Don't check that the operands of numeric comparison operators are
2684 integers when their expansion is in "skipping" mode (fixes bug
2685 introduced by 4.64-PH/07).
2687 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2688 child addresses, Exim now panics and dies. Previously, because the count
2689 is held in a short int, deliveries were likely to be lost. As such a
2690 large number of recipients for a single message is ridiculous
2691 (performance will be very, very poor), I have chosen to impose a limit
2692 rather than extend the field.
2698 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2699 leftover -K file (the existence of which was triggered by #402).
2700 While we were at it, introduced process PID as part of the -K
2701 filename. This should rule out race conditions when creating
2704 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2705 processing considerably. Previous code took too long for large mails,
2706 triggering a timeout which in turn triggers #401.
2708 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2709 in the DK code in transports.c. sendfile() is not really portable,
2710 hence the _LINUX specificness.
2712 TF/01 In the add_headers option to the mail command in an Exim filter,
2713 there was a bug that Exim would claim a syntax error in any
2714 header after the first one which had an odd number of characters
2717 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2718 callout verification, Exim cached a "reject" for the entire domain. This
2719 is correct for most verifications, but it is not correct for a recipient
2720 verification with use_sender or use_postmaster set, because in that case
2721 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2722 case of MAIL FROM:<> rejection from other early rejections (e.g.
2723 rejection of HELO). When verifying a recipient using a non-null MAIL
2724 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2725 Whatever the result of the callout, the value of the domain cache is
2726 left unchanged (for any other kind of callout, getting as far as trying
2727 RCPT means that the domain itself is ok).
2729 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2732 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2733 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2734 session, the daemon ignores ECONNECT errors and logs others; it now
2735 ignores EPIPE as well.
2737 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2738 (quoted-printable decoding).
2740 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2741 later the small subsequent patch to fix an introduced bug.
2743 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2745 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2747 PH/08 An error is now given if message_size_limit is specified negative.
2749 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2750 to be given (somewhat) arbitrary names.
2752 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2755 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2756 miscellaneous code fixes
2758 PH/10 Added the log_reject_target ACL modifier to specify where to log
2761 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2762 hostname. This is wrong, because it relates to the incoming message (and
2763 probably the interface on which it is arriving) and not to the outgoing
2764 callout (which could be using a different interface). This has been
2765 changed to use the value of the helo_data option from the smtp transport
2766 instead - this is what is used when a message is actually being sent. If
2767 there is no remote transport (possible with a router that sets up host
2768 addresses), $smtp_active_hostname is used.
2770 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2771 tweaks were necessary in order to get it to work (see also 21 below):
2772 (a) The code assumed that strncpy() returns a negative number on buffer
2773 overflow, which isn't the case. Replaced with Exim's string_format()
2775 (b) There were several signed/unsigned issues. I just did the minimum
2776 hacking in of casts. There is scope for a larger refactoring.
2777 (c) The code used strcasecmp() which is not a standard C function.
2778 Replaced with Exim's strcmpic() function.
2779 (d) The code set only $1; it now sets $auth1 as well.
2780 (e) A simple test gave the error "authentication client didn't specify
2781 service in request". It would seem that Dovecot has changed its
2782 interface. Fortunately there's a specification; I followed it and
2783 changed what the client sends and it appears to be working now.
2785 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2788 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2789 address A is aliased to B and C, where B exists and C does not. Without
2790 -v the output is "A verified" because verification stops after a
2791 successful redirection if more than one address is generated. However,
2792 with -v the child addresses are also verified. Exim was outputting "A
2793 failed to verify" and then showing the successful verification for C,
2794 with its parentage. It now outputs "B failed to verify", showing B's
2795 parentage before showing the successful verification of C.
2797 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2798 look up a TXT record in a specific list after matching in a combined
2801 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2802 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2803 they consult the DNS. I had assumed they would set it the way they
2804 wanted; and indeed my experiments on Linux seem to show that in some
2805 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2806 To be on the safe side, however, I have now made the interface to
2807 host_find_byname() similar to host_find_bydns(), with an argument
2808 containing the DNS resolver options. The host_find_byname() function now
2809 sets these options at its start, just as host_find_bydns() does. The smtp
2810 transport options dns_qualify_single and dns_search_parents are passed to
2811 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2812 of host_find_byname() use the default settings of RES_DEFNAMES
2813 (qualify_single) but not RES_DNSRCH (search_parents).
2815 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2816 spool_read_header() do less string testing, by means of a preliminary
2817 switch on the second character of optional "-foo" lines. (This is
2818 overdue, caused by the large number of possibilities that now exist.
2819 Originally there were few.) While I was there, I also converted the
2820 str(n)cmp tests so they don't re-test the leading "-" and the first
2821 character, in the hope this might squeeze out yet more improvement.
2823 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2824 flag allowing group syntax was set by the header_syntax check but not
2825 turned off, possible causing trouble later; (2) The flag was not being
2826 set at all for the header_verify test, causing "group"-style headers to
2827 be rejected. I have now set it in this case, and also caused header_
2828 verify to ignore an empty address taken from a group. While doing this, I
2829 came across some other cases where the code for allowing group syntax
2830 while scanning a header line wasn't quite right (mostly, not resetting
2831 the flag correctly in the right place). These bugs could have caused
2832 trouble for malformed header lines. I hope it is now all correct.
2834 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2835 with the "reply" argument non-NULL. The code, however (which originally
2836 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2837 but it didn't always do it. This confused somebody who was copying the
2838 code for some other use. I have removed all the tests.
2840 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2841 feature that was used to support insecure browsers during the U.S. crypto
2842 embargo. It requires special client support, and Exim is probably the
2843 only MTA that supported it -- and would never use it because real RSA is
2844 always available. This code has been removed, because it had the bad
2845 effect of slowing Exim down by computing (never used) parameters for the
2846 RSA_EXPORT functionality.
2848 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2849 authenticator to fail if there's a tab character in the incoming data
2850 (there should never be unless someone is messing about, as it's supposed
2851 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2852 if the connection is using TLS or if the remote IP is the same as the
2853 local IP, and the "valid-client-cert option" if a client certificate has
2856 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2857 authenticators. This can be used for authorization after authentication
2858 succeeds. (In the case of plaintext, it servers for both authentication
2861 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2862 if any retry times were supplied.
2864 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2865 connection, where there is no client IP address to check. In this
2866 situation, the verify now always succeeds.
2868 PH/25 Applied John Jetmore's -Mset patch.
2870 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2872 PH/27 In a string expansion for a processed (not raw) header when multiple
2873 headers of the same name were present, leading whitespace was being
2874 removed from all of them, but trailing whitespace was being removed only
2875 from the last one. Now trailing whitespace is removed from each header
2876 before concatenation. Completely empty headers in a concatenation (as
2877 before) are ignored.
2879 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2880 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2882 PH/29 [Removed. This was a change that I later backed out, and forgot to
2883 correct the ChangeLog entry (that I had efficiently created) before
2884 committing the later change.]
2886 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2887 address errors (4xx response to RCPT) over the same connection as other
2888 messages routed to the same hosts. Such deliveries are always "forced",
2889 so retry times are not inspected. This resulted in far too many retries
2890 for the affected addresses. The effect occurred only when there were more
2891 hosts than the hosts_max_try setting in the smtp transport when it had
2892 the 4xx errors. Those hosts that it had tried were not added to the list
2893 of hosts for which the message was waiting, so if all were tried, there
2894 was no problem. Two fixes have been applied:
2896 (i) If there are any address or message errors in an SMTP delivery, none
2897 of the hosts (tried or untried) are now added to the list of hosts
2898 for which the message is waiting, so the message should not be a
2899 candidate for sending over the same connection that was used for a
2900 successful delivery of some other message. This seems entirely
2901 reasonable: after all the message is NOT "waiting for some host".
2902 This is so "obvious" that I'm not sure why it wasn't done
2903 previously. Hope I haven't missed anything, but it can't do any
2904 harm, as the worst effect is to miss an optimization.
2906 (ii) If, despite (i), such a delivery is accidentally attempted, the
2907 routing retry time is respected, so at least it doesn't keep
2908 hammering the server.
2910 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2911 in ${readsocket because some servers need this prod.
2913 PH/32 Added some extra debug output when updating a wait-xxx database.
2915 PH/33 The hint "could be header name not terminated by colon", which has been
2916 given for certain expansion errors for a long time, was not being given
2917 for the ${if def:h_colon_omitted{... case.
2919 PH/34 The spec says: "With one important exception, whenever a domain list is
2920 being scanned, $domain contains the subject domain." There was at least
2921 one case where this was not true.
2923 PH/35 The error "getsockname() failed: connection reset by peer" was being
2924 written to the panic log as well as the main log, but it isn't really
2925 panic-worthy as it just means the connection died rather early on. I have
2926 removed the panic log writing for the ECONNRESET error when getsockname()
2929 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2930 runs only) independently of the message's sender address. This meant
2931 that, if the 4xx error was in fact related to the sender, a different
2932 message to the same recipient with a different sender could confuse
2933 things. In particular, this can happen when sending to a greylisting
2934 server, but other circumstances could also provoke similar problems.
2935 I have changed the default so that the retry time for these errors is now
2936 based a combination of the sender and recipient addresses. This change
2937 can be overridden by setting address_retry_include_sender=false in the
2940 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2941 remote server are returned as part of bounce messages. This was not
2942 happening for LMTP over a pipe (the lmtp transport), but now it is the
2943 same for both kinds of LMTP.
2945 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2946 in header lines that were in fact CNAMEs. This is no longer the case.
2948 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2949 and queue runs started by the daemon processed all messages. This has
2950 been fixed so that -R and -S can now usefully be given with -q<time>.
2952 PH/40 Import PCRE release 6.7 (fixes some bugs).
2954 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2956 PH/42 Give an error if -q is specified more than once.
2958 PH/43 Renamed the variables $interface_address and $interface_port as
2959 $received_ip_address and $received_port, to make it clear that these
2960 values apply to message reception, and not to the outgoing interface when
2961 a message is delivered. (The old names remain recognized, of course.)
2963 PH/44 There was no timeout on the connect() call when using a Unix domain
2964 socket in the ${readsocket expansion. There now is.
2966 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2967 be meaningful with "accept".
2969 SC/01 Eximstats V1.43
2970 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2972 SC/02 Eximstats V1.44
2973 Use a glob alias rather than an array ref in the generated
2974 parser. This improves both readability and performance.
2976 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2977 Collect SpamAssassin and rejection statistics.
2978 Don't display local sender or destination tables unless
2979 there is data to show.
2980 Added average volumes into the top table text output.
2982 SC/04 Eximstats V1.46
2983 Collect data on the number of addresses (recipients)
2984 as well as the number of messages.
2986 SC/05 Eximstats V1.47
2987 Added 'Message too big' to the list of mail rejection
2988 reasons (thanks to Marco Gaiarin).
2990 SC/06 Eximstats V1.48
2991 Mainlog lines which have GMT offsets and are too short to
2992 have a flag are now skipped.
2994 SC/07 Eximstats V1.49 (Alain Williams)
2995 Added the -emptyok flag.
2997 SC/08 Eximstats V1.50
2998 Fixes for obtaining the IP address from reject messages.
3000 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
3001 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
3002 whitespace changes from 4.64-PH/27
3004 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
3007 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
3008 are found, allow negative numbers in numeric criteria)
3010 JJ/06 exipick.20061117.2, added new $message_body_missing variable
3012 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
3013 to match changes made in 4.64-PH/43
3015 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
3017 PH/47 Put in an explicit test for a DNS lookup of an address record where the
3018 "domain" is actually an IP address, and force a failure. This locks out
3019 those revolvers/nameservers that support "A-for-A" lookups, in
3020 contravention of the specifications.
3022 PH/48 When a host name was looked up from an IP address, and the subsequent
3023 forward lookup of the name timed out, the host name was left in
3024 $sender_host_name, contrary to the specification.
3026 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3027 restricted to single-key lookups, Exim was not diagnosing an error if
3028 * or *@ was used with a query-style lookup.
3030 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3032 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3033 long ago, but noone interested enough thought of it. Let's just say that
3034 the "1.1" means that there are some new functions that weren't there at
3035 some point in the past.
3037 PH/51 Error processing for expansion failure of helo_data from an smtp
3038 transport during callout processing was broken.
3040 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3041 tested/used via the -bh/-bhc/-bs options.
3043 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3044 bug, fixed in subsequent PCRE releases).
3046 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3047 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3049 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3055 SC/01 Use a glob alias rather than an array ref in eximstats generated
3056 parser. This improves both readability and performance.
3058 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3059 Don't display local sender or destination tables in eximstats unless
3060 there is data to show.
3061 Added average volumes into the eximstats top table text output.
3063 SC/03 Collect data on the number of addresses (recipients) as well
3064 as the number of messages in eximstats.
3066 TF/01 Correct an error in the documentation for the redirect router. Exim
3067 does (usually) call initgroups() when daemonizing.
3069 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3070 with consistent privilege compared to when running as a daemon.
3072 TF/03 Note in the spec that $authenticated_id is not set for local
3073 submissions from trusted users.
3075 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3076 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3078 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3079 by adding some example configuration directives to the default
3080 configuration file. A little bit of work is required to uncomment the
3081 directives and define how usernames and passwords are checked, but
3082 there is now a framework to start from.
3084 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3085 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3086 without this. I don't know how relevant this is to other LDAP libraries.
3088 PH/02 Add the verb name to the "unknown ACL verb" error.
3090 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3092 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3094 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3095 directory not expanded when it should be if an expanded home directory
3096 was set for the address (which is overridden by the transport).
3098 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3101 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3102 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3103 because it is too late at that time, and has no effect.
3105 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3106 security issue with \' (bugzilla #107). I could not use the
3107 PQescapeStringConn() function, because it needs a PGconn value as one of
3110 PH/08 When testing addresses using -bt, indicate those final addresses that
3111 are duplicates that would not cause an additional delivery. At least one
3112 person was confused, thinking that -bt output corresponded to deliveries.
3113 (Suppressing duplicates isn't a good idea as you lose the information
3114 about possibly different redirections that led to the duplicates.)
3116 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3117 systems where poll() doesn't work, in particular OS X.
3119 PH/10 Added more information to debugging output for retry time not reached.
3121 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3122 operations in malware.c.
3124 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3127 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3128 syslog_duplication set false, log lines that would normally be written
3129 both the the main log and to the reject log were not written to syslog at
3132 PH/14 In the default configuration, change the use of "message" in ACL warn
3133 statements to "add_header".
3135 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3136 not followed by a command (e.g. "seen endif").
3138 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3139 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3142 PH/17 Added extra conditions to the default value of delay_warning_condition
3146 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3147 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3148 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3151 The Auto-Submitted: and various List- headers are standardised, whereas I
3152 don't think Precedence: ever was.
3154 PH/18 Refactored debugging code in route_finduser() to show more information,
3155 in particular, the error code if getpwnam() issues one.
3157 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3158 This is apparently needed in addition to the PH/07 change above to avoid
3159 any possible encoding problems.
3161 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3162 but not after initializing Perl.
3164 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3165 output them only if debugging. By default they are written stderr,
3166 apparently, which is not desirable.
3168 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3171 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3174 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3176 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3177 authenticated or an ident call has been made. Suppress the default
3178 values for $authenticated_id and $authenticated_sender (but permit -oMai
3179 and -oMas) when testing with -bh.
3181 PH/24 Re-jigged the order of the tests in the default configuration so that the
3182 tests for valid domains and recipients precede the DNS black list and CSA
3183 tests, on the grounds that those ones are more expensive.
3185 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3186 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3187 This bug exists in every version of Exim that I still have, right back to
3190 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3191 However, an attempt to turn on case-sensitivity in a regex key by
3192 including (?-i) didn't work because the subject string was already
3193 lowercased, and the effects were non-intuitive. It turns out that a
3194 one-line patch can be used to allow (?-i) to work as expected.
3200 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3201 other effects) broke the use of negated acl sub-conditions.
3203 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3206 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3207 "Deny" causes Exim to reject the incoming connection with a 554 error.
3208 Unfortunately, if there is a major crisis, such as a disk failure,
3209 tcp-wrappers gives "deny", whereas what one would like would be some
3210 kind of temporary error. A kludge has been added to help with this.
3211 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3212 554 error is used if errno is still zero or contains ENOENT (which occurs
3213 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3216 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3218 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3219 errors. Otherwise a message that provokes a temporary error (when other
3220 messages do not) can cause a whole host to time out.
3222 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3223 addresses were routed directly to files or pipes from a redirect router.
3224 File deliveries just didn't batch; pipe deliveries might have suffered
3227 PH/06 A failure to get a lock for a hints database would erroneously always say
3228 "Failed to get write lock", even when it was really a read lock.
3230 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3231 of 0600. This has been changed to use the value of the lockfile_mode
3232 option (which defaults to 0600).
3234 PH/08 Applied small patch from the Sieve maintainer.
3236 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3237 folder from quota calculations, a direct delivery into this folder messed
3238 up the contents of the maildirsize file. This was because the regex was
3239 used only to exclude .Trash (or whatever) when the size of the mailbox
3240 was calculated. There was no check that a delivery was happening into an
3241 excluded directory. This bug has been fixed by ignoring all quota
3242 processing for deliveries into excluded directories.
3244 PH/10 Added the maildirfolder_create_regex option to appendfile.
3250 PH/01 The code for finding all the local interface addresses on a FreeBSD
3251 system running IPv6 was broken. This may well have applied to all BSD
3252 systems, as well as to others that have similar system calls. The broken
3253 code found IPv4 interfaces correctly, but gave incorrect values for the
3254 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3255 that it would not match correctly against @[] and not recognize the IPv6
3258 PH/02 The ipliteral router was not recognizing addresses of the form user@
3259 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3261 PH/03 Added disable_ipv6.
3263 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3264 decoded form, because it is most often used to construct To: headers
3265 lines in autoreplies, and the decoded form may well be syntactically
3266 invalid. However, $reply_address has leading white space removed, and all
3267 newlines turned into spaces so that the autoreply transport does not
3270 PH/05 If group was specified without a user on a router, and no group or user
3271 was specified on a transport, the group from the router was ignored.
3273 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3274 for visible compile-time settings that can be used to change these
3275 numbers, for those that want even more. Backwards compatibility with old
3276 spool files has been maintained. However, going back to a previous Exim
3277 release will lost any variables that are in spool files.
3279 PH/07 Two small changes when running in the test harness: increase delay when
3280 passing a TCP/IP connection to a new process, in case the original
3281 process has to generate a bounce, and remove special handling of
3282 127.0.0.2 (sic), which is no longer necessary.
3284 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3285 be the same on different OS.
3287 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3290 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3291 whether --show-vars was specified or not
3293 JJ/02 exipick: Added support for new ACL variable spool format introduced
3296 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3297 syntactically invalid From: or Reply-to: line, and a filter used this to
3298 generate an autoreply, and therefore failed to obtain an address for the
3299 autoreply, Exim could try to deliver to a non-existent relative file
3300 name, causing unrelated and misleading errors. What now happens is that
3301 it logs this as a hard delivery error, but does not attempt to create a
3304 PH/11 The exinext utility has a -C option for testing purposes, but although
3305 the given file was scanned by exinext itself; it wasn't being passed on
3306 when Exim was called.
3308 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3309 an end-of-file indication when reading a command response.
3311 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3312 compiled. In many other places in Exim, IPv6 addresses are always
3313 recognized, so I have changed this. It also means that IPv4 domain
3314 literals of the form [IPV4:n.n.n.n] are now always recognized.
3316 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3317 used if the router is not running as root, for example, when verifying at
3318 ACL time, or when using -bh. The debugging output from this situation was
3319 non-existent - all you got was a failure to exec. I have made two
3322 (a) Failures to set uid/gid, the current directory, or a process leader
3323 in a subprocess such as that created by queryprogram now generate
3324 suitable debugging output when -d is set.
3326 (b) The queryprogram router detects when it is not running as root,
3327 outputs suitable debugging information if -d is set, and then runs
3328 the subprocess without attempting to change uid/gid.
3330 PH/15 Minor change to Makefile for building test_host (undocumented testing
3333 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3334 additional section of a DNS packet that returns MX or SRV records.
3335 Instead, it always explicitly searches for A/AAAA records. This avoids
3336 major problems that occur when a DNS server includes only records of one
3337 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3338 fixed another bug: if SRV records were looked up and the corresponding
3339 address records were *not* found in the additional section, the port
3340 values from the SRV records were lost.
3342 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3343 using the correct key (the original address) when searching the retry
3344 rules in order to find which one to use for generating the retry hint.
3346 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3347 adding the default one. Similarly, if it contains a Reply-To: header, the
3348 errors_reply_to option, if set, is not used.
3350 PH/19 When calculating a retry time, Exim used to measure the "time since
3351 failure" by looking at the "first failed" field in the retry record. Now
3352 it does not use this if it is later than than the arrival time of the
3353 message. Instead it uses the arrival time. This makes for better
3354 behaviour in cases where some deliveries succeed, thus re-setting the
3355 "first failed" field. An example is a quota failure for a huge message
3356 when small messages continue to be delivered. Without this change, the
3357 "time since failure" will always be short, possible causing more frequent
3358 delivery attempts for the huge message than are intended.
3359 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3361 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3362 $1, $2, $3) because the numerical variables can be reset during some
3363 expansion items (e.g. "match"), thereby losing the authentication data.
3365 PH/21 Make -bV show the size of off_t variables so that the test suite can
3366 decide whether to run tests for quotas > 2G.
3368 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3369 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3370 filecount value is greater than 2G or if a quota value is greater than 2G
3371 on a system where the size of off_t is not greater than 4, a panic error
3374 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3375 never match. The debug and -bh output now contains an explicit error
3376 message indicating a malformed IPv4 address or mask.
3378 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3379 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3380 PH/23 above applies.
3382 PH/25 Do not write to syslog when running in the test harness. The only
3383 occasion when this arises is a failure to open the main or panic logs
3384 (for which there is an explicit test).
3386 PH/26 Added the /no_tell option to "control=freeze".
3388 PH/27 If a host name lookup failed very early in a connection, for example, if
3389 the IP address matched host_lookup and the reverse lookup yielded a name
3390 that did not have a forward lookup, an error message of the form "no IP
3391 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3392 could be logged. Now it outputs the IP address instead of "NULL".
3394 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3395 allows the sender and the authenticated sender to be set when
3396 submitting a message from within Exim. Since child_open_exim() is
3397 documented for local_scan(), the new function should be too.
3399 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3400 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3401 results in an empty string is now treated as unset.
3403 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3405 PH/31 Added sender_verify_fail logging option.
3407 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3408 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3409 tidied the source and removed it altogether.
3411 PH/33 When a queue run was abandoned because the load average was too high, a
3412 log line was always written; now it is written only if the queue_run log
3413 selector is set. In addition, the log line for abandonment now contains
3414 information about the queue run such as the pid. This is always present
3415 in "start" and "stop" lines but was omitted from the "abandon" line.
3417 PH/34 Omit spaces between a header name and the colon in the error message that
3418 is given when verify = headers_syntax fails (if there are lots of them,
3419 the message gets confusing).
3421 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3422 names, as there are now some PTR records that contain slashes. This check
3423 is only to protect against broken name servers that fall over on strange
3424 characters, so the fact that it applies to all lookups doesn't matter.
3426 PH/36 Now that the new test suite is complete, we can remove some of the
3427 special code in Exim that was needed for the old test suite. For example,
3428 sorting DNS records because real resolvers return them in an arbitrary
3429 order. The new test suite's fake resolver always returns records in the
3432 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3433 bounces) except when queue_only is set, to avoid logging races between
3434 the different processes.
3436 PH/38 Panic-die if .include specifies a non-absolute path.
3438 PH/39 A tweak to the "H" retry rule from its user.
3440 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3441 a label. They prevented compilation on older perls.
3443 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3444 a warning to be raised on newish perls.
3446 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3447 on queue. Changes to match documented behaviour of showing count of
3448 messages matching specified criteria.
3450 PH/40 Changed the default ident timeout from 30s to 5s.
3452 PH/41 Added support for the use of login_cap features, on those BSD systems
3453 that have them, for controlling the resources used by pipe deliveries.
3455 PH/42 The content-scanning code uses fopen() to create files in which to put
3456 message data. Previously it was not paying any attention to the mode of
3457 the files. Exim runs with umask(0) because the rest of the code creates
3458 files with open(), and sets the required mode explicitly. Thus, these
3459 files were ending up world-writeable. This was not a big issue, because,
3460 being within the spool directory, they were not world-accessible. I have
3461 created a function called modefopen, which takes an additional mode
3462 argument. It sets umask(777), creates the file, chmods it to the required
3463 mode, then resets the umask. All the relevant calls to fopen() in the
3464 content scanning code have been changed to use this function.
3466 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3467 to 24 hours. This avoids potential overflow problems when processing G
3468 and H retry rules. I suspect nobody ever tinkers with this value.
3470 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3472 PH/45 When the plaintext authenticator is running as a client, the server's
3473 challenges are checked to ensure they are valid base64 strings. By
3474 default, the authentication attempt is cancelled if an invalid string is
3475 received. Setting client_ignore_invalid_base64 true ignores these errors.
3476 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3477 they are received. Thus, the responses can be made to depend on the
3478 challenges. If an invalid string is ignored, an empty string is placed in
3481 PH/46 Messages that are created by the autoreply transport now contains a
3482 References: header, in accordance with RFCs 2822 and 3834.
3484 PH/47 Added authenticated_sender_force to the smtp transport.
3486 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3488 PH/49 Installed latest patch from the Sieve maintainer.
3490 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3491 was also set, the appendfile transport was unnecessarily scanning a
3492 directory of message files (e.g. for maildir delivery) to find the count
3493 of files (along with the size), even though it did not need this
3494 information. It now does the scan only if it needs to find either the
3495 size of the count of files.
3497 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3499 PH/52 Two bugs concerned with error handling when the smtp transport is
3502 (i) Exim was not creating retry information for temporary errors given
3503 for individual recipients after the DATA command when the smtp transport
3504 was used in LMTP mode. This meant that they could be retried too
3505 frequently, and not timed out correctly.
3507 (ii) Exim was setting the flag that allows error details to be returned
3508 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3509 recipients that were returned after the DATA command.
3511 PH/53 This is related to PH/52, but is more general: for any failing address,
3512 when detailed error information was permitted to be returned to the
3513 sender, but the error was temporary, then after the final timeout, only
3514 "retry timeout exceeded" was returned. Now it returns the full error as
3515 well as "retry timeout exceeded".
3517 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3518 do this, and (what is worse) MTAs that accept it.
3520 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3521 will now be deprecated.
3523 PH/56 New os.c-cygwin from the Cygwin maintainer.
3525 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3526 formats (previously only available in exim formats via -bpr, -bpru,
3527 and -bpra. Now also available in native and exiqgrep formats)
3529 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3530 with very large, slow to parse queues
3532 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3534 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3536 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3537 responses to authentication challenges, though it was showing the
3538 challenges; (ii) I've removed the CR characters from the debug output for
3541 PH/58 Allow for the insertion of a newline as well as a space when a string
3542 is turned into more than one encoded-word during RFC 2047 encoding. The
3543 Sieve code now uses this.
3545 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3546 data_4xx, lost_connection, tls_required.
3548 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3549 message was being sent as an SMTP response.
3551 PH/61 Add -l and -k options to exicyclog.
3553 PH/62 When verifying, if an address was redirected to one new address, so that
3554 verification continued, and the new address failed or deferred after
3555 having set something in $address_data, the value of $address_data was not
3556 passed back to the ACL. This was different to the case when no
3557 redirection occurred. The value is now passed back in both cases.
3559 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3560 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3561 use login_cap.h, so on its own it isn't the distinguishing feature. The
3562 new name refers directly to the setclassresources() function.
3564 PH/65 Added configuration files for NetBSD3.
3566 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3568 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3569 is preferred over IPv4.
3571 PH/68 The bounce_return_message and bounce_return_body options were not being
3572 honoured for bounces generated during the reception of non-SMTP messages.
3573 In particular, this applied to messages rejected by the ACL. This bug has
3574 been fixed. However, if bounce_return_message is true and bounce_return_
3575 body is false, the headers that are returned for a non-SMTP message
3576 include only those that have been read before the error was detected.
3577 (In the case of an ACL rejection, they have all been read.)
3579 PH/69 The HTML version of the specification is now built in a directory called
3580 spec_html instead of spec.html, because the latter looks like a path with
3581 a MIME-type, and this confuses some software.
3583 PH/70 Catch two compiler warnings in sieve.c.
3585 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3586 function verify_get_ident() calls ip_connect() to connect a socket, but
3587 if the "connect()" function timed out, ip_connect() used to close the
3588 socket. However, verify_get_ident() also closes the socket later, and in
3589 between Exim writes to the log, which may get opened at this point. When
3590 the socket was closed in ip_connect(), the log could get the same file
3591 descriptor number as the socket. This naturally causes chaos. The fix is
3592 not to close the socket in ip_connect(); the socket should be closed by
3593 the function that creates it. There was only one place in the code where
3594 this was missing, in the iplookup router, which I don't think anybody now
3595 uses, but I've fixed it anyway.
3597 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3598 well as to direct DNS lookups. Otherwise the handling of names in host
3599 lists is inconsistent and therefore confusing.
3605 PH/01 Two changes to the default runtime configuration:
3607 (1) Move the checks for relay_from_hosts and authenticated clients from
3608 after to before the (commented out) DNS black list checks.
3610 (2) Add control=submission to the relay_from_hosts and authenticated
3611 clients checks, on the grounds that messages accepted by these
3612 statements are most likely to be submissions.
3614 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3616 (1) Generate an error if the third argument for the ${prvs expansion is
3619 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3622 (3) Reset the variables that are obtained from the first argument of
3623 ${prvscheck and used in the second argument before leaving the code,
3624 because their memory is reclaimed, so using them afterwards may do
3627 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3628 one (it's much easier than Tom thought :-).
3630 (5) Because of (4), we can now allow for the use of $prvscheck_result
3631 inside the third argument.
3633 PH/03 For some reason, the default setting of PATH when running a command from
3634 a pipe transport was just "/usr/bin". I have changed it to
3637 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3638 anything to be listed in the output from -bV.
3640 PH/05 When a filter generated an autoreply, the entire To: header line was
3641 quoted in the delivery log line, like this:
3643 => >A.N.Other <ano@some.domain> <original@ddress> ...
3645 This has been changed so that it extracts the operative address. There
3646 may be more than one such address. If so, they are comma-separated, like
3649 => >ano@some.domain,ona@other.domain <original@ddress> ...
3651 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3652 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3653 not being looked up in the rDNS to get a host name, Exim was showing the
3654 IP address twice in Received: lines, even though the IP addresses were
3655 identical. For example:
3657 Received: from [1.2.3.4] (helo=[1.2.3.4])
3659 However, if the real host name was known, it was omitting the HELO data
3660 if it matched the actual IP address. This has been tidied up so that it
3661 doesn't show the same IP address twice.
3663 PH/07 When both +timestamp and +memory debugging was on, the value given by
3664 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3665 called by the debug printing, thereby overwriting the timestamp buffer.
3666 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3668 PH/08 When the original message was included in an autoreply transport, it
3669 always said "this is a copy of the message, including all the headers",
3670 even if body_only or headers_only was set. It now gives an appropriate
3673 PH/09 Applied a patch from the Sieve maintainer which:
3675 o fixes some comments
3676 o adds the (disabled) notify extension core
3677 o adds some debug output for the result of if/elsif tests
3678 o points to the current vacation draft in the documentation
3679 and documents the missing references header update
3683 o fixes a bug in processing the envelope test (when testing
3684 multiple envelope elements, the last element determined the
3687 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3688 Electronic Mail") by including:
3690 Auto-submitted: auto-generated
3692 in the messages that it generates (bounce messages and others, such as
3693 warnings). In the case of bounce messages for non-SMTP messages, there was
3694 also a typo: it was using "Auto_submitted" (underscore instead of
3695 hyphen). Since every message generated by Exim is necessarily in response
3696 to another message, thes have all been changed to:
3698 Auto-Submitted: auto-replied
3700 in accordance with these statements in the RFC:
3702 The auto-replied keyword:
3704 - SHOULD be used on messages sent in direct response to another
3705 message by an automatic process,
3707 - MUST NOT be used on manually-generated messages,
3709 - MAY be used on Delivery Status Notifications (DSNs) and Message
3710 Disposition Notifications (MDNs),
3712 - MUST NOT be used on messages generated by automatic or periodic
3713 processes, except for messages which are automatic responses to
3716 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3717 to the default Received: header definition.
3719 PH/12 Added log selector acl_warn_skipped (default on).
3721 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3722 variables because (a) they are in the wrong storage pool and (b) even if
3723 they were copied, it wouldn't work properly because of the caching.
3725 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3726 checking when decoding. Apparently there are clients that generate
3727 overlong encoded strings. Why am I not surprised?
3729 PH/15 If the first argument of "${if match_address" was not empty, but did not
3730 contain an "@" character, Exim crashed. Now it writes a panic log message
3731 and treats the condition as false.
3733 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3735 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3736 extension "envelope-auth". The code is finished and in agreement with
3737 other implementations, but there is no documentation so far and in fact,
3738 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3739 not changing the active code.
3741 Print executed "if" and "elsif" statements when debugging is used. This
3742 helps a great deal to understand what a filter does.
3744 Document more things not specified clearly in RFC3028. I had all this
3745 sorted out, when out of a sudden new issues came to my mind. Oops."
3747 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3750 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3751 canonical form (as documented). However, after a host name lookup from
3752 the IP address, check_host() was doing a simple string comparison with
3753 addresses acquired from the DNS when checking that the found name did
3754 have the original IP as one of its addresses. Since any found IPv6
3755 addresses are likely to be in abbreviated form, the comparison could
3756 fail. Luckily, there already exists a function for doing the comparison
3757 by converting both addresses to binary, so now that is used instead of
3758 the text comparison.
3760 PH/20 There was another similar case to PH/19, when a complete host name was
3761 given in a host list; looking up its IP address could give an abbreviated
3762 form, whereas the current host's name might or might not be abbreviated.
3763 The same fix has been applied.
3769 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3770 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3773 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3774 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3776 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3778 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3779 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3780 Also, for "Auto-Submitted", treat anything other than "no" as
3781 non-personal, in accordance with RFC 3834. (Previously it treated
3782 anything starting "auto-" as non-personal.)
3784 TF/01 The control=submission/name=... option had a problem with syntax
3785 errors if the name included a slash character. The /name= option
3786 now slurps the rest of the string, so it can include any characters
3787 but it must come last in the list of options (after /sender_retain
3790 PH/05 Some modifications to the interface to the fake nameserver for the new
3798 TK/01 Added the "success_on_redirect" address verification option. See
3799 NewStuff for rationale and an example.
3801 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3803 PH/02 Patch to exigrep to allow it to work on syslog lines.
3805 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3806 fread() to skip over the body file's header line, because in Cygwin the
3807 header line is locked and is inaccessible.
3809 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3810 co-exist for some time) to make it clear that it is the Exim ID that is
3811 referenced, not the Message-ID: header line.
3813 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3814 string_format() function, because snprintf() does not exist on all
3817 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3818 ${stat: expansion item.
3820 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3821 protocol synchronization error", to keep the pedants happy.
3823 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3824 well as for IRIX systems, when gcc is being used. See the host.c source
3827 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3829 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3832 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3833 transport and to the smtp transport in LMTP mode.
3835 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3837 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3838 run a filter in a subprocess. This could lead to confusion in subsequent
3839 lookups in the parent process. There should also be a search_tidyup() at
3840 the end of the subprocess.
3842 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3843 only if the host matched helo_try_verify_hosts, which caused the
3844 verification to occur when the EHLO/HELO command was issued. The ACL just
3845 tested the remembered result. Now, if a previous verification attempt has
3846 not happened, "verify = helo" does it there and then.
3848 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3850 TK/03 Fix log output including CR from clamd.
3852 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3853 exist provoked a memory error which could cause a segfault.
3855 PH/15 Installed PCRE 6.2
3857 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3859 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3860 of the problem. Specifically, suggested +O2 rather than +O1 for the
3863 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3865 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3866 record, the port was not correctly logged when the outgoing_port log
3867 selector was set (it logged the transort's default port).
3869 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3870 fallback_hosts, and "hosts" in the smtp transport.
3872 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3873 host errors such as "Connection refused".
3875 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3876 authentication with radiusclient 0.4.9:
3878 - Error returned from rc_read_config was caught wrongly
3879 - Username/password not passed on to radius server due to wrong length.
3881 The presumption is that some radiusclient API changes for 4.51/PH/17
3882 were not taken care of correctly. The code is still untested by me (my
3883 Linux distribution still has 0.3.2 of radiusclient), but it was
3884 contributed by a Radius user.
3886 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3887 expanding the "port" option of the smtp transport.
3889 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3890 while reading a MIME header. Thanks to Tom Hughes for a patch.
3892 PH/24 Include config.h inside local_scan.h so that configuration settings are
3895 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3896 that in an ACL for RCPT (for example), you can examine exactly what was
3899 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3900 commands, but it was not correctly comparing the address with the actual
3901 client host address. Thus, it would show the EHLO address in Received:
3902 header lines when this was not necessary.
3904 PH/27 Added the % operator to ${eval:}.
3906 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3907 it should be ignoring failures (because with -C, for example, it has lost
3908 privilege). It wasn't ignoring creation failures other than "already
3911 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3914 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3915 because an input line was too long, either on its own, or by virtue of
3916 too many continuations, the temporary file was not being removed, and the
3917 return code was incorrect.
3919 PH/31 Missing "BOOL" in function definition in filtertest.c.
3921 PH/32 Applied Sieve patches from the maintainer.
3923 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3925 PH/33 Added "verify = not_blind".
3927 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3928 Local/Makefile (with some defaults set). These are used in built scripts
3929 such as exicyclog, but they have never been used in the exim_install
3930 script (though there are many overriding facilities there). I have
3931 arranged that the exim_install script now takes note of these two
3934 PH/35 Installed configuration files for Dragonfly.
3936 PH/36 When a locally submitted message by a trusted user did not contain a
3937 From: header, and the sender address was obtained from -f or from an SMTP
3938 MAIL command, and the trusted user did not use -F to supply a sender
3939 name, $originator_name was incorrectly used when constructing a From:
3940 header. Furthermore, $originator_name was used for submission mode
3941 messages from external hosts without From: headers in a similar way,
3942 which is clearly wrong.
3944 PH/37 Added control=suppress_local_fixups.
3946 PH/38 When log_selector = +received_sender was set, and the addition of the
3947 sender made the log line's construction buffer exactly full, or one byte
3948 less than full, an overflow happened when the terminating "\n" was
3951 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3952 when the result of a list match is failure because a DNS lookup failed.
3954 PH/40 RM_COMMAND is now used in the building process.
3956 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3957 the "build-* directories that it finds.
3959 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3960 address in a domain literal was a prefix of an interface address.
3962 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3963 when verifying a sender address, unless rewrite_headers is false.
3965 PH/44 Wrote a long comment about why errors_to addresses are verified as
3966 recipients, not senders.
3968 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3969 the ratelimit ACL was added.
3971 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3973 PH/46 Added extra information about PostgreSQL errors to the error string.
3975 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3976 avoiding the need to install special zones in a real server. This is
3977 backwards compatible; if it can't find the fake resolver, it drops back.
3978 Thus, both old and new test suites can be run.
3980 TF/02 Added util/ratelimit.pl
3982 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3983 clock is set back in time.
3985 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3986 Candler <B.Candler@pobox.com>.
3988 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3989 OK for addresses that are the result of redirections.
3991 PH/48 A number of further additions for the benefit of the new test suite,
3992 including a fake gethostbyname() that interfaces to the fake DNS resolver
3995 TF/06 The fix for widen_domains has also been applied to qualify_single and
3996 search_parents which are the other dnslookup options that can cause
3999 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
4002 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
4004 TF/07 Exim produced the error message "an SRV record indicated no SMTP
4005 service" if it encountered an MX record with an empty target hostname.
4006 The message is now "an MX or SRV record indicated no SMTP service".
4008 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
4009 if the DNS of the sending site is misconfigured. This is quite a
4010 common situation. This change restores the behaviour of treating a
4011 helo verification defer as a failure.
4013 PH/51 If self=fail was set on a router, the bounce message did not include the
4014 actual error message.
4020 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
4022 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
4023 command itself ended in error, the underlying message about the transport
4024 filter timeout was being overwritten with the pipe command error. Now the
4025 underlying error message should be appended to the second error message.
4027 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4029 PH/02 Reduce the amount of output that "make" produces by default. Full output
4030 can still be requested.
4032 PH/03 The warning log line about a condition test deferring for a "warn" verb
4033 was being output only once per connection, rather than after each
4034 occurrence (because it was using the same function as for successful
4035 "warn" verbs). This seems wrong, so I have changed it.
4037 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4038 it should not have, which might have caused a crash in the right
4039 circumstances, but probably never did.
4041 PH/04 Installed a modified version of Tony Finch's patch to make submission
4042 mode fix the return path as well as the Sender: header line, and to
4043 add a /name= option so that you can make the user's friendly name appear
4046 TF/03 Added the control = fakedefer ACL modifier.
4048 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4049 Mark Lowes for thorough testing.
4051 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4053 TK/03 Merged latest SRS patch from Miles Wilton.
4055 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4056 with the definition in sysexits.h (which is #included earlier).
4057 Fortunately, Exim does not actually use EX_OK. The code used to try to
4058 preserve the sysexits.h value, by assuming that macro definitions were
4059 scanned for macro replacements. I have been disabused of this notion,
4060 so now the code just undefines EX_OK before #including unistd.h.
4062 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4063 in the smtp transport. When a block could not be written in a single
4064 write() function, the timeout was being re-applied to each part-write.
4065 This seems wrong - if the receiver was accepting one byte at a time it
4066 would take for ever. The timeout is now adjusted when this happens. It
4067 doesn't have to be particularly precise.
4069 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4070 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4072 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4073 without a domain if the check to <postmaster@domain> fails.
4075 SC/01 Eximstats: added -xls and the ability to specify output files
4076 (patch written by Frank Heydlauf).
4078 SC/02 Eximstats: use FileHandles for outputting results.
4080 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4082 SC/04 Eximstats: fixed display of large numbers with -nvr option
4084 SC/05 Eximstats: fixed merging of reports with empty tables.
4086 SC/06 Eximstats: added the -include_original_destination flag
4088 SC/07 Eximstats: removed tabs and trailing whitespace.
4090 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4092 TK/06 MBOX spool code: Add real "From " MBOX separator line
4093 so the .eml file is really in mbox format (even though
4094 most programs do not really care). Patch from Alex Miller.
4096 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4097 The latter is generated from $received_to and is only set if the
4098 message has one envelope recipient. SA can use these headers,
4099 obviously out-of-the-box. Patch from Alex Miller.
4101 PH/08 The ${def test on a variable was returning false if the variable's
4102 value was "0", contrary to what the specification has always said!
4103 The result should be true unless the variable is empty.
4105 PH/09 The syntax error of a character other than { following "${if
4106 def:variable_name" (after optional whitespace) was not being diagnosed.
4107 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4108 accidental colon was present, for example, could give incorrect results.
4110 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4111 result is used (not including appendfile, where other changes are about
4114 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4115 This involved changing a lot of size variables from int to off_t. It
4116 should work with maildirs and everything.
4118 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4119 spamd dying while we are connected to it.
4121 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4124 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4125 function for BDB 4.3.
4127 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4129 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4130 the sources for PCRE have been re-arranged and more files are now
4133 PH/15 The code I had for printing potentially long long variables in PH/11
4134 above was not the best (it lost precision). The length of off_t variables
4135 is now inspected at build time, and an appropriate printing format (%ld
4136 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4137 to be "long long int" or "long int". This is needed for the internal
4138 formatting function string_vformat().
4140 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4141 the configuration file to be ":syslog", then the script "guesses" where
4142 the logs files are, rather than using the compiled in default. In our
4143 case the guess is not the same as the compiled default, so the script
4144 suddenly stopped working when I started to use syslog. The patch checks
4145 to see if log_file_path is "". If so, it attempts to read it from exim
4146 with no configuration file to get the compiled in version, before it
4147 falls back to the previous guessing code."
4149 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4150 implementing BATV in an Exim configuration. See NewStuff for the gory
4153 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4154 Makefile that are specific to HP-UX.
4156 PH/18 If the "use_postmaster" option was set for a recipient callout together
4157 with the "random" option, the postmaster address was used as the MAIL
4158 FROM address for the random test, but not for the subsequent recipient
4159 test. It is now used for both.
4161 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4162 patch removes a few documentation additions to RFC 3028, because the
4163 latest draft now contains them. It adds the new en;ascii-case comparator
4164 and a new error check for 8bit text in MIME parts. Comparator and
4165 require names are now matched exactly. I enabled the subaddress
4166 extension, but it is not well tested yet (read: it works for me)."
4168 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4169 rework some of the code of TK/09 above to avoid the hardwired use of
4170 "%lld" and "long long". Replaced the call to snprintf() with a call to
4173 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4174 records point to non-existent hosts", "retry timeout exceeded", and
4175 "retry time not reached for any host after a long failure period".
4177 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4178 experimental DomainKeys support:
4180 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4181 (2) On an error such as an illegally used "control", the wrong name for
4182 the control was given.
4184 These problems did NOT occur unless DomainKeys support was compiled.
4186 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4188 PH/24 Added ${if match_ip condition.
4190 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4191 databases so that it will be absolutely obvious if a crash occurs in the
4192 DB library. This is a regular occurrence (often caused by mis-matched
4195 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4196 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4197 picked up on a user's system that detects such things. There doesn't seem
4198 to be a gcc warning option for this - only an attribute that has to be
4199 put on the function's prototype. It seems that in Fedora Core 4 they have
4200 set this on a number of new functions. No doubt there will be more in due
4203 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4204 specify a transport. However, if an address that was verified by such a
4205 router was the subject of a callout, Exim crashed because it tried to
4206 read the rcpt_include_affixes from the non-existent transport. Now it
4207 just assumes that the setting of that option is false. This bug was
4208 introduced by 4.51/PH/31.
4210 PH/28 Changed -d+all to exclude +memory, because that information is very
4211 rarely of interest, but it makes the output a lot bigger. People tend to
4212 do -d+all out of habit.
4214 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4215 code in os-type was giving problems when libc.so lives in lib64, like on
4218 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4219 aren't the modern standard, and it seems that some systems' include files
4220 don't always have them. Exim was already checking for some of the newer
4221 ones like T_AAAA, and defining it itself. I've added checks for all the
4222 record types that Exim uses.
4224 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4225 not automatically generating a new one, as it is supposed to. This
4226 prevented TLS from working. If the file did exist, but contained invalid
4227 data, a new version was generated, as expected. It was only the case of a
4228 non-existent file that was broken.
4230 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4231 with a change in libdomainkeys > 0.64.
4233 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4234 from DNS. If the selector record carries the flag, it now has
4235 precedence over the domain-wide flag.
4237 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4239 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4240 the use of a port name that isn't defined in /etc/services) occurred, the
4241 message was deferred as in a normal delivery, and thus remained on the
4242 spool, instead of being failed because of the mua_wrapper setting. This
4243 is now fixed, and I tidied up some of the mua_wrapper messages at the
4246 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4247 the messages in a hash of arrays rather than using individual hashes.
4248 This is a bit cleaner and results in dramatic memory savings, albeit
4249 at a slight CPU cost.
4251 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4252 as requested by Marc Sherman.
4254 SC/10 Eximstats: added histograms for user specified patterns as requested
4257 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4259 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4260 fopen() in the content-scanning modules that did not already have it.
4266 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4267 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4269 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4271 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4273 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4274 to test Sieve filters that use "vacation".
4276 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4277 that changes the way the GnuTLS parameters are stored in the cache file.
4278 The new format can be generated externally. For backward compatibility,
4279 if the data in the cache doesn't make sense, Exim assumes it has read an
4280 old-format file, and it generates new data and writes a new file. This
4281 means that you can't go back to an older release without removing the
4284 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4285 work if there are any delivery delays because "one_time" forces the
4286 parent to be marked "delivered", so its unseen clone is never tried
4287 again. For this reason, Exim now forbids the simultaneous setting of
4290 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4291 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4292 entry if you want to know the details. The fix, however, neglected to
4293 consider the case when local delivery batching is involved. The test for
4294 "previously delivered" was not happening when checking to see if an
4295 address could be batched with a previous (undelivered) one; under
4296 certain circumstances this could lead to multiple deliveries to the same
4299 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4300 in its include files, and this causes problems building Exim.
4302 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4303 header_syntax) but Exim was just ignoring anything given after a slash.
4304 In particular, this caused confusion with an attempt to use "verify =
4305 reverse_host_lookup/defer_ok". An error is now given when options are
4306 supplied for verify items that do not have them. (Maybe reverse_host_
4307 lookup should have a defer_ok option, but that's a different point.)
4309 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4310 defined by RFC 821) to 2048, because there were problems with some AUTH
4311 commands, and RFC 1869 says the size should be increased for extended
4312 SMTP commands that take arguments.
4314 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4317 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4318 "unknown" error; now it says that the functionality isn't in the binary.
4320 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4321 an address' error message when a string expansion fails (syntax or
4322 whatever). Otherwise the password may appear in the log. Following change
4323 PH/42 below, there is no longer a chance of it appearing in a bounce
4326 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4328 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4329 of its addresses was ever tried. (Bugzilla bug #2.)
4331 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4332 the result incorrectly in the debug output. (It correctly added a newline
4333 to what was transported.)
4335 TF/01 Added $received_time.
4337 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4338 commented out examples of how to interface to a virus scanner and to
4339 SpamAssassin. Also added commented examples of av_scanner and
4340 spamd_address settings.
4342 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4343 and controls are allowed in which ACLs. There were a couple of minor
4344 errors. Some of the entries in the conditions table (which is a table of
4345 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4346 negation of where the condition IS allowed.
4348 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4350 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4351 header file does not have a version number, so I've had to invent a new
4352 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4353 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4354 radiusclient), but it was contributed by a Radius user.
4356 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4357 files or directories, for OpenSSL.
4359 PH/19 When an Exim process that is running as root has to create an Exim log
4360 file, it does so in a subprocess that runs as exim:exim so as to get the
4361 ownership right at creation (otherwise, other Exim processes might see
4362 the file with the wrong ownership). There was no test for failure of this
4363 fork() call, which would lead to the process getting stuck as it waited
4364 for a non-existent subprocess. Forks do occasionally fail when resources
4365 run out. I reviewed all the other calls to fork(); they all seem to check
4368 PH/20 When checking for unexpected SMTP input at connect time (before writing
4369 the banner), Exim was not dealing correctly with a non-positive return
4370 from the read() function. If the client had disconnected by this time,
4371 the result was a log entry for a synchronization error with an empty
4372 string after "input=" when read() returned zero. If read() returned -1
4373 (an event I could not check), uninitialized data bytes were printed.
4374 There were reports of junk text (parts of files, etc) appearing after
4377 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4379 PH/22 Added support for macro redefinition, and (re)definition in between
4380 driver and ACL definitions.
4382 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4383 forgetting to use the resulting value; it was using the unexpanded value.
4385 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4386 hadn't been configured. The fix is from Juergen Kreileder, who
4387 understands it better than I do:
4389 "Here's what I see happening with three configured cyrus_sasl
4390 authenticators configured (plain, login, cram-md5):
4392 On startup auth_cyrus_sasl_init() gets called for each of these.
4393 This means three calls to sasl_listmech() without a specified mech_list.
4394 => SASL tests which mechs of all available mechs actually work
4395 => three warnings about OTP not working
4396 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4398 With the patch, sasl_listmech() also gets called three times. But now
4399 SASL's mech_list option is set to the server_mech specified in the the
4400 authenticator. Or in other words, the answer from sasl_listmech()
4401 gets limited to just the mech you're testing for (which is different
4403 => the return list contains just 'plain' or 'login', 'cram-md5' or
4404 nothing depending on the value of ob->server_mech.
4406 I've just tested the patch: Authentication still works fine,
4407 unavailable mechs specified in the exim configuration are still
4408 caught, and the auth.log warnings about OTP are gone."
4410 PH/25 When debugging is enabled, the contents of the command line are added
4411 to the debugging output, even when log_selector=+arguments is not
4414 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4415 answer is "GNU", and only if the return is "GNU/something" is the answer
4418 PH/27 $acl_verify_message is now set immediately after the failure of a
4419 verification in an ACL, and so is available in subsequent modifiers. In
4420 particular, the message can be preserved by coding like this:
4422 warn !verify = sender
4423 set acl_m0 = $acl_verify_message
4425 Previously, $acl_verify_message was set only while expanding "message"
4426 and "log_message" when a very denied access.
4428 PH/28 Modified OS/os.c-Linux with
4430 -#ifndef OS_LOAD_AVERAGE
4431 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4433 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4434 nomenclature these days.)
4436 PH/29 Installed patch from the Sieve maintainer that adds the options
4437 sieve_useraddress and sieve_subaddress to the redirect router.
4439 PH/30 In these circumstances:
4440 . Two addresses routed to the same list of hosts;
4441 . First host does not offer TLS;
4442 . First host accepts first address;
4443 . First host gives temporary error to second address;
4444 . Second host offers TLS and a TLS session is established;
4445 . Second host accepts second address.
4446 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4447 and peerdn, if requested) that were in fact used only for the second
4450 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4451 attention to any local part prefix or suffix that was matched by the
4452 router that accepted the address. It now behaves in the same way as it
4453 does for delivery: the affixes are removed from the local part unless
4454 rcpt_include_affixes is set on the transport.
4456 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4457 timeout during the DATA phase of an incoming message.
4459 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4460 applied a patch sanctioned by the Sieve maintainer.
4462 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4463 the uid or gid is negative. A case of a negative gid caused this to be
4464 noticed. The fix allows for either to be negative.
4466 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4467 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4470 PH/36 The change PH/12 above was broken. Fixed it.
4472 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4473 the grounds that routing the same address twice would always produce the
4474 same answer. This might have been true once, but it is certainly no
4475 longer true now. Routing a child address may depend on the previous
4476 routing that produced that child. Some complicated redirection strategies
4477 went wrong when messages had multiple recipients, and made Exim's
4478 behaviour dependent on the order in which the addresses were given.
4480 I have moved the duplicate checking until after the routing is complete.
4481 Exim scans the addresses that are assigned to local and remote
4482 transports, and removes any duplicates. This means that more work will be
4483 done, as duplicates will always all be routed, but duplicates are
4484 presumably rare, so I don't expect this is of any significance.
4486 For deliveries to pipes, files, and autoreplies, the duplicate checking
4487 still happens during the routing process, since they are not going to be
4490 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4491 It corrects a timeout issue with spamd. This is Ian's comment: "The
4492 background is that sometimes spamd either never reads data from a
4493 connection it has accepted, or it never writes response data. The exiscan
4494 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4495 blindly assumes that writes won't block so it may never time out."
4497 PH/39 Allow G after quota size as well as K and M.
4499 PH/40 The value set for $authenticated_id in an authenticator may not contain
4500 binary zeroes or newlines because the value is written to log lines and
4501 to spool files. There was no check on this. Now the value is run through
4502 the string_printing() function so that such characters are converted to
4503 printable escape sequences.
4505 PH/41 $message_linecount is a new variable that contains the total number of
4506 lines in the message. Compare $body_linecount, which is the count for the
4509 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4510 bounce and delay warning messages, except in certain special cases, which
4513 (a) An SMTP error message from a remote host;
4514 (b) A message specified in a :fail: redirection;
4515 (c) A message specified in a "fail" command in a system filter;
4516 (d) A message specified in a FAIL return from the queryprogram router;
4517 (e) A message specified by the cannot_route_message router option.
4519 In these cases only, Exim does include the error details in bounce and
4520 warning messages. There are also a few cases where bland messages such
4521 as "unrouteable address" or "local delivery error" are given.
4523 PH/43 $value is now also set for the "else" part of a ${run expansion.
4525 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4526 being worked on, but at least Exim now implements the latest version to
4529 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4530 process to complete was treated as a delivery failure, a timeout while
4531 writing the message to the pipe was logged, but erroneously treated as a
4532 successful delivery. Such timeouts include transport filter timeouts. For
4533 consistency with the overall process timeout, these timeouts are now
4534 treated as errors, giving rise to delivery failures by default. However,
4535 there is now a new Boolean option for the pipe transport called
4536 timeout_defer, which, if set TRUE, converts the failures into defers for
4537 both kinds of timeout. A transport filter timeout is now identified in
4540 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4541 systems where "make" and "gmake" are different, calling "gmake" at top
4542 level broke things. I've arranged for the value of $(MAKE) to be passed
4543 from the Makefile to this script so that it can call the same version of
4547 A note about Exim versions 4.44 and 4.50
4548 ----------------------------------------
4550 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4551 changes of various kinds. As a consequence, a big documentation update was
4552 needed. This delayed the release for rather longer than seemed good, especially
4553 in the light of a couple of (minor) security issues. Therefore, the changes
4554 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4555 release. So 4.44 and 4.50 are in effect two different branches that both start
4558 I have left the 4.50 change log unchanged; it contains all the changes since
4559 4.43. The change log for 4.44 is below; many of its items are identical to
4560 those for 4.50. This seems to be the most sensible way to preserve the
4561 historical information.
4567 1. Minor wording change to the doc/README.SIEVE file.
4569 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4570 computation of the current number of files was incorrect.
4572 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4573 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4576 4. Give more explanation in the error message when the command for a transport
4577 filter fails to execute.
4579 5. There are several places where Exim runs a non-Exim command in a
4580 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4581 was being done only for the command run by the queryprogram router. It is
4582 now done for all such subprocesses. The other cases are: ${run, transport
4583 filters, and the commands run by the lmtp and pipe transports.
4585 6. Added CONFIGURE_GROUP build-time option.
4587 7. Some older OS have a limit of 256 on the maximum number of file
4588 descriptors. Exim was using setrlimit() to set 1000 as a large value
4589 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4590 systems. I've change it so that if it can't get 1000, it tries for 256.
4592 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4593 was an oversight, and furthermore, ever since the addition of extra
4594 controls (e.g. 4.43/32), the checks on when to allow different forms of
4595 "control" were broken. There should now be diagnostics for all cases when a
4596 control that does not make sense is encountered.
4598 9. Added the /retain_sender option to "control=submission".
4600 10. $recipients is now available in the predata ACL (oversight).
4602 11. Tidy the search cache before the fork to do a delivery from a message
4603 received from the command line. Otherwise the child will trigger a lookup
4604 failure and thereby defer the delivery if it tries to use (for example) a
4605 cached ldap connection that the parent has called unbind on.
4607 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4608 of $address_data from the recipient verification was clobbered by the
4609 sender verification.
4611 13. The value of address_data from a sender verification is now available in
4612 $sender_address_data in subsequent conditions in the ACL statement.
4614 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4616 15. Added a new option "connect=<time>" to callout options, to set a different
4619 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4620 was its contents. (It was OK if the option was not defined at all.)
4622 17. A "Completed" log line is now written for messages that are removed from
4623 the spool by the -Mrm option.
4625 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4626 information about exactly what failed.
4628 19. Added -dd to debug only the daemon process.
4630 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4631 handles timeouts, both on the server side and network timeouts. Renamed the
4632 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4634 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4635 It is now set to "smtps".
4637 22. $host_address is now set to the target address during the checking of
4638 ignore_target_hosts.
4640 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4641 being passed; this would have caused $sender_host_name to have been used if
4642 matching the list had actually called for a host name (not very likely,
4643 since this list is usually IP addresses). A host name is now passed as
4646 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4647 code that specifies a non-restarting handler (typically sigaction() in
4648 modern systems) in an attempt to fix a rare and obscure crash bug.
4650 25. Narrowed the window for a race in the daemon that could cause it to ignore
4651 SIGCHLD signals. This is not a major problem, because they are used only to
4652 wake it up if nothing else does.
4654 26. A malformed maildirsize file could cause Exim to calculate negative values
4655 for the mailbox size or file count. Odd effects could occur as a result.
4656 The maildirsize information is now recalculated if the size or filecount
4659 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4660 support for a long time. Removed HAVE_SYS_VFS_H.
4662 28. Installed the latest version of exipick from John Jetmore.
4664 29. In an address list, if the pattern was not a regular expression, an empty
4665 subject address (from a bounce message) matched only if the pattern was an
4666 empty string. Non-empty patterns were not even tested. This was the wrong
4667 because it is perfectly reasonable to use an empty address as part of a
4668 database query. An empty address is now tested by patterns that are
4669 lookups. However, all the other forms of pattern expect the subject to
4670 contain a local part and a domain, and therefore, for them, an empty
4671 address still always fails if the pattern is not itself empty.
4673 30. Exim went into a mad DNS loop when attempting to do a callout where the
4674 host was specified on an smtp transport, and looking it up yielded more
4675 than one IP address.
4677 31. Re-factored the code for checking spool and log partition space into a
4678 function that finds that data and another that does the check. The former
4679 is then used to implement four new variables: $spool_space, $log_space,
4680 $spool_inodes, and $log_inodes.
4682 32. The RFC2047 encoding function was originally intended for short strings
4683 such as real names; it was not keeping to the 75-character limit for
4684 encoded words that the RFC imposes. It now respects the limit, and
4685 generates multiple encoded words if necessary. To be on the safe side, I
4686 have increased the buffer size for the ${rfc2047: expansion operator from
4689 33. It is now permitted to omit both strings after an "if" condition; if the
4690 condition is true, the result is "true". As before, when the second string
4691 is omitted, a false condition yields an empty string. This makes it less
4692 cumbersome to write custom ACL and router conditions.
4694 34. Failure to deliver a bounce message always caused it to be frozen, even if
4695 there was an errors_to setting on the router. The errors_to setting is now
4698 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4699 canonical form (fully expanded) before being placed in
4700 $sender_host_address.
4702 36. The table in the code that translates DNS record types into text (T_A to
4703 "A" for instance) was missing entries for NS and CNAME. It is just possible
4704 that this could have caused confusion if both these types were looked up
4705 for the same domain, because the text type is used as part of Exim's
4706 per-process caching. But the chance of anyone hitting this buglet seems
4709 37. The dnsdb lookup has been extended in a number of ways.
4711 (1) There is a new type, "zns", which walks up the domain tree until it
4712 finds some nameserver records. It should be used with care.
4714 (2) There is a new type, "mxh", which is like "mx" except that it returns
4715 just the host names, not the priorities.
4717 (3) It is now possible to give a list of domains (or IP addresses) to be
4718 looked up. The behaviour when one of the lookups defers can be
4719 controlled by a keyword.
4721 (4) It is now possible to specify the separator character for use when
4722 multiple records are returned.
4724 38. The dnslists ACL condition has been extended: it is now possible to supply
4725 a list of IP addresses and/or domains to be looked up in a particular DNS
4728 39. Added log_selector=+queue_time_overall.
4730 40. When running the queue in the test harness, wait just a tad after forking a
4731 delivery process, to get repeatability of debugging output.
4733 41. Include certificate and key file names in error message when GnuTLS fails
4734 to set them up, because the GnuTLS error message doesn't include the name
4735 of the failing file when there is a problem reading it.
4737 42. Allow both -bf and -bF in the same test run.
4739 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4741 44. The "Exiscan patch" is now merged into the mainline Exim source.
4743 45. Sometimes the final signoff response after QUIT could fail to get
4744 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4745 < 0 before doing a fflush(). This bug looks as though it goes back to the
4746 introduction of TLS in release 3.20, but "sometimes" must have been rare
4747 because the tests only now provoked it.
4749 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4750 (this can affect the format of dates).
4752 47. exim_tidydb, when checking for the continued existence of a message for
4753 which it has found a message-specific retry record, was not finding
4754 messages that were in split spool directories. Consequently, it was
4755 deleting retry records that should have stayed in existence.
4757 48. Steve fixed some bugs in eximstats.
4759 49. The SPA authentication driver was not abandoning authentication and moving
4760 on to the next authenticator when an expansion was forced to fail,
4761 contradicting the general specification for all authenticators. Instead it
4762 was generating a temporary error. It now behaves as specified.
4764 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4765 (the order specifies the preference for clients). The order is now AES256,
4766 AES128, 3DES, ARCFOUR128.
4768 51. Small patch to Sieve code - explicitly set From: when generating an
4771 52. Exim crashed if a remote delivery caused a very long error message to be
4772 recorded - for instance if somebody sent an entire SpamAssassin report back
4773 as a large number of 550 error lines. This bug was coincidentally fixed by
4774 increasing the size of one of Exim's internal buffers (big_buffer) that
4775 happened as part of the Exiscan merge. However, to be on the safe side, I
4776 have made the code more robust (and fixed the comments that describe what
4779 53. Now that there can be additional text after "Completed" in log lines (if
4780 the queue_time_overall log selector is set), a one-byte patch to exigrep
4781 was needed to allow it to recognize "Completed" as not the last thing in
4784 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4785 patch that reportedly fixes this has been added. I am not expert enough to
4786 create a test for it. This is what the patch creator wrote:
4788 "I found a little strange behaviour of ldap code when working with
4789 Windows 2003 AD Domain, where users was placed in more than one
4790 Organization Units. When I tried to give exim partial DN, the exit code
4791 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4792 But simultaneously result of request was absolutely normal ldap result,
4793 so I produce this patch..."
4795 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4796 REFERENCE, so I have modified the code to exclude the patch when that macro
4799 55. Some experimental protocols are using DNS PTR records for new purposes. The
4800 keys for these records are domain names, not reversed IP addresses. The
4801 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4802 leaves it alone. Component reversal etc. now happens only for IP addresses.
4805 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4807 57. Double the size of the debug message buffer (to 2048) so that more of very
4808 long debug lines gets shown.
4810 58. The exicyclog utility now does better if the number of log files to keep
4811 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4813 59. Two changes related to the smtp_active_hostname option:
4815 (1) $smtp_active_hostname is now available as a variable.
4816 (2) The default for smtp_banner uses $smtp_active_hostname instead
4817 of $primary_hostname.
4819 60. The host_aton() function is supposed to be passed a string that is known
4820 to be a valid IP address. However, in the case of IPv6 addresses, it was
4821 not checking this. This is a hostage to fortune. Exim now panics and dies
4822 if the condition is not met. A case was found where this could be provoked
4823 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4824 components; fortuitously, this particular loophole had already been fixed
4825 by change 4.50/55 above.
4827 If there are any other similar loopholes, the new check in host_aton()
4828 itself should stop them being exploited. The report I received stated that
4829 data on the command line could provoke the exploit when Exim was running as
4830 exim, but did not say which command line option was involved. All I could
4831 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4832 running as the user.
4835 61. There was a buffer overflow vulnerability in the SPA authentication code
4836 (which came originally from the Samba project). I have added a test to the
4837 spa_base64_to_bits() function which I hope fixes it.
4840 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4841 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4843 63. The daemon start-up calls getloadavg() while still root for those OS that
4844 need the first call to be done as root, but it missed one case: when
4845 deliver_queue_load_max is set with deliver_drop_privilege. This is
4846 necessary for the benefit of the queue runner, because there is no re-exec
4847 when deliver_drop_privilege is set.
4849 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4850 This has been fixed.
4852 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4853 was in use, was not putting the data itself into the right store pool;
4854 consequently, it could be overwritten for a subsequent message in the same
4855 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4858 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4860 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4861 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4862 the function were treating the return as a boolean value, which happened to
4863 work because 0=false and not-0=true, but is not correct code.
4865 68. The host_aton() function was not handling scoped IPv6 addresses (those
4866 with, for example, "%eth0" on the end) correctly.
4868 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4869 negated items (that is, ~something) in unsigned ints. Some compilers
4870 apparently mutter when there is no cast.
4872 70. If an address verification called from an ACL failed, and did not produce a
4873 user-specific message (i.e. there was only a "system" message), nothing was
4874 put in $acl_verify_message. In this situation, it now puts the system
4877 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4878 session; change 4.31/43 added the unwanted input to the log line - except
4879 that it did not do this in the start of session case. It now does.
4881 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4882 This is wrong and can cause the other end to generate a synchronization
4883 error if it is another Exim or anything else that does the synchronization
4884 check. A QUIT command is no longer sent after a timeout.
4886 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4887 during host lookups.
4889 74. The defer_ok option of callout verification was not working if it was used
4890 when verifying addresses in header lines, that is, for this case:
4892 verify = header_sender/callout=defer_ok
4894 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4895 those file descriptors could be used for SMTP connections. If anything
4896 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4897 could be sent to the SMTP client, causing chaos. The daemon now opens
4898 stdin, stdout, and stderr to /dev/null when it puts itself into the
4901 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4902 log by default. The user can override this with suitable Perl magic.
4904 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4905 the log message when discard triggers, was not working for the DATA ACL or
4906 for the non-SMTP ACL.
4908 78. Error message wording change in sieve.c.
4910 79. If smtp_accept_max_per_host was set, the number of connections could be
4911 restricted to fewer than expected, because the daemon was trying to set up
4912 a new connection before checking whether the processes handling previous
4913 connections had finished. The check for completed processes is now done
4914 earlier. On busy systems, this bug wouldn't be noticed because something
4915 else would have woken the daemon, and it would have reaped the completed
4918 80. If a message was submitted locally by a user whose login name contained one
4919 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4920 It caused a spool format error. I have fixed the spool reading code. A
4921 related problem was that the "from" clause in the Received: line became
4922 illegal because of the space(s). It is now covered by ${quote_local_part.
4924 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4927 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4929 83. Patch from Sieve maintainer to fix unterminated string problem in
4930 "vacation" handling.
4932 84. Some minor changes to the Linux configuration files to help with other
4933 OS variants using glibc.
4935 85. One more patch for Sieve to update vacation handling to latest spec.
4938 ----------------------------------------------------
4939 See the note above about the 4.44 and 4.50 releases.
4940 ----------------------------------------------------
4946 1. Change 4.43/35 introduced a bug that caused file counts to be
4947 incorrectly computed when quota_filecount was set in an appendfile
4950 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4951 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4954 3. Give more explanation in the error message when the command for a transport
4955 filter fails to execute.
4957 4. There are several places where Exim runs a non-Exim command in a
4958 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4959 was being done only for the command run by the queryprogram router. It is
4960 now done for all such subprocesses. The other cases are: ${run, transport
4961 filters, and the commands run by the lmtp and pipe transports.
4963 5. Some older OS have a limit of 256 on the maximum number of file
4964 descriptors. Exim was using setrlimit() to set 1000 as a large value
4965 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4966 systems. I've change it so that if it can't get 1000, it tries for 256.
4968 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4969 was an oversight, and furthermore, ever since the addition of extra
4970 controls (e.g. 4.43/32), the checks on when to allow different forms of
4971 "control" were broken. There should now be diagnostics for all cases when a
4972 control that does not make sense is encountered.
4974 7. $recipients is now available in the predata ACL (oversight).
4976 8. Tidy the search cache before the fork to do a delivery from a message
4977 received from the command line. Otherwise the child will trigger a lookup
4978 failure and thereby defer the delivery if it tries to use (for example) a
4979 cached ldap connection that the parent has called unbind on.
4981 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4982 of $address_data from the recipient verification was clobbered by the
4983 sender verification.
4985 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4986 was its contents. (It was OK if the option was not defined at all.)
4988 11. A "Completed" log line is now written for messages that are removed from
4989 the spool by the -Mrm option.
4991 12. $host_address is now set to the target address during the checking of
4992 ignore_target_hosts.
4994 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4995 being passed; this would have caused $sender_host_name to have been used if
4996 matching the list had actually called for a host name (not very likely,
4997 since this list is usually IP addresses). A host name is now passed as
5000 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5001 code that specifies a non-restarting handler (typically sigaction() in
5002 modern systems) in an attempt to fix a rare and obscure crash bug.
5004 15. Narrowed the window for a race in the daemon that could cause it to ignore
5005 SIGCHLD signals. This is not a major problem, because they are used only to
5006 wake it up if nothing else does.
5008 16. A malformed maildirsize file could cause Exim to calculate negative values
5009 for the mailbox size or file count. Odd effects could occur as a result.
5010 The maildirsize information is now recalculated if the size or filecount
5013 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5014 support for a long time. Removed HAVE_SYS_VFS_H.
5016 18. Updated exipick to current release from John Jetmore.
5018 19. Allow an empty sender to be matched against a lookup in an address list.
5019 Previously the only cases considered were a regular expression, or an
5022 20. Exim went into a mad DNS lookup loop when doing a callout where the
5023 host was specified on the transport, if the DNS lookup yielded more than
5026 21. The RFC2047 encoding function was originally intended for short strings
5027 such as real names; it was not keeping to the 75-character limit for
5028 encoded words that the RFC imposes. It now respects the limit, and
5029 generates multiple encoded words if necessary. To be on the safe side, I
5030 have increased the buffer size for the ${rfc2047: expansion operator from
5033 22. Failure to deliver a bounce message always caused it to be frozen, even if
5034 there was an errors_to setting on the router. The errors_to setting is now
5037 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5038 canonical form (fully expanded) before being placed in
5039 $sender_host_address.
5041 24. Updated eximstats to version 1.33
5043 25. Include certificate and key file names in error message when GnuTLS fails
5044 to set them up, because the GnuTLS error message doesn't include the name
5045 of the failing file when there is a problem reading it.
5047 26. Expand error message when OpenSSL has problems setting up cert/key files.
5050 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5051 (this can affect the format of dates).
5053 28. exim_tidydb, when checking for the continued existence of a message for
5054 which it has found a message-specific retry record, was not finding
5055 messages that were in split spool directories. Consequently, it was
5056 deleting retry records that should have stayed in existence.
5058 29. eximstats updated to version 1.35
5059 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5060 1.35 - bugfix such that pie charts by volume are generated correctly
5062 30. The SPA authentication driver was not abandoning authentication and moving
5063 on to the next authenticator when an expansion was forced to fail,
5064 contradicting the general specification for all authenticators. Instead it
5065 was generating a temporary error. It now behaves as specified.
5067 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5068 (the order specifies the preference for clients). The order is now AES256,
5069 AES128, 3DES, ARCFOUR128.
5071 31. Small patch to Sieve code - explicitly set From: when generating an
5074 32. Exim crashed if a remote delivery caused a very long error message to be
5075 recorded - for instance if somebody sent an entire SpamAssassin report back
5076 as a large number of 550 error lines. This bug was coincidentally fixed by
5077 increasing the size of one of Exim's internal buffers (big_buffer) that
5078 happened as part of the Exiscan merge. However, to be on the safe side, I
5079 have made the code more robust (and fixed the comments that describe what
5082 33. Some experimental protocols are using DNS PTR records for new purposes. The
5083 keys for these records are domain names, not reversed IP addresses. The
5084 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5085 leaves it alone. Component reversal etc. now happens only for IP addresses.
5088 34. The host_aton() function is supposed to be passed a string that is known
5089 to be a valid IP address. However, in the case of IPv6 addresses, it was
5090 not checking this. This is a hostage to fortune. Exim now panics and dies
5091 if the condition is not met. A case was found where this could be provoked
5092 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5093 components; fortuitously, this particular loophole had already been fixed
5094 by change 4.50/55 or 4.44/33 above.
5096 If there are any other similar loopholes, the new check in host_aton()
5097 itself should stop them being exploited. The report I received stated that
5098 data on the command line could provoke the exploit when Exim was running as
5099 exim, but did not say which command line option was involved. All I could
5100 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5101 running as the user.
5104 35. There was a buffer overflow vulnerability in the SPA authentication code
5105 (which came originally from the Samba project). I have added a test to the
5106 spa_base64_to_bits() function which I hope fixes it.
5109 36. The daemon start-up calls getloadavg() while still root for those OS that
5110 need the first call to be done as root, but it missed one case: when
5111 deliver_queue_load_max is set with deliver_drop_privilege. This is
5112 necessary for the benefit of the queue runner, because there is no re-exec
5113 when deliver_drop_privilege is set.
5115 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5116 was in use, was not putting the data itself into the right store pool;
5117 consequently, it could be overwritten for a subsequent message in the same
5118 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5121 38. Sometimes the final signoff response after QUIT could fail to get
5122 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5123 < 0 before doing a fflush(). This bug looks as though it goes back to the
5124 introduction of TLS in release 3.20, but "sometimes" must have been rare
5125 because the tests only now provoked it.
5131 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5132 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5133 Now it returns an integer. A number of calls were still expecting a T/F
5134 return. Fortuitously, in all cases, the tests worked in OK situations,
5135 which is the norm. However, things would have gone wrong on any write
5136 failures on the smtp file descriptor. This function is used when sending
5137 messages over SMTP and also when doing verify callouts.
5139 2. When Exim is called to do synchronous delivery of a locally submitted
5140 message (the -odf or -odi options), it no longer closes stderr before doing
5143 3. Implemented the mua_wrapper option.
5145 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5147 5. Implemented the functions header_remove(), header_testname(),
5148 header_add_at_position(), and receive_remove_recipient(), and exported them
5151 6. If an ACL "warn" statement specified the addition of headers, Exim already
5152 inserted X-ACL-Warn: at the start if there was no header name. However, it
5153 was not making this test for the second and subsequent header lines if
5154 there were newlines in the string. This meant that an invalid header could
5155 be inserted if Exim was badly configured.
5157 7. Allow an ACL "warn" statement to add header lines at the start or after all
5158 the Received: headers, as well as at the end.
5160 8. Added the rcpt_4xx retry error code.
5162 9. Added postmaster_mailfrom=xxx to callout verification option.
5164 10. Added mailfrom=xxxx to the callout verification option, for verify=
5167 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5168 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5170 12. Inserted some casts to stop certain compilers warning when using pointer
5171 differences as field lengths or precisions in printf-type calls (mostly
5172 affecting debugging statements).
5174 13. Added optional readline() support for -be (dynamically loaded).
5176 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5177 same clock tick as a message's arrival, so that its received time was the
5178 same as the "first fail" time on the retry record, and that message
5179 remained on the queue past the ultimate address timeout, every queue runner
5180 would try a delivery (because it was past the ultimate address timeout) but
5181 after another failure, the ultimate address timeout, which should have then
5182 bounced the address, did not kick in. This was a "< instead of <=" error;
5183 in most cases the first failure would have been in the next clock tick
5184 after the received time, and all would be well.
5186 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5187 being recognized when the domain list was tested by the match_domain
5188 condition in an expansion string.
5190 16. Added the ${str2b64: operator.
5192 17. Exim was always calling setrlimit() to set a large limit for the number of
5193 processes, without checking whether the existing limit was already
5194 adequate. (It did check for the limit on file descriptors.) Furthermore,
5195 errors from getrlimit() and setrlimit() were being ignored. Now they are
5196 logged to the main and panic logs, but Exim does carry on, to try to do its
5197 job under whatever limits there are.
5199 18. Imported PCRE 5.0.
5201 19. Trivial typo in log message " temporarily refused connection" (the leading
5204 20. If the log selector return_path_on_delivery was set and an address was
5205 redirected to /dev/null, the delivery process crashed because it assumed
5206 that a return path would always be set for a "successful" delivery. In this
5207 case, the whole delivery is bypassed as an optimization, and therefore no
5210 21. Internal re-arrangement: the function for sending a challenge and reading
5211 a response while authentication was assuming a zero-terminated challenge
5212 string. It's now changed to take a pointer and a length, to allow for
5213 binary data in such strings.
5215 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5217 23. Exim was not respecting finduser_retries when seeking the login of the
5218 uid under which it was called; it was always trying 10 times. (The default
5219 setting of finduser_retries is zero.) Also, it was sleeping after the final
5220 failure, which is pointless.
5222 24. Implemented tls_on_connect_ports.
5224 25. Implemented acl_smtp_predata.
5226 26. If the domain in control=submission is set empty, Exim assumes that the
5227 authenticated id is a complete email address when it generates From: or
5228 Sender: header lines.
5230 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5231 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5232 chown and chgrp in /bin and hostname in /usr/bin.
5234 28. Exim was keeping the "process log" file open after each use, just as it
5235 does for the main log. This opens the possibility of it remaining open for
5236 long periods when the USR1 signal hits a daemon. Occasional processlog
5237 errors were reported, that could have been caused by this. Anyway, it seems
5238 much more sensible not to leave this file open at all, so that is what now
5241 29. The long-running daemon process does not normally write to the log once it
5242 has entered its main loop, and it closes the log before doing so. This is
5243 so that log files can straightforwardly be renamed and moved. However,
5244 there are a couple of unusual error situations where the daemon does write
5245 log entries, and I had neglected to close the log afterwards.
5247 30. The text of an SMTP error response that was received during a remote
5248 delivery was being truncated at 512 bytes. This is too short for some of
5249 the long messages that one sometimes sees. I've increased the limit to
5252 31. It is now possible to make retry rules that apply only when a message has a
5253 specific sender, in particular, an empty sender.
5255 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5256 it possible to be selective about when SMTP synchronization is enforced.
5258 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5260 32. Implemented hosts_connection_nolog.
5262 33. Added an ACL for QUIT.
5264 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5267 35. Added mailbox_size and mailbox_filecount to appendfile.
5269 36. Added control = no_multiline_responses to ACLs.
5271 37. There was a bug in the logic of the code that waits for the clock to tick
5272 in the case where the clock went backwards by a substantial amount such
5273 that the microsecond fraction of "now" was more than the microsecond
5274 fraction of "then" (but the whole seconds number was less).
5276 38. Added support for the libradius Radius client library this is found on
5277 FreeBSD (previously only the radiusclient library was supported).
5283 1. When certain lookups returned multiple values in the form name=value, the
5284 quoting of the values was not always being done properly. Specifically:
5285 (a) If the value started with a double quote, but contained no whitespace,
5287 (b) If the value contained whitespace other than a space character (i.e.
5288 tabs or newlines or carriage returns) it was not quoted.
5289 This fix has been applied to the mysql and pgsql lookups by writing a
5290 separate quoting function and calling it from the lookup code. The fix
5291 should probably also be applied to nisplus, ibase and oracle lookups, but
5292 since I cannot test any of those, I have not disturbed their existing code.
5294 2. A hit in the callout cache for a specific address caused a log line with no
5295 reason for rejecting RCPT. Now it says "Previous (cached) callout
5296 verification failure".
5298 3. There was an off-by-one bug in the queryprogram router. An over-long
5299 return line was truncated at 256 instead of 255 characters, thereby
5300 overflowing its buffer with the terminating zero. As well as fixing this, I
5301 have increased the buffer size to 1024 (and made a note to document this).
5303 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5304 when Exim is waiting for an SMTP response from a remote server, Exim
5305 restarts its select() call on the socket, thereby resetting its timeout.
5306 This is not a problem when such interrupts are rare. Somebody set up a cron
5307 job to run exiwhat every 2 minutes, which is less than the normal select()
5308 timeout (5 or 10 minutes). This meant that the select() timeout never
5309 kicked in because it was always reset. I have fixed this by comparing the
5310 time when an interrupt arrives with the time at the start of the first call
5311 to select(). If more time than the timeout has elapsed, the interrupt is
5312 treated as a timeout.
5314 5. Some internal re-factoring in preparation for the addition of Sieve
5315 extensions (by MH). In particular, the "personal" test is moved to a
5316 separate function, and given an option for scanning Cc: and Bcc: (which is
5317 not set for Exim filters).
5319 6. When Exim created an email address using the login of the caller as the
5320 local part (e.g. when creating a From: or Sender: header line), it was not
5321 quoting the local part when it contained special characters such as @.
5323 7. Installed new OpenBSD configuration files.
5325 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5326 try to make them clearer.
5328 9. Callout options, other than the timeout value, were being ignored when
5329 verifying sender addresses in header lines. For example, when using
5331 verify = header_sender/callout=no_cache
5333 the cache was (incorrectly) being used.
5335 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5336 only the Cygwin environment.
5338 11. When return_path_on_delivery was set as a log selector, if different remote
5339 addresses in the same message used different return paths and parallel
5340 remote delivery occurred, the wrong values would sometimes be logged.
5341 (Whenever a remote delivery process finished, the return path value from
5342 the most recently started remote delivery process was logged.)
5344 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5345 lines when AUTH and/or TLS are in use. This is the "received protocol"
5346 field. Exim used to use "asmtp" for authenticated SMTP, without any
5347 indication (in the protocol name) for TLS use. Now it follows the RFC and
5348 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5349 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5350 names appear in log lines as well as in Received: header lines.
5352 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5353 extensions, and comparison tests, and to fix some bugs.
5355 14. Changes to the "personal" filter test:
5357 (1) The test was buggy in that it was just doing the equivalent of
5358 "contains" tests on header lines. For example, if a user's address was
5359 anne@some.where, the "personal" test would incorrectly be true for
5361 To: susanne@some.where
5363 This test is now done by extracting each address from the header in turn,
5364 and checking the entire address. Other tests that are part of "personal"
5365 are now done using regular expressions (for example, to check local parts
5366 of addresses in From: header lines).
5368 (2) The list of non-personal local parts in From: addresses has been
5369 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5370 taken from the Sieve specification recommendations.
5372 (3) If the message contains any header line starting with "List-" it is
5373 treated as non-personal.
5375 (4) The test for "circular" in the Subject: header line has been removed
5376 because it now seems ill-conceived.
5378 15. Minor typos in src/EDITME comments corrected.
5380 16. Installed latest exipick from John Jetmore.
5382 17. If headers_add on a router specified a text string that was too long for
5383 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5384 of string_sprintf() is now avoided.
5386 18. $message_body_size was not set (it was always zero) when running the DATA
5387 ACL and the local_scan() function.
5389 19. For the "mail" command in an Exim filter, no default was being set for
5390 the once_repeat time, causing a random time value to be used if "once" was
5391 specified. (If the value happened to be <= 0, no repeat happened.) The
5392 default is now 0s, meaning "never repeat". The "vacation" command was OK
5393 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5394 (I found it when inspecting the code).
5396 20. There is now an overall timeout for performing a callout verification. It
5397 defaults to 4 times the callout timeout, which applies to individual SMTP
5398 commands during the callout. The overall timeout applies when there is more
5399 than one host that can be tried. The timeout is checked before trying the
5400 next host. This prevents very long delays if there are a large number of
5401 hosts and all are timing out (e.g. when the network connections are timing
5402 out). The value of the overall timeout can be changed by specifying an
5403 additional sub-option for "callout", called "maxwait". For example:
5405 verify = sender/callout=5s,maxwait=20s
5407 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5408 to the end before writing, but this should make it even safer).
5410 22. Exim was forgetting that it had advertised PIPELINING for the second and
5411 subsequent messages on an SMTP connection. It was also not resetting its
5412 memory on STARTTLS and an internal HELO.
5414 23. When Exim logs an SMTP synchronization error within a session, it now
5415 records whether PIPELINING has been advertised or not.
5417 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5418 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5419 rather than long int.
5421 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5423 26. Added the never_mail option to autoreply.
5429 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5430 crash if the getsockname() call failed; this can happen if a connection is
5431 closed very soon after it is established. The problem was simply in the
5432 order in which certain operations were done, causing Exim to try to write
5433 to the SMTP stream before it had set up the file descriptor. The bug has
5434 been fixed by making things happen in the correct order.
5440 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5441 before the connection was closed, thus losing the rejection response.
5443 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5444 some early Solaris releases, but causes trouble in current releases where
5445 socklen_t is defined.
5447 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5450 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5453 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5454 that releasing the top of it at the end releases what was used for sub-
5455 expansions (unless the block got too big). However, discard this block if
5456 the first thing is a variable or header, so that we can use its block when
5457 it is dynamic (useful for very large $message_headers, for example).
5459 6. Lookups now cache *every* query, not just the most recent. A new, separate
5460 store pool is used for this. It can be recovered when all lookup caches are
5461 flushed. Lookups now release memory at the end of their result strings.
5462 This has involved some general refactoring of the lookup sources.
5464 7. Some code has been added to the store_xxx() functions to reduce the amount
5465 of flapping under certain conditions.
5467 8. log_incoming_interface used to affect only the <= reception log lines. Now
5468 it causes the local interface and port to be added to several more SMTP log
5469 lines, for example "SMTP connection from", and rejection lines.
5471 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5473 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5475 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5476 could be overwritten at the end of the current message (or the start of a
5477 new message if it was set in a HELO ACL). The value is now preserved for
5478 the duration of the SMTP connection.
5480 12. If a transport had a headers_rewrite setting, and a matching header line
5481 contained an unqualified address, that address was qualified, even if it
5482 did not match any rewriting rules. The underlying bug was that the values
5483 of the flags that permit the existence of unqualified sender and recipient
5484 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5485 non-local messages, and by -bnq for local messages) were not being
5486 preserved with the message after it was received.
5488 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5489 "next input=" as part of the text comprising the host identity instead of
5490 the correct text. The code was using the same buffer for two different
5491 strings. However, depending on which order the printing function evaluated
5492 its arguments, the bug did not always show up. Under Linux, for example, my
5493 test suite worked just fine.
5495 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5496 doesn't work with some older versions of Perl. It has been changed to "my",
5497 which in any case is probably the better facility to use.
5499 15. A really picky compiler found some instances of statements for creating
5500 error messages that either had too many or two few arguments for the format
5503 16. The size of the buffer for calls to the DNS resolver has been increased
5504 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5505 for addresses that have a lot of PTR records. This alleviates a problem; it
5506 does not fully solve it.
5508 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5509 buffer now truncates the list and logs the incident, which is the same
5510 action as happens when Exim is looking up a host name and its aliases.
5511 Previously in this situation something unpredictable would happen;
5512 sometimes it was "internal error: store_reset failed".
5514 18. If a server dropped the connection unexpectedly when an Exim client was
5515 using GnuTLS and trying to read a response, the client delivery process
5516 crashed while trying to generate an error log message.
5518 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5519 string, for example:
5521 warn message = H1: something\nH2: something
5523 the text was added as a single header line from Exim's point of view
5524 though it ended up OK in the delivered message. However, searching for the
5525 second and subsequent header lines using $h_h2: did not work. This has been
5526 fixed. Similarly, if a system filter added multiple headers in this way,
5527 the routers could not see them.
5529 20. Expanded the error message when iplsearch is called with an invalid key to
5530 suggest using net-iplsearch in a host list.
5532 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5533 ACLs are no longer actually imposed (and a message to that effect is
5536 22. If a "gecos" field in a passwd entry contained escaped characters, in
5537 particular, if it contained a \" sequence, Exim got it wrong when building
5538 a From: or a Sender: header from that name. A second bug also caused
5539 incorrect handling when an unquoted " was present following a character
5540 that needed quoting.
5542 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5543 was not being matched caselessly.
5545 24. Arranged for all hyphens in the exim.8 source to be escaped with
5548 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5549 properly. Recipient callout cache records were still being keyed to include
5550 the sender, even when use_sender was set false. This led to far more
5551 callouts that were necessary. The sender is no longer included in the key
5552 when use_sender is false.
5554 26. Added "control = submission" modifier to ACLs.
5556 27. Added the ${base62d: operator to decode base 62 numbers.
5558 28. dnsdb lookups can now access SRV records.
5560 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5561 the configuration file.
5563 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5564 (-v) mode. This makes the output for a verbose queue run more intelligible.
5566 31. Added a use_postmaster feature to recipient callouts.
5568 32. Added the $body_zerocount variable, containing the number of binary zero
5569 bytes in the message body.
5571 33. The time of last modification of the "new" subdirectory is now used as the
5572 "mailbox time last read" when there is a quota error for a maildir
5575 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5577 35. Added +ignore_unknown as a special item in host lists.
5579 36. Code for decoding IPv6 addresses in host lists is now included, even if
5580 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5581 address was recognized as an IP address, but was then not correctly decoded
5582 into binary, causing unexpected and incorrect effects when compared with
5589 1. Very minor rewording of debugging text in manualroute to say "list of
5590 hosts" instead of "hostlist".
5592 2. If verify=header_syntax was set, and a header line with an unqualified
5593 address (no domain) and a large number of spaces between the end of the
5594 name and the colon was received, the reception process suffered a buffer
5595 overflow, and (when I tested it) crashed. This was caused by some obsolete
5596 code that should have been removed. The fix is to remove it!
5598 3. When running in the test harness, delay a bit after writing a bounce
5599 message to get a bit more predictability in the log output.
5601 4. Added a call to search_tidyup() just before forking a reception process. In
5602 theory, someone could use a lookup in the expansion of smtp_accept_max_
5603 per_host which, without the tidyup, could leave open a database connection.
5605 5. Added the variables $recipient_data and $sender_data which get set from a
5606 lookup success in an ACL "recipients" or "senders" condition, or a router
5607 "senders" option, similar to $domain_data and $local_part_data.
5609 6. Moved the writing of debug_print from before to after the "senders" test
5612 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5613 problems for message scanning, either using a data ACL, or using
5614 local_scan() because the Received: header was not generated till after they
5615 were called (in order to set the time as the time of reception completion).
5616 I have revised the way this works. The header is now generated after the
5617 body is received, but before the ACL or local_scan() are called. After they
5618 are run, the timestamp in the header is updated.
5624 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5625 before starting a queue runner without re-exec. This happened only when
5626 deliver_drop_privilege was set or when the Exim user was set to root. The
5627 effect of the bug was that timeouts during subsequent deliveries caused
5628 crashes instead of being properly handled. The handler is now left at its
5629 default (and expected) setting.
5631 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5632 message, again when deliver_drop_privilege is set or Exim is run as root.
5633 The bug described in (1) was not present in this case, but the tidying up
5634 of the other signals was missing. I have made the two cases consistent.
5636 3. The ignore_target_hosts setting on a manualroute router was being ignored
5637 for hosts that were looked up using the /MX notation.
5639 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5642 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5643 operated on the sender address. After changing the $sender_address to <>
5644 for the sender address verify, Exim was re-instated it as the original
5645 (before rewriting) address, but remembering that it had rewritten it, so it
5646 wasn't rewriting it again. This bug also had the effect of breaking the
5647 sender address verification caching when the sender address was rewritten.
5649 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5650 This has been changed so that if the ip literal address matches
5651 ignore_target_hosts, the router declines.
5653 7. Added expansion conditions match_domain, match_address, and match_local_
5654 part (NOT match_host).
5656 8. The placeholder for the Received: header didn't have a length field set.
5658 9. Added code to Exim itself and to exim_lock to test for a specific race
5659 condition that could lead to file corruption when using MBX delivery. The
5660 issue is with the lockfile that is created in /tmp. If this file is removed
5661 after a process has opened it but before that process has acquired a lock,
5662 there is the potential for a second process to recreate the file and also
5663 acquire a lock. This could lead to two Exim processes writing to the file
5664 at the same time. The added code performs the same test as UW imapd; it
5665 checks after acquiring the lock that its file descriptor still refers to
5666 the same named file.
5668 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5669 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5672 11. Added the smtp_active_hostname option. If used, this will typically be made
5673 to depend on the incoming interface address. Because $interface_address is
5674 not set up until the daemon has forked a reception process, error responses
5675 that can happen earlier (such as "too many connections") no longer contain
5678 12. If an expansion in a condition on a "warn" statement fails because a lookup
5679 defers, the "warn" statement is abandoned, and the next ACL statement is
5680 processed. Previously this caused the whole ACL to be aborted.
5682 13. Added the iplsearch lookup type.
5684 14. Added ident_timeout as a log selector.
5686 15. Added tls_certificate_verified as a log selector.
5688 16. Added a global option tls_require_ciphers (compare the smtp transport
5689 option of the same name). This controls incoming TLS connections.
5691 17. I finally figured out how to make tls_require_ciphers do a similar thing
5692 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5693 before starting the TLS session.
5695 18. Tabs are now shown as \t in -bP output.
5697 19. If the log selector return_path_on_delivery was set, Exim crashed when
5698 bouncing a message because it had too many Received: header lines.
5700 20. If two routers both had headers_remove settings, and the first one included
5701 a superfluous trailing colon, the final name in the first list and the
5702 first name in the second list were incorrectly joined into one item (with a
5703 colon in the middle).
5709 1. Added -C and -D options to the exinext utility, mainly to make it easier
5710 to include in the automated testing, but these could be helpful when
5711 multiple configurations are in use.
5713 2. The exinext utility was not formatting the output nicely when there was
5714 an alternate port involved in the retry record key, nor when there was a
5715 message id as well (for retries that were specific to a specific message
5716 and a specific host). It was also confused by IPv6 addresses, because of
5717 the additional colons they contain. I have fixed the IPv4 problem, and
5718 patched it up to do a reasonable job for IPv6.
5720 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5721 delivery, the log line now contains "pipelined" if PIPELINING was used.
5723 4. An SMTP transport process used to panic and die if the bind() call to set
5724 an explicit outgoing interface failed. This has been changed; it is now
5725 treated in the same way as a connect() failure.
5727 5. A reference to $sender_host_name in the part of a conditional expansion
5728 that was being skipped was still causing a DNS lookup. This no longer
5731 6. The def: expansion condition was not recognizing references to header lines
5732 that used bh_ and bheader_.
5734 7. Added the _cache feature to named lists.
5736 8. The code for checking quota_filecount in the appendfile transport was
5737 allowing one more file than it should have been.
5739 9. For compatibility with Sendmail, the command line option
5747 and sets the incoming protocol and host name (for trusted callers). The
5748 host name and its colon can be omitted when only the protocol is to be set.
5749 Note the Exim already has two private options, -pd and -ps, that refer to
5750 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5751 "s", but I don't think that's a major issue.
5753 10. A number of refactoring changes to the code, none of which should affect
5756 (a) The number of logging options was getting close to filling up the
5757 32-bit word that was used as a bit map. I have split them into two classes:
5758 those that are passed in the argument to log_write(), and those that are
5759 only ever tested independently outside of that function. These are now in
5760 separate 32-bit words, so there is plenty of room for expansion again.
5761 There is no change in the user interface or the logging behaviour.
5763 (b) When building, for example, log lines, the code previously used a
5764 macro that called string_cat() twice, in order to add two strings. This is
5765 not really sufficiently general. Furthermore, there was one instance where
5766 it was actually wrong because one of the argument was used twice, and in
5767 one call a function was used. (As it happened, calling the function twice
5768 did not affect the overall behaviour.) The macro has been replaced by a
5769 function that can join an arbitrary number of extra strings onto a growing
5772 (c) The code for expansion conditions now uses a table and a binary chop
5773 instead of a serial search (which was left over from when there were very
5774 few conditions). Also, it now recognizes conditions like "pam" even when
5775 the relevant support is not compiled in: a suitably worded error message is
5776 given if an attempt is made to use such a condition.
5778 11. Added ${time_interval:xxxxx}.
5780 12. A bug was causing one of the ddress fields not to be passed back correctly
5781 from remote delivery subprocesses. The field in question was not being
5782 subsequently used, so this caused to problems in practice.
5784 13. Added new log selectors queue_time and deliver_time.
5786 14. Might have fixed a bug in maildirsizefile handling that threw up
5787 "unexpected character" debug warnings, and recalculated the data
5788 unnecessarily. In any case, I expanded the warning message to give more
5791 15. Added the message "Restricted characters in address" to the statements in
5792 the default ACL that block characters like @ and % in local parts.
5794 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5795 Three changes have been made:
5797 (a) There was a serious bug; a negative response to MAIL caused the whole
5798 recipient domain to be cached as invalid, thereby blocking all messages
5799 to all local parts at the same domain, from all senders. This bug has
5800 been fixed. The domain is no longer cached after a negative response to
5801 MAIL if the sender used is not empty.
5803 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5806 (c) A new callout option, "use_sender" has been added for people who want
5807 the modified behaviour.
5813 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5816 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5817 indeed breaks things for older releases.
5819 3. Added additional logging to the case where there is a problem reading data
5820 from a filter that is running in a subprocess using a pipe, in order to
5821 try to track down a specific problem.
5823 4. Testing facility fudge: when running in the test harness and attempting
5824 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5825 getting "No route to host". Convert this to a timeout.
5827 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5830 6. Some OS don't have socklen_t but use size_t instead. This affects the
5831 fifth argument of getsockopt() amongst other things. This is now
5832 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5833 can be set for individual OS. I have set it for SunOS5, OSF1, and
5834 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5835 some earlier ones do not.
5837 7. Change 4.30/15 was not doing the test caselessly.
5839 8. The standard form for an IPv6 address literal was being rejected by address
5840 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5841 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5842 this, as well as the form without the "IPv6" on the front (but only when
5843 address literals are enabled, of course).
5845 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5847 10. Exim crashed if a message with an empty sender address specified by -f
5848 encountered a router with an errors_to setting. This could be provoked only
5849 by a command such as
5853 where an empty string was supplied; "<>" did not hit this bug.
5855 11. Installed PCRE release 4.5.
5857 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5858 remained set. It is now erased.
5860 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5861 times from message ids (which are base 36 rather than the normal 62).
5863 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5864 were being counted as actual protocol errors, and logged if the log
5865 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5866 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5867 and DATA following a set of rejected RCPTs do not count as protocol errors.
5868 In other words, Exim assumes they were pipelined, though this may not
5869 actually be the case. Of course, in all cases the client gets an
5870 appropriate error code.
5872 15. If a lookup fails in an ACL condition, a message about the failure may
5873 be available; it is used if testing the ACL cannot continue, because most
5874 such messages specify what the cause of the deferral is. However, some
5875 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5876 that caused an old message to be retained and used if a later statement
5877 caused a defer, replacing the real cause of the deferral.
5879 16. If an IP address had so many PTR records that the DNS lookup buffer
5880 was not large enough to hold them, Exim could crash while trying to process
5881 the truncated data. It now detects and logs this case.
5883 17. Further to 4.21/58, another change has been made: if (and only if) the
5884 first line of a message (the first header line) ends with CRLF, a bare LF
5885 in a subsequent header line has a space inserted after it, so as not to
5886 terminate the header.
5888 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5889 unnecessarily, used atoi() instead of strtol(), and didn't check the
5890 termination when getting file sizes from file names by regex.
5892 19. Completely re-implemented the support for maildirsize files, in the light
5893 of a number of problems with the previous contributed implementation
5894 (4.30/29). In particular:
5896 . If the quota is zero, the maildirsize file is maintained, but no quota is
5899 . If the maildir directory does not exist, it is created before any attempt
5900 to write a maildirsize file.
5902 . The quota value in the file is just a cache; if the quota is changed in
5903 the transport, the new value overrides.
5905 . A regular expression is available for excluding directories from the
5908 20. The autoreply transport checks the characters in options that define the
5909 message's headers; it allows continued headers, but it was checking with
5910 isspace() after an embedded newline instead of explicitly looking for a
5913 21. If all the "regular" hosts to which an address was routed had passed their
5914 expiry times, and had not reached their retry times, the address was
5915 bounced, even if fallback hosts were defined. Now Exim should go on to try
5918 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5919 equivalent code in the SMTP transport. Some hosts send humungous responses
5920 to HELO/EHLO, more than 1024 it seems.
5922 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5923 gives compiler warnings in some environments. I've now done it "properly",
5926 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5927 (because of problems with the built-in one) was declared to return uschar *
5928 instead of char *, causing compiler failure.
5930 25. Fixed a file descriptor leak when processing alias/forward files.
5932 26. Fixed a minor format string issue in dbfn.c.
5934 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5936 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5937 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5940 29. When a sender address is verified, it is cached, to save repeating the test
5941 when there is more than one recipient in a message. However, when the
5942 verification involves a callout, it is possible for different callout
5943 options to be set for different recipients. It is too complicated to keep
5944 track of this in the cache, so now Exim always runs a verification when a
5945 callout is required, relying on the callout cache for the optimization.
5946 The overhead is duplication of the address routing, but this should not be
5949 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5950 to be verified with callout=postmaster, and the main callout worked but the
5951 postmaster check failed, the verification correctly failed. However, if a
5952 subsequent RCPT command asked for sender verification *without* the
5953 postmaster check, incorrect caching caused this verification also to fail,
5956 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5957 it was not caching the DNS options (qualify_single, search_parents) that
5958 were used when the lookup failed. A subsequent lookup with different
5959 options therefore always gave the same answer, though there were cases
5960 where it should not have. (Example: a "domains = !$mx_any" option on a
5961 dnslookup router: the "domains" option is always processed without any
5962 widening, but the router might have qualify_single set.) Now Exim uses the
5963 cached value only when the same options are set.
5965 32. Added John Jetmore's "exipick" utility to the distribution.
5967 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5968 than a timeout (e.g. a certificate is required, and is not provided), an
5969 Exim server now closes the connection immediately. Previously it waited for
5970 the client to close - but if the client is SSL, it seems that they each
5971 wait for each other, leading to a delay before one of them times out.
5973 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5974 maintained 0.8.x compatibility because I don't think many are using it, and
5975 it is clearly obsolete.
5977 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5980 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5981 client certificate was expired. A simple patch fixes this, though I don't
5982 understand the full logic of why the verify callback is called multiple
5985 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5986 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5987 which causes problems with some clients (such as the Certicom SSL Plus
5988 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5989 disables the coutermeasure allowing Eudora to connect."
5991 38. Exim was not checking that a write() to a log file succeeded. This could
5992 lead to Bad Things if a log got too big, in particular if it hit a file
5993 size limit. Exim now panics and dies if it cannot write to a log file, just
5994 as it does if it cannot open a log file.
5996 39. Modified OS/Makefile-Linux so that it now contains
5998 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
6000 The two -D definitions ensure that Exim is compiled with large file
6001 support, which makes it possible to handle log files that are bigger than
6004 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
6005 instance) a domain was checked against a named list that involved a lookup,
6006 causing $domain_data to be set, then another domain was checked against the
6007 same list, then the first domain was re-checked, the value of $domain_data
6008 after the final check could be wrong. In particular, if the second check
6009 failed, it could be set empty. This bug probably also applied to
6012 41. The strip_trailing_dot option was not being applied to the address given
6013 with the -f command-line option.
6015 42. The code for reading a message's header from the spool was incrementing
6016 $received_count, but never initializing it. This meant that the value was
6017 incorrect (doubled) while delivering a message in the same process in which
6018 it was received. In the most common configuration of Exim, this never
6019 happens - a fresh exec is done - but it can happen when
6020 deliver_drop_privilege is set.
6022 43. When Exim logs an SMTP synchronization error - client data sent too soon -
6023 it now includes up to 150 characters of the unexpected data in the log
6026 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6027 and building data strings. The size of both of these buffers was 10 000
6028 bytes - far larger than anybody would *ever* want, thought I. Needless to
6029 say, somebody hit the limit. I have increased the maximum line length to
6030 20 000 and the maximum data length of concatenated lines to 100 000. I have
6031 also fixed two bugs, because there was no checking on these buffers. Tsk,
6032 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6033 buffer is too small.
6035 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6036 lsearch lookups. Now it does.
6038 46. When parsing a route_list item in a manualroute router, a fixed-length
6039 buffer was used for the list of hosts. I made this 1024 bytes long,
6040 thinking that nobody would ever have a list of hosts that long. Wrong.
6041 Somebody had a whole pile of complicated expansion conditions, and the
6042 string was silently truncated, leading to an expansion error. It turns out
6043 that it is easier to change to an unlimited length (owing to other changes
6044 that have happened since this code was originally written) than to build
6045 structure for giving a limitation error. The length of the item that
6046 expands into the list of hosts is now unlimited.
6048 47. The lsearch lookup could not handle data where the length of text line was
6049 more than 4095 characters. Such lines were truncated, leading to shortened
6050 data being returned. It should now handle lines of any length.
6052 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6053 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6056 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6057 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6058 no longer happens in comment lines. A list of replacements is now placed
6059 at the head of all of the source files, except those whose only change is
6060 to replace PERL_COMMAND in the very first #! line.
6062 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6063 messages on the queue, with a bottom-up merge sort, using code contributed
6064 by Michael Haardt. This should make operations like -bp somewhat faster on
6065 large queues. It won't affect queue runners, except when queue_run_in_order
6068 51. Installed eximstats 1.31 in the distribution.
6070 52. Added support for SRV lookups to the dnslookup router.
6072 53. If an ACL referred to $message_body or $message_body_end, the value was not
6073 reset for any messages that followed in the same SMTP session.
6075 54. The store-handling optimization for building very long strings was not
6076 differentiating between the different store pools. I don't think this
6077 actually made any difference in practice, but I've tidied it.
6079 55. While running the routers to verify a sender address, $sender_address
6080 was still set to the sender address. This is wrong, because when routing to
6081 send a bounce to the sender, it would be empty. Therefore, I have changed
6082 it so that, while verifying a sender address, $sender_address is set to <>.
6083 (There is no change to what happens when verifying a recipient address.)
6085 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6086 target A or AAAA records (if not already returned) without resetting the
6087 qualify_single or search_parents options of the DNS resolver. These are
6088 inappropriate in this case because the targets of MX and SRV records must
6089 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6090 target that, when qualified, matched something in the local domain. These
6091 two options are now turned off when doing these lookups.
6093 57. It seems that at least some releases of Reiserfs (which does not have the
6094 concept of a fixed number of inodes) returns zero and not -1 for the
6095 number of available inodes. This interacted badly with check_spool_inodes,
6096 which assumed that -1 was the "no such thing" setting. What I have done is
6097 to check that the total number of inodes is greater than zero before doing
6098 the test of how many are available.
6100 58. When a "warn" ACL statement has a log_message modifier, the message is
6101 remembered, and not repeated. This is to avoid a lot of repetition when a
6102 message has many recipients that cause the same warning to be written.
6103 However, Exim was preserving the list of already written lines for an
6104 entire SMTP session, which doesn't seem right. The memory is now reset if a
6105 new message is started.
6107 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6108 debugging output unless log_rewrite was also set.
6110 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6111 of (int)(handle) when we know that handle contains (void *)(-1).
6113 61. The Exim daemon panic-logs an error return when it closes the incoming
6114 connection. However "connection reset by peer" seems to be common, and
6115 isn't really an error worthy of noting specially, so that particular error
6118 62. When Exim is trying to find all the local interfaces, it used to panic and
6119 die if the ioctl to get the interface flags failed. However, it seems that
6120 on at least one OS (Solaris 9) it is possible to have an interface that is
6121 included in the list of interfaces, but for which you get a failure error
6122 for this call. This happens when the interface is not "plumbed" into a
6123 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6124 failure of the "get flags" call assumes that the interface is down.
6126 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6127 makes life easier for people who are doing arithmetic on fields extracted
6128 from dates, where you often get leading zeros that should not be
6129 interpreted as octal.
6131 64. Added qualify_domain to the redirect router, to override the global
6134 65. If a pathologically long header line contained very many addresses (the
6135 report of this problem mentioned 10 000) and each of them was rewritten,
6136 Exim could use up a very large amount of memory. (It kept on making new
6137 copies of the header line as it rewrote, and never released the old ones.)
6138 At the expense of a bit more processing, the header rewriting function has
6139 been changed so that it no longer eats memory in this way.
6141 66. The generation of the Received: header has been moved from the time that a
6142 message starts to be received, to the time that it finishes. The timestamp
6143 in the Received: header should now be very close to that of the <= log
6144 line. There are two side-effects of this change:
6146 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6147 logged header lines no longer include the local Received: line, because
6148 it has not yet been created. The same applies to a copy of the message
6149 that is returned to a non-SMTP sender when a message is rejected.
6151 (b) When a filter file is tested using -bf, no additional Received: header
6152 is added to the test message. After some thought, I decided that this
6155 This change does not affect the value of $received_for. It is still set
6156 after address rewriting, but before local_scan() is called.
6158 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6160 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6161 gave an unhelpful panic error message, and a defer error. I have managed to
6162 change this behaviour so that it now rejects any supplied certificate,
6163 which seems right, as the list of acceptable certificates is empty.
6165 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6166 gave an unhelpful defer error. I have not managed to make this reject any
6167 supplied certificates, but the error message it gives is "no certificate
6168 supplied", which is not helpful.
6170 70. exigrep's output now also includes lines that are not associated with any
6171 message, but which match the given pattern. Implemented by a patch from
6172 Martin Sluka, which also tidied up the Perl a bit.
6174 71. Recipient callout verification, like sender verification, was using <> in
6175 the MAIL FROM command. This isn't really the right thing, since the actual
6176 sender may affect whether the remote host accepts the recipient or not. I
6177 have changed it to use the actual sender in the callout; this means that
6178 the cache record is now keyed on a recipient/sender pair, not just the
6179 recipient address. There doesn't seem to be a real danger of callout loops,
6180 since a callout by the remote host to check the sender would use <>.
6181 [SEE ABOVE: changed after hitting problems.]
6183 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6184 temporary errors. However, in the case of such a code being given after
6185 the end of a data transmission (i.e. after ".") Exim was failing to write
6186 a retry record for the message. (Yes, there was some broken host that was
6187 actually sending 8xx at this point.)
6189 73. An unknown lookup type in a host list could cause Exim to panic-die when
6190 the list was checked. (An example that provoked this was putting <; in the
6191 middle of a list instead of at the start.) If this happened during a DATA
6192 ACL check, a -D file could be left lying around. This kind of configuration
6193 error no longer causes Exim to die; instead it causes a defer error. The
6194 incident is still logged to the main and panic logs.
6196 74. Buglet left over from Exim 3 conversion. The message "too many messages
6197 in one connection" was written to the rejectlog but not the mainlog, except
6198 when address rewriting (yes!) was being logged.
6200 75. Added write_rejectlog option.
6202 76. When a system filter was run not as root (that is, when system_filter_user
6203 was set), the values of the $n variables were not being returned to the
6204 main process; thus, they were not subsequently available in the $sn
6207 77. Added +return_path_on_delivery log selector.
6209 78. A connection timeout was being treated differently from recipients deferred
6210 when testing hosts_max_try with a message that was older than the host's
6211 retry timeout. (The host should not be counted, thus allowing all hosts to
6212 be tried at least once before bouncing.) This may have been the cause of an
6213 occasionally reported bug whereby a message would remain on the queue
6214 longer than the retry timeout, but would be bounced if a delivery was
6215 forced. I say "may" because I never totally pinned down the problem;
6216 setting up timeout/retry tests is difficult. See also the next item.
6218 79. The ultimate address timeout was not being applied to errors that involved
6219 a combination of host plus message (for example, a timeout on a MAIL
6220 command). When an address resolved to a number of possible hosts, and they
6221 were not all tried for each delivery (e.g. because of hosts_max_try), a
6222 message could remain on the queue longer than the retry timeout.
6224 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6227 81. Fixed an obscure SMTP outgoing bug which required at least the following
6228 conditions: (a) there was another message waiting for the same server;
6229 (b) the server returned 5xx to all RCPT commands in the first message so
6230 that the message was not completed; (c) the server dropped the connection
6231 or gave a negative response to the RSET that Exim sends to abort the
6232 transaction. The observed case was a dropped connection after DATA that had
6233 been sent in pipelining mode. That is, the server had advertised PIPELINING
6234 but was not implementing it correctly. The effect of the bug was incorrect
6235 behaviour, such as trying another host, and this could lead to a crash.
6241 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6242 and daemon.c were passed as pointers to ints; they should have been
6243 pointers to socklen_t variables (which are typically unsigned ints).
6245 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6248 3. Fixed a really odd bug that affected only the testing scheme; patching a
6249 certain fixed string in the binary changed the value of another string that
6250 happened to be identical to the end of the original first string.
6252 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6253 name", it returns that address as the IP address. On some operating
6254 systems (e.g. Solaris), it also passes back the IP address string as the
6255 "host name". However, on others (e.g. Linux), it passes back an empty
6256 string. Exim wasn't checking for this, and was changing the host name to an
6257 empty string, assuming it had been canonicalized.
6259 5. Although rare, it is permitted to have more than one PTR record for a given
6260 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6261 all the names associated with an address, because they do in Solaris.
6262 However, it seems that they do not in Linux for data that comes from the
6263 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6264 I found this out when I moved to a new Linux workstation and tried to run
6265 the Exim test suite.
6267 To get round this problem I have changed the code so that it now does its
6268 own call to the DNS to look up PTR records when searching for a host name.
6269 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6270 addresses that are only in /etc/hosts are still found.
6272 This behaviour is, however, controlled by an option called host_lookup_
6273 order, which defaults to "bydns:byaddr". If people want to use the other
6274 order, or indeed, just use one or the other means of lookup, they can
6275 specify it in this variable.
6277 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6278 some operating systems, this comes back from gethostbyaddr() as an empty
6279 string, and this is what Exim used to test for. However, it seems that in
6280 other systems, "." is yielded. Exim now tests for this case too.
6282 7. The values of check_spool_space and check_log_space are now held internally
6283 as a number of kilobytes instead of an absolute number of bytes. If a
6284 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6285 kilobyte. This means that much larger values can be stored.
6287 8. Exim monitor: an attempt to get the action menu when not actually pointing
6288 at a message produces an empty menu entitled "No message selected". This
6289 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6290 no entries in it ("Shell widget menu has zero width and/or height"). So I
6291 have added a single, blank menu entry in this case.
6293 9. Added ${quote_local_part.
6295 10. MIME decoding is now applied to the contents of Subject: header lines when
6298 11. Now that a reference to $sender_host_address automatically causes a reverse
6299 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6300 host lookup before query-style lookups in lists that might use this
6301 variable. This has therefore been abolished, and the "net-" prefix is no
6302 longer necessary for query-style lookups.
6304 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6305 have been a typo for LFLAGS, so it has been changed.
6307 13. The install script calls Exim with "-C /dev/null" in order to find the
6308 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6309 to be output. However, since Exim outputs its version number before the
6310 error, it didn't break the script. It just looked ugly. I fixed this by
6311 always allowing "-C /dev/null" if the caller is root.
6313 14. Ignore overlarge ACL variable number when reading spool file - insurance
6314 against a later release with more variables having written the file.
6316 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6317 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6318 this, as well as the form without the "IPv6" on the front.
6320 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6321 OS/Makefile-Darwin file.
6323 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6324 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6326 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6327 variable. However, some OS use ints for inodes. I've added cast to long int
6328 to get rid of the compiler warning.
6330 19. I had forgotten to lock out "/../" in configuration file names when
6331 ALT_CONFIG_PREFIX was set.
6333 20. Routers used for verification do not need to specify transports. However,
6334 if such a router generated a host list, and callout was configured, Exim
6335 crashed, because it could not find a port number from the (non-existent)
6336 transport. It now assumes port 25 in this circumstance.
6338 21. Added the -t option to exigrep.
6340 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6341 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6342 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6343 them is set, LOOKUP_LSEARCH is forced.
6345 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6347 24. Added sender and host information to the "rejected by local_scan()" log
6348 line; previously there was no indication of these.
6350 25. Added .include_if_exists.
6352 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6353 Linux. It turns out that not all file systems support this. Apparently some
6354 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6355 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6356 not supported on the file descriptor, is now ignored when Exim is trying to
6357 sync a directory. This applies only to Linux.
6359 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6361 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6362 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6365 29. Added support for maildirsize files from supplied patch (modified a bit).
6367 30. The use of :fail: followed by an empty string could lead Exim to respond to
6368 sender verification failures with (e.g.):
6370 550 Verification failed for <xxx>
6371 550 Sender verify failed
6373 where the first response line was missing the '-' that indicates it is not
6374 the final line of the response.
6376 31. The loop for finding the name of the user that called Exim had a hardwired
6377 limit of 10; it now uses the value of finduser_retries, which is used for
6378 all other user lookups.
6380 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6383 33. Exim was neglecting to zero errno before one call of strtol() when
6384 expanding a string and expecting an integer value. On some systems this
6385 resulted in spurious "integer overflow" errors. Also, it was casting the
6386 result into an int without checking.
6388 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6389 did not work. The code looks as if it has *never* worked, though it appears
6390 to have been documented since at least release 1.62. I have made it work.
6392 35. The "timeout_DNS" error in retry rules, also documented since at least
6393 1.62, also never worked. As it isn't clear exactly what this means, and
6394 clearly it isn't a major issue, I have abolished the feature by treating it
6395 as "timeout", and writing a warning to the main and panic logs.
6397 36. The display of retry rules for -brt wasn't always showing the error code
6400 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6401 timeout_connect_A, timeout_connect_MX.
6403 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6404 to the empty sender.
6406 39. The daemon was not analysing the content of -oX till after it had closed
6407 stderr and disconnected from the controlling terminal. This meant that any
6408 syntax errors were only noted on the panic log, and the return code from
6409 the command was 0. By re-arranging the code a little, I've made the
6410 decoding happen first, so such errors now appear on stderr, and the return
6411 code is 1. However, the actual setting up of the sockets still happens in
6412 the disconnected process, so errors there are still only recorded on the
6415 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6416 connections (as happens on some IP stacks) was logged at start up time as
6417 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6418 it from "IPv6 and IPv4", which means that two separate sockets are being
6421 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6422 says whether AF_INET or AF_INET6 was passed as an argument.
6424 42. Exiwhat output was messed up when time zones were included in log
6427 43. Exiwhat now gives more information about the daemon's listening ports,
6428 and whether -tls-on-connect was used.
6430 44. The "port" option of the smtp transport is now expanded.
6432 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6433 silently ignored. Now an error message is written to the main and panic
6436 46. There's a new ACL modifier called "logwrite" which writes to a log file
6437 as soon as it is encountered.
6439 47. Added $local_user_uid and $local_user_gid at routing time.
6441 48. Exim crashed when trying to verify a sender address that was being
6444 49. Exim was recognizing only a space character after ".include". It now also
6445 recognizes a tab character.
6447 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6448 extracting the relevant information from the specification. The man page no
6449 longer contains scrambled data for the -d option, and I've added a section
6450 at the front about calling Exim under different names.
6452 51. Added "extra_headers" argument to the "mail" command in filter files.
6454 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6457 53. Installed eximstats 1.29.
6459 54. Added transport_filter_timeout as a generic transport option.
6461 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6462 Cc: header lines. This was required by RFC 822, but it not required by RFC
6465 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6466 incoming messages that did not have them. Now it does so only if the
6467 message originates locally, that is, if there is no associated remote host
6468 address. When Resent- header lines are present, this applies to the Resent-
6469 lines rather than the non-Resent- lines.
6471 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6472 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6474 58. Messages for configuration errors now include the name of the main
6475 configuration file - useful now that there may be more than one file in a
6476 list (.included file names were always shown).
6478 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6479 for those rare installations that do not start the daemon as root or run it
6480 setuid root. I've cut out the call to initgroups() if the daemon is not
6483 60. The Exim user and group can now be bound into the binary as text strings
6484 that are looked up at the start of Exim's processing.
6486 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6488 62. Added $mailstore_basename variable.
6490 63. Installed patch to sieve.c from Michael Haardt.
6492 64. When Exim failed to open the panic log after failing to open the main log,
6493 the original message it was trying to log was written to stderr and debug
6494 output, but if they were not available (the usual case in production), it
6495 was lost. Now it is written to syslog before the two lines that record the
6496 failures to open the logs.
6498 65. Users' Exim filters run in subprocesses under the user's uid. It is
6499 possible for a "deliver" command or an alias in a "personal" command to
6500 provoke an address rewrite. If logging of address rewriting is configured,
6501 this fails because the process is not running as root or exim. There may be
6502 a better way of dealing with this, but for the moment (because 4.30 needs
6503 to be released), I have disabled address rewrite logging when running a
6504 filter in a non-root, non-exim process.
6510 1. The buildconfig auxiliary program wasn't quoting the value set for
6511 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6512 not defined. This bug was masked in 4.22 by the effect that was fixed in
6515 2. Some messages that were rejected after a message id was allocated were
6516 shown as "incomplete" by exigrep. It no longer does this for messages that
6517 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6519 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6520 have allow_domain_literals set, the ID did not get logged in the <= line.
6521 Domain literals are now always recognized in Message-ID: header lines.
6523 4. The first argument for a ${extract expansion item is the key name or field
6524 number. Leading and trailing spaces in this item were not being ignored,
6525 causing some misleading effects.
6527 5. When deliver_drop_privilege was set, single queue runner processes started
6528 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6529 same command in the process it spins off) were not dropping privilege.
6531 6. When the daemon running as "exim" started a queue runner, it always
6532 re-executed Exim in the spun-off process. This is a waste of effort when
6533 deliver_drop_privilege is set. The new process now just calls the
6534 queue-runner function directly.
6540 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6543 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6544 ignored. Though the use of -oP was forcing the writing of a pid file, it
6545 was always written to the default place.
6547 3. If the message "no IP address found for host xxxx" is generated during
6548 incoming verification, it is now followed by identification of the incoming
6549 connection (so you can more easily find what provoked it).
6551 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6553 5. Added some features to "harden" Exim a bit more against certain attacks:
6555 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6556 be put in Local/Makefile. This is like the never_users runtime option,
6557 but it cannot be overridden. The default setting is "root".
6559 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6560 prefix string with which any file named in a -C command line option
6563 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6564 is retained for -C and -D only if the caller of Exim is root. Without
6565 it, the exim user may also use -C and -D and retain privilege.
6567 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6568 command line option is disabled.
6570 6. Macro names set by the -D option must start with an upper case letter, just
6571 like macro names defined in the configuration file.
6573 7. Added "dereference=" facility to LDAP.
6575 8. Two instances of the typo "uknown" in the source files are fixed.
6577 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6578 the Configure-Makefile script screwed up while processing it.
6580 10. Incorporated PCRE 4.4.
6582 11. The SMTP synchronization check was not operating right at the start of an
6583 SMTP session. For example, it could not catch a HELO sent before the client
6584 waited for the greeting. There is now a check for outstanding input at the
6585 point when the greeting is written. Because of the duplex, asynchronous
6586 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6587 way, but not yet received, when the check is performed.
6589 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6590 on TCP/IP sockets, because this apparently causes some broken clients to
6593 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6594 unchanged) from the Cygwin maintainer.
6596 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6597 was supported instead of testing for mbx. Effectively a typo.
6599 15. The spa authenticator server code was not checking that the input it
6600 received was valid base64.
6602 16. The debug output line for the "set" modifier in ACLs was not showing the
6603 name of the variable that was being set.
6605 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6607 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6608 lookup on its own. Something else was needed to trigger the lookup. For
6609 example, a match in host_lookup or the need for a host name in a host list.
6610 Now, if $sender_host_name is referenced and the host name has not yet been
6611 looked up, a lookup is performed. If the lookup fails, the variable remains
6612 empty, and $host_lookup_failed is set to "1".
6614 19. Added "eqi" as a case-independent comparison operator.
6616 20. The saslauthd authentication condition could segfault if neither service
6617 nor realm was specified.
6619 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6620 error message that was logged was misleading, and incoming SMTP
6621 connections were dropped. The message is now more accurate, and temporary
6622 errors are given to SMTP connections.
6624 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6625 (except RSET) with a 421 error, until QUIT is received. However, it was
6626 failing to send a response to QUIT.
6628 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6629 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6630 ACL is now run after the helo_try_verify_hosts code.
6632 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6633 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6634 case-independent, so other case variants are also recognized). Apparently
6635 some systems use these upper case variants.
6637 25. If more than two messages were waiting for the same host, and a transport
6638 filter was specified for the transport, Exim sent two messages over the
6639 same TCP/IP connection, and then failed with "socket operation on non-
6640 socket" when it tried to send the third.
6642 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6644 27. The extern definition of crypt16() in expand.c was not being excluded when
6645 the OS had its own crypt16() function.
6647 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6648 as a preferred synonym for return_size_limit, both as an option and as an
6651 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6653 30. Changed the default configuration ACL to relax the local part checking rule
6654 for addresses that are not in any local domains. For these addresses,
6655 slashes and pipe symbols are allowed within local parts, but the sequence
6656 /../ is explicitly forbidden.
6658 31. SPA server authentication was not clearing the challenge buffer before
6661 32. log_message in a "warn" ACL statement was writing to the reject log as
6662 well as to the main log, which contradicts the documentation and doesn't
6663 seem right (because no rejection is happening). So I have stopped it.
6665 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6666 However, I am unable to do any testing of this.
6668 34. Fixed an infelicity in the appendfile transport. When checking directories
6669 for a mailbox, to see if any needed to be created, it was accidentally
6670 using path names with one or more superfluous leading slashes; tracing
6671 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6673 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6674 contents are added to the log line that is written for every discarded
6675 recipient. (Previously a log_message setting was ignored.)
6677 36. The ${quote: operator now quotes the string if it is empty.
6679 37. The install script runs exim in order to find its version number. If for
6680 some reason other than non-existence or emptiness, which it checks, it
6681 could not run './exim', it was installing it with an empty version number,
6682 i.e. as "exim-". This error state is now caught, and the installation is
6685 38. An argument was missing from the function that creates an error message
6686 when Exim fails to connect to the socket for saslauthd authentication.
6687 This could cause Exim to crash, or give a corrupted message.
6689 39. Added isip, isip4, and isip6 to ${if conditions.
6691 40. The ACL variables $acl_xx are now saved with the message, and can be
6692 accessed later in routers, transports, and filters.
6694 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6695 strings in the file are not string-expanded.
6697 42. If a MAIL command specified a SIZE value that was too large to fit into an
6698 int variable, the check against message_size_limit failed. Such values are
6699 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6700 day this will have to be increased, but I don't think I want to be around
6701 when emails are that large.
6708 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6709 iconv() is not standard in FreeBSD.
6711 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6712 IPv6 enabled. The observed symptom was a segmentation fault on return from
6713 the function os_common_find_running_interfaces() in src/os.c.
6715 3. In the check_special_case() function in daemon.c I had used "errno" as an
6716 argument name, which causes warnings on some systems. This was basically a
6717 typo, since it was named "eno" in the comments!
6719 4. The code that waits for the clock to tick (at a resolution of some fraction
6720 of a second) so as to ensure message-id uniqueness was always waiting for
6721 at least one whole tick, when it could have waited for less. [This is
6722 almost certainly not relevant at current processor speeds, where it is
6723 unlikely to ever wait at all. But we try to future-proof.]
6725 5. The function that sleeps for a time interval that includes fractions of a
6726 second contained a race. It did not block SIGALRM between setting the
6727 timer, and suspending (a couple of lines later). If the interval was short
6728 and the sigsuspend() was delayed until after it had expired, the suspension
6729 never ended. On busy systems this could lead to processes getting stuck for
6732 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6733 process, before it forks any delivery processes. The open lookup caching
6734 mechanism meant that the open file or database connection was passed into
6735 the delivery process. The problem was that delivery processes always tidy
6736 up cached lookup data. This could cause a problem for the next delivery
6737 process started by the queue runner, because the external queue runner
6738 process does not know about the closure. So the next delivery process
6739 still has data in the lookup cache. In the case of a file lookup, there was
6740 no problem because closing a file descriptor in a subprocess doesn't affect
6741 the parent. However, if the lookup was caching a connection to a database,
6742 the connection was closed, and the second delivery process was likely to
6743 see errors such as "PGSQL: query failed: server closed the connection
6744 unexpectedly". The problem has been fixed by closing all cached lookups
6745 in a queue runner before running a delivery process.
6747 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6748 seem to have the "const" qualifier which it has on other OS. I've
6751 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6752 *of the same type* (client or server) with the same public name that an
6753 error should be diagnosed.
6755 9. When Exim looked up a host name for an IP address, but failed to find the
6756 original IP address when looking up the host name (a safety check), it
6757 output the message "<ip address> does not match any IP for NULL", which was
6758 confusing, to say the least. The bug was that the host name should have
6759 appeared instead of "NULL".
6761 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6762 user that is built into the binary, and the -C or -D options is used, root
6763 privilege is dropped before the configuration file is read. In addition,
6764 logging is switched to stderr instead of the normal log files. If the
6765 configuration then re-defines the Exim user, the unprivileged environment
6766 is probably not what is expected, so Exim logs a panic warning message (but
6769 However, if deliver_drop_privilege is set, the unprivileged state may well
6770 be exactly what is intended, so the warning has been cut out in that case,
6771 and Exim is allowed to try to write to its normal log files.
6777 1. smtp_return_error_details was not giving details for temporary sender
6778 or receiver verification errors.
6780 2. Diagnose a configuration error if two authenticators have the same public
6783 3. Exim used not to create the message log file for a message until the first
6784 delivery attempt. This could be confusing when incoming messages were held
6785 for policy or load reasons. The message log file is now created at the time
6786 the message is received, and an initial "Received" line is written to it.
6788 4. The automatically generated man page for command line options had a minor
6789 bug that caused no ill effects; however, a more serious problem was that
6790 the procedure for building the man page automatically didn't always
6791 operate. Consequently, release 4.20 contains an out-of-date version. This
6792 shouldn't happen again.
6794 5. When building Exim with embedded Perl support, the script that builds the
6795 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6796 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6798 6. The freeze_tell option was not being used for messages that were frozen on
6799 arrival, either by an ACL or by local_scan().
6801 7. Added the smtp_incomplete_transaction log selector.
6803 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6804 was accepting AUTH without a new EHLO.
6806 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6807 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6810 10. Logging of TCP/IP connections (when configured) now happens in the main
6811 daemon process instead of the child process, so that the TCP/IP connection
6812 count is more accurate (but it can never be perfect).
6814 11. The use of "drop" in a nested ACL was not being handled correctly in the
6815 outer ACL. Now, if condition failure induced by the nested "drop" causes
6816 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6817 or "require"), the connection is dropped.
6819 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6820 that yield "discard" can now be used with an "accept" or a "discard" verb,
6821 but an error is generated for any others (because I can't see a useful way
6822 to define what should happen).
6824 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6825 are now processed in the same way as lines in the Exim configuration file.
6826 In particular, continuation lines are supported.
6828 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6830 15. Added -ti meaning -t -i.
6832 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6833 domains, and warn by logging if others are found.
6835 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6836 returned from GIFCONF when Exim is trying to find the list of interfaces on
6837 a host. The code in os.c has been modified to copy each ifreq to an aligned
6838 structure in all cases.
6840 Also, in some cases, the returned ifreq's were being copied to a 'struct
6841 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6842 means the last couple of bytes of an IPv6 address could be chopped if the
6843 ifreq contained only a normal sockaddr (14 bytes storage).
6845 18. Named domain lists were not supported in the hosts_treat_as_local option.
6846 An entry such as +xxxx was not recognized, and was treated as a literal
6849 19. Ensure that header lines added by a DATA ACL are included in the reject log
6850 if the ACL subsequently rejects the message.
6852 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6853 MD5 (which is deprecated).
6855 21. When testing a filter file using -bf, Exim was writing a message when it
6856 took the sender from a "From " line in the message, but it was not doing so
6857 when it took $return_path from a Return-Path: header line. It now does.
6859 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6860 with a valid header line field name (a series of printing characters
6861 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6863 23. Changed "disc" in the source to "disk" to conform to the documentation and
6864 the book and for uniformity.
6866 24. Ignore Sendmail's -Ooption=value command line item.
6868 25. When execve() failed while trying to run a command in a pipe transport,
6869 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6870 could be confused with a return value of 69 from the command itself. This
6871 has been changed to 127, the value the shell returns if it is asked to run
6872 a non-existent command. The wording for the related log line suggests a
6873 non-existent command as the problem.
6875 26. If received_header_text expands to an empty string, do not add a Received:
6876 header line to the message. (Well, it adds a token one on the spool, but
6877 marks it "old" so that it doesn't get used or transmitted.)
6879 27. Installed eximstats 1.28 (addition of -nt option).
6881 28. There was no check for failure on the call to getsockname() in the daemon
6882 code. This can fail if there is a shortage of resources on the system, with
6883 ENOMEM, for example. A temporary error is now given on failure.
6885 29. Contrary to the C standard, it seems that in some environments, the
6886 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6887 program. Exim now does this explicitly; it affects the formatting of
6888 timestamps using strftime().
6890 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6891 complaints. I've now initialized all the variables, to avoid this.
6893 32. Header lines added by a system filter were not being "seen" during
6894 transport-time rewrites.
6896 33. The info_callback() function passed to OpenSSL is set up with type void
6897 (*)(SSL *, int, int), as described somewhere. However, when calling the
6898 function (actually a macro) that sets it up, the type void(*)() is
6899 expected. I've put in a cast to prevent warnings from picky compilers.
6901 34. If a DNS black list lookup found a CNAME record, but there were no A
6902 records associated with the domain it pointed at, Exim crashed.
6904 35. If a DNS black list lookup returned more than one A record, Exim ignored
6905 all but the first. It now scans all returned addresses if a particular IP
6906 value is being sought. In this situation, the contents of the
6907 $dnslist_value variable are a list of all the addresses, separated by a
6910 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6911 to accept a host name and all its aliases if the forward lookup for any of
6912 them yielded the IP address of the incoming connection. Now it accepts only
6913 those names whose forward lookup yields the correct IP address. Any other
6914 names are discarded. This closes a loophole whereby a rogue DNS
6915 administrator could create reverse DNS records to break through a
6916 wildcarded host restriction in an ACL.
6918 37. If a user filter or a system filter that ran in a subprocess used any of
6919 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6920 the wrong values were passed to the pipe command ($thisaddress had the
6921 value of $0, $0 had the value of $1, etc). This bug was introduced by
6922 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6924 38. Improved the line breaking for long SMTP error messages from ACLs.
6925 Previously, if there was no break point between 40 and 75 characters, Exim
6926 left the rest of the message alone. Two changes have been made: (a) I've
6927 reduced the minimum length to 35 characters; (b) if it can't find a break
6928 point between 35 and 75 characters, it looks ahead and uses the first one
6929 that it finds. This may give the occasional overlong line, but at least the
6930 remaining text gets split now.
6932 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6933 file descriptors might be low, and that setting 1000 would always raise it.
6934 It turns out that in some environments, the limit is already over 1000 and
6935 that lowering it causes trouble. So now Exim takes care not to decrease it.
6937 40. When delivering a message, the value of $return_path is set to $sender_
6938 address at the start of routing (routers may change the value). By an
6939 oversight, this default was not being set up when an address was tested by
6940 -bt or -bv, which affected the outcome if any router or filter referred to
6943 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6944 log only when "message" or "log_message" are set. However, if one of the
6945 conditions was an address verification, or a call to a nested ACL, the
6946 messages generated by the underlying test were being passed through. This
6947 no longer happens. The underlying message is available in $acl_verify_
6948 message for both "message" and "log_message" expansions, so it can be
6949 passed through if needed.
6951 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6952 new expansion $bh_ to give the encoded byte string without charset
6953 translation. Translation happens only if iconv() is available; HAVE_ICONV
6954 indicates this at build time. HEADERS_CHARSET gives the charset to
6955 translate to; headers_charset can change it in the configuration, and
6956 "headers charset" can change it in an individual filter file.
6958 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6959 that creates RFC 2047 encoded "words" labels them as that charset instead
6960 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6961 expansion operator; (ii) when Exim creates a From: line for a local
6962 message; (iii) when a header line is rewritten to include a "phrase" part.
6964 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6965 buggy, causing it to skip the first lines of messages whose message ID
6966 ended in 'D'. This would not have bitten before Exim release 4.14, because
6967 message IDs were unlikely to end in 'D' before then. The effect was to have
6968 incorrect size information for certain domains.
6970 45. #include "config.h" was missing at the start of the crypt16.c module. This
6971 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6974 46. If there was a timeout during a "random" callout check, Exim treated it as
6975 a failure of the random address, and carried on sending RSET and the real
6976 address. If the delay was just some slowness somewhere, the response to the
6977 original RCPT would be taken as a response to RSET and so on, causing
6978 mayhem of various kinds.
6980 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6981 when I implemented it. It didn't allow for the fact that some option values
6982 may legitimately be negative (e.g. size_addition), and it didn't even do
6983 the right test for positive values.
6985 48. Domain names in DNS records are case-independent. Exim always looks them up
6986 in lower case. Some resolvers return domain names in exactly the case they
6987 appear in the zone file, that is, they may contain uppercase letters. Not
6988 all resolvers do this - some return always lower case. Exim was treating a
6989 change of case by a resolver as a change of domain, similar to a widening
6990 of a domain abbreviation. This triggered its re-routing code and so it was
6991 trying to route what was effectively the same domain again. This normally
6992 caused routing to fail (because the router wouldn't handle the domain
6993 twice). Now Exim checks for this case specially, and just changes the
6994 casing of the domain that it ultimately uses when it transmits the message
6997 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
7000 50. If a filter generated a file delivery with a non-absolute name (possible if
7001 no home directory exists for the router), the forbid_file option was not
7004 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
7005 the existing equality matching.
7007 52. Exim was using ints instead of ino_t variables in some places where it was
7008 dealing with inode numbers.
7010 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
7011 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
7012 called TMPDIR, and if it finds it is different, it changes its value.
7014 54. The smtp_printf() function is now made available to local_scan() so
7015 additional output lines can be written before returning. There is also an
7016 smtp_fflush() function to enable the detection of a dropped connection.
7017 The variables smtp_input and smtp_batched_input are exported to
7020 55. Changed the default runtime configuration: the message "Unknown user"
7021 has been removed from the ACL, and instead placed on the localuser router,
7022 using the cannot_route_message feature. This means that any verification
7023 failures that generate their own messages won't get overridden. Similarly,
7024 the "Unrouteable address" message that was in the ACL for unverifiable
7025 relay addresses has also been removed.
7027 56. Added hosts_avoid_esmtp to the smtp transport.
7029 57. The exicyclog script was not checking for the esoteric option
7030 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7031 will work only if exicyclog is run under the appropriate euid.
7033 58. Following a discussion on the list, the rules by which Exim recognises line
7034 endings on incoming messages have been changed. The -dropcr and drop_cr
7035 options are now no-ops, retained only for backwards compatibility. The
7036 following line terminators are recognized: LF CRLF CR. However, special
7037 processing applies to CR:
7039 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7040 nor a local message in the state where . is a terminator.
7042 (ii) If a bare CR is encountered in a header line, an extra space is added
7043 after the line terminator so as not to end the header. The reasoning
7044 behind this is that bare CRs in header lines are most likely either
7045 to be mistakes, or people trying to play silly games.
7047 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7048 was being incorrectly given as 18 bytes larger than it should have been.
7049 This is a VOB (very old bug).
7051 60. This may never have affected anything current, but just in case it has:
7052 When the local host is found other than at the start of a list of hosts,
7053 the local host, those with the same MX, and any that follow, are discarded.
7054 When the list in question was part of a longer list of hosts, the following
7055 hosts (not currently being processed) were also being discarded. This no
7056 longer happens. I'm not sure if this situation could ever has previously
7059 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7062 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7063 header. This is something that is recommended in a new Internet Draft, and
7064 is something that is documented as being done by Sendmail. There are two
7065 possible values. For messages generated by the autoreply transport, Exim
7068 Auto-Submitted: auto-replied
7070 whereas for all other generated messages (e.g. bounces) it adds
7072 Auto-Submitted: auto-generated
7074 63. The "personal" condition in filters now includes a test for the
7075 Auto-Submitted: header. If it contains the string "auto-" the message it
7076 not considered personal.
7078 64. Added rcpt_include_affixes as a generic transport option.
7080 65. Added queue_only_override (default true).
7082 66. Added the syslog_duplication option.
7084 67. If what should have been the first header line of a message consisted of
7085 a space followed by a colon, Exim was mis-interpreting it as a header line.
7086 It isn't of course - it is syntactically invalid and should therefore be
7087 treated as the start of the message body. The misbehaviour could have
7088 caused a number of strange effects, including loss of data in subsequent
7089 header lines, and spool format errors.
7091 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7092 client host had authenticated. This control can now be exercised by an ACL
7093 for more flexibility.
7095 69. By default, callouts do not happen when testing with -bh. There is now a
7096 variant, -bhc, which does actually run the callout code, including
7097 consulting and updating the callout cache.
7099 70. Added support for saslauthd authentication, courtesy of Alexander
7102 71. If statvfs() failed on the spool or log directories while checking their
7103 size for availability, Exim confusingly gave the error "space shortage".
7104 Furthermore, in debugging mode it crashed with a floating point exception.
7105 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7106 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7107 serious problem, Exim now writes to the main and panic logs when this
7108 happens, with details of the failure. It then refuses to accept the
7109 incoming message, giving the message "spool directory problem" or "log
7110 directory problem" with a 421 code for SMTP messages.
7112 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7113 0, 1, and 2 exist, because some OS complain for execs without them (see
7114 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7115 descriptors. However, the code omitted to check that the open succeeded,
7116 causing mysterious errors if for some reason the permissions on /dev/null
7117 got screwed. Now Exim writes a message to the main and panic logs, and
7118 bombs out if it can't open /dev/null.
7120 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7121 interact so that it is all more flexible. It is supposed to remain
7122 backwards compatible. Also added extra_local_interfaces.
7124 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7125 to bomb out with an assertion failure - to the client this appears as a
7126 connection drop. This problem occurs in the part of the code that was taken
7127 from the Samba project. Fortunately, the assertion is in a very simple
7128 function, so I have fixed this by reproducing the function inline in the
7129 one place where it is called, and arranging for authentication to fail
7130 instead of killing the process with assert().
7132 75. The SPA client code was not working when the server requested OEM rather
7133 than Unicode encoding.
7135 76. Added code to make require_files with a specific uid setting more usable in
7136 the case where statting the file as root fails - usually a non-root-mounted
7137 NFS file system. When this happens and the failure is EACCES, Exim now
7138 forks a subprocess and does the per-uid checking as the relevant uid.
7140 77. Added process_log_path.
7142 78. If log_file_path was not explicitly set, a setting of check_log_space or
7143 check_log_inodes was ignored.
7145 79. If a space check for the spool or log partitions fails, the incident is now
7146 logged. Of course, in the latter case the data may get lost...
7148 80. Added the %p formatting code to string_format() so that it can be used to
7149 print addresses in debug_print(). Adjusted all the address printing in the
7150 debugging in store.c to use %p rather than %d.
7152 81. There was a concern that a line of code in smtp_in.c could overflow a
7153 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7154 initially expressed, the concern was not well-founded, because trailing
7155 spaces are removed early. However, if the trailing spaces were followed by
7156 a NULL, they did not get removed, so the overflow was possible. Two fixes
7159 (a) I re-wrote the offending code in a cleaner fashion.
7160 (b) If an incoming SMTP command contains a NULL character, it is rejected
7163 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7164 runs initgroups(), so that if the Exim user is in any additional groups,
7165 they will be used during message reception.
7171 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob