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 JH/01 OpenSSL: With debug enabled output keying information sufficient, server
12 side, to decode a TLS 1.3 packet capture.
14 JH/02 OpenSSL: Suppress the sending of (stateful) TLS1.3 session tickets.
15 Previously the default library behaviour applied, sending two, each in
18 JH/03 Debug output for ACL now gives the config file name and line number for
21 JH/04 The default received_header_text now uses the RFC 8314 tls cipher clause.
23 JH/05 DKIM: ensure that dkim_domain elements are lowercased before use.
25 JH/06 Fix buggy handling of autoreply bounce_return_size_limit, and a possible
26 buffer overrun for (non-chunking) other transports.
28 JH/07 GnuTLS: Our use of late (post-handshake) certificate verification, under
29 TLS1.3, means that a server rejecting a client certificate is not visible
30 to the client until the first read of encrypted data (typically the
31 response to EHLO). Add detection for that case and treat it as a failed
32 TLS connection attempt, so that the normal retry-in-clear can work (if
35 JB/01 Bug 2375: fix expansions of 822 addresses having comments in local-part
36 and/or domain. Found and fixed by Jason Betts.
38 JH/08 Add hardening against SRV & TLSA lookups the hit CNAMEs (a nonvalid
39 configuration). If a CNAME target was not a wellformed name pattern, a
42 JH/09 Logging: Fix initial listening-on line for multiple ports for an IP when
43 the OS reports them interleaved with other addresses.
45 JH/10 OpenSSL: Fix aggregation of messages. Previously, when PIPELINING was
46 used both for input and for a verify callout, both encrypted, SMTP
47 responses being sent by the server could be lost. This resulted in
48 dropped connections and sometimes bounces generated by a peer sending
51 JH/11 Harden plaintext authenticator against a badly misconfigured client-send
52 string. Previously it was possible to cause undefined behaviour in a
53 library routine (usually a crash). Found by "zerons".
55 JH/12 Bug 2384: fix "-bP smtp_receive_timeout". Previously it returned no
58 JH/13 Bug 2386: Fix builds with Dane under LibreSSL 2.9.0 onward. Some old
59 API was removed, so update to use the newer ones.
61 JH/14 Bug 1891: Close the log file if receiving a non-smtp message, without
62 any timeout set, is taking a long time. Previously we would hang on to a
63 rotated logfile "forever" if the input was arriving with long gaps
64 (a previous attempt to fix addressed lack, for a long time, of initial
67 HS/01 Bug 2390: Use message_id for tempfile creation to avoid races in a
68 shared (NFS) environment. The length of the tempfile name is now
69 4 + 16 ("hdr.$message_exim_id") which might break on file
70 systems which restrict the file name length to lower values.
73 HS/01 Bug 2390: Use message_id for tempfile creation to avoid races in a
74 shared (NFS) environment.
76 HS/02 Bug 2392: exigrep does case sensitive *option* processing (as it
77 did for all versions <4.90). Notably -M, -m, --invert, -I may be
80 JH/15 Use unsigned when creating bitmasks in macros, to avoid build errors
81 on some platforms for bit 31.
83 JH/16 GnuTLS: rework ciphersuite strings under recent library versions. Thanks
84 to changes apparently associated with TLS1.3 handling some of the APIs
85 previously used were either nonfunctional or inappropriate. Strings
86 like TLS1.3:ECDHE_SECP256R1__RSA_PSS_RSAE_SHA256__AES_256_GCM__AEAD:256
87 and TLS1.2:ECDHE_SECP256R1__RSA_SHA256__AES_128_CBC__SHA256:128 replace
88 the previous TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256 .
89 This affects log line X= elements, the $tls_{in,out}_cipher variables,
90 and the use of specific cipher names in the encrypted= ACL condition.
92 JH/17 OpenSSL: the default openssl_options now disables ssl_v3.
94 JH/18 GnuTLS: fix $tls_out_ocsp under hosts_request_ocsp. Previously the
95 verification result was not updated unless hosts_require_ocsp applied.
97 JH/19 Bug 2398: fix listing of a named-queue. Previously, even with the option
98 queue_list_requires_admin set to false, non-admin users were denied the
101 JH/20 Bug 2389: fix server advertising of usable certificates, under GnuTLS in
102 directory-of-certs mode. Previously they were advertised despite the
105 JH/21 The smtp transport option "hosts_noproxy_tls" is now unset by default.
106 A single TCP connection by a client will now hold a TLS connection open
107 for multiple message deliveries, by default. Previoud the default was to
110 JH/22 The smtp transport option "hosts_try_dane" now enables all hosts by
111 default. If built with the facility, DANE will be used. The facility
112 SUPPORT_DANE is now enabled in the prototype build Makefile "EDITME".
114 JH/23 The build default is now for TLS to be included; the SUPPORT_TLS define
115 is replaced with DISABLE_TLS. Either USE_GNUTLS or (the new) USE_OPENSSL
116 must be defined and you must still, unless you define DISABLE_TLS, manage
117 the the include-dir and library-file requirements that go with that
118 choice. Non-TLS builds are still supported.
120 JH/24 Fix duplicated logging of peer name/address, on a transport connection-
123 JH/25 The smtp transport option "hosts_try_fastopen" now enables all hosts by
124 default. If the platform supports and has the facility enabled, it will
125 be requested on all coneections.
127 JH/26 The PIPE_CONNECT facility is promoted from experimental status and is now
128 controlled by the build-time option SUPPORT_PIPE_CONNECT.
130 PP/01 Unbreak heimdal_gssapi, broken in 4.92.
132 JH/27 Bug 2404: Use the main-section configuration option "dsn_from" for
133 success-DSN messages. Previously the From: header was always the default
134 one for these; the option was ignored.
136 JH/28 Fix the timeout on smtp response to apply to the whole response.
137 Previously it was reset for every read, so a teergrubing peer sending
138 single bytes within the time limit could extend the connection for a
139 long time. Credit to Qualsys Security Advisory Team for the discovery.
141 JH/29 Fix DSN Final-Recipient: field. Previously it was the post-routing
142 delivery address, which leaked information of the results of local
143 forwarding. Change to the original envelope recipient address, per
146 JH/30 Bug 2411: Fix DSN generation when RFC 3461 failure notification is
147 requested. Previously not bounce was generated and a log entry of
148 error ignored was made.
150 JH/31 Avoid re-expansion in ${sort } expansion. (CVE-2019-13917)
152 JH/32 Introduce a general tainting mechanism for values read from the input
153 channel, and values derived from them. Refuse to expand any tainted
154 values, to catch one form of exploit.
156 JH/33 Bug 2413: Fix dkim_strict option. Previously the expansion result
157 was unused and the unexpanded text used for the test. Found and
158 fixed by Ruben Jenster.
160 JH/34 Fix crash after TLS shutdown. When the TCP/SMTP channel was left open,
161 an attempt to use a TLS library read routine dereffed a nul pointer,
168 JH/01 Remove code calling the customisable local_scan function, unless a new
169 definition "HAVE_LOCAL_SCAN=yes" is present in the Local/Makefile.
171 JH/02 Bug 1007: Avoid doing logging from signal-handlers, as that can result in
172 non-signal-safe functions being used.
174 JH/03 Bug 2269: When presented with a received message having a stupidly large
175 number of DKIM-Signature headers, disable DKIM verification to avoid
176 a resource-consumption attack. The limit is set at twenty.
178 JH/04 Add variables $arc_domains, $arc_oldest_pass for ARC verify. Fix the
179 report of oldest_pass in ${authres } in consequence, and separate out
180 some descriptions of reasons for verification fail.
182 JH/05 Bug 2273: Cutthrough delivery left a window where the received messsage
183 files in the spool were present and unlocked. A queue-runner could spot
184 them, resulting in a duplicate delivery. Fix that by doing the unlock
185 after the unlink. Investigation by Tim Stewart. Take the opportunity to
186 add more error-checking on spoolfile handling while that code is being
189 PP/01 Refuse to open a spool data file (*-D) if it's a symlink.
190 No known attacks, no CVE, this is defensive hardening.
192 JH/06 Bug 2275: The MIME ACL unlocked the received message files early, and
193 a queue-runner could start a delivery while other operations were ongoing.
194 Cutthrough delivery was a common victim, resulting in duplicate delivery.
195 Found and investigated by Tim Stewart. Fix by using the open message data
196 file handle rather than opening another, and not locally closing it (which
197 releases a lock) for that case, while creating the temporary .eml format
198 file for the MIME ACL. Also applies to "regex" and "spam" ACL conditions.
200 JH/07 Bug 177: Make a random-recipient callout success visible in ACL, by setting
201 $sender_verify_failure/$recipient_verify_failure to "random".
203 JH/08 When generating a selfsigned cert, use serial number 1 since zero is not
206 JH/09 Bug 2274: Fix logging of cmdline args when starting in an unlinked cwd.
207 Previously this would segfault.
209 JH/10 Fix ARC signing for case when DKIM signing failed. Previously this would
212 JH/11 Bug 2264: Exim now only follows CNAME chains one step by default. We'd
213 like zero, since the resolver should be doing this for us, But we need one
214 as a CNAME but no MX presence gets the CNAME returned; we need to check
215 that doesn't point to an MX to declare it "no MX returned" rather than
216 "error, loop". A new main option is added so the older capability of
217 following some limited number of chain links is maintained.
219 JH/12 Add client-ip info to non-pass iprev ${authres } lines.
221 JH/13 For receent Openssl versions (1.1 onward) use modern generic protocol
222 methods. These should support TLS 1.3; they arrived with TLS 1.3 and the
223 now-deprecated earlier definitions used only specified the range up to TLS
224 1.2 (in the older-version library docs).
226 JH/14 Bug 2284: Fix DKIM signing for body lines starting with a pair of dots.
228 JH/15 Rework TLS client-side context management. Stop using a global, and
229 explicitly pass a context around. This enables future use of TLS for
230 connections to service-daemons (eg. malware scanning) while a client smtp
231 connection is using TLS; with cutthrough connections this is quite likely.
233 JH/16 Fix ARC verification to do AS checks in reverse order.
235 JH/17 Support a "tls" option on the ${readsocket } expansion item.
237 JH/18 Bug 2287: Fix the protocol name (eg utf8esmtp) for multiple messages
238 using the SMTPUTF8 option on their MAIL FROM commands, in one connection.
239 Previously the "utf8" would be re-prepended for every additional message.
241 JH/19 Reject MAIL FROM commands with SMTPUTF8 when the facility was not advertised.
242 Previously thery were accepted, resulting in issues when attempting to
243 forward messages to a non-supporting MTA.
245 PP/02 Let -n work with printing macros too, not just options.
247 JH/20 Bug 2296: Fix cutthrough for >1 address redirection. Previously only
248 one parent address was copied, and bogus data was used at delivery-logging
249 time. Either a crash (after delivery) or bogus log data could result.
250 Discovery and analysis by Tim Stewart.
252 PP/03 Make ${utf8clean:} expansion operator detect incomplete final character.
253 Previously if the string ended mid-character, we did not insert the
254 promised '?' replacement.
256 PP/04 Documentation: current string operators work on bytes, not codepoints.
258 JH/21 Change as many as possible of the global flags into one-bit bitfields; these
259 should pack well giving a smaller memory footprint so better caching and
260 therefore performance. Group the declarations where this can't be done so
261 that the byte-sized flag variables are not interspersed among pointer
262 variables, giving a better chance of good packing by the compiler.
264 JH/22 Bug 1896: Fix the envelope from for DMARC forensic reports to be possibly
265 non-null, to avoid issues with sites running BATV. Previously reports were
266 sent with an empty envelope sender so looked like bounces.
268 JH/23 Bug 2318: Fix the noerror command within filters. It wasn't working.
269 The ignore_error flag wasn't being returned from the filter subprocess so
270 was not set for later routers. Investigation and fix by Matthias Kurz.
272 JH/24 Bug 2310: Raise a msg:fail:internal event for each undelivered recipient,
273 and a msg:complete for the whole, when a message is manually removed using
274 -Mrm. Developement by Matthias Kurz, hacked on by JH.
276 JH/25 Avoid fixed-size buffers for pathnames in DB access. This required using
277 a "Gnu special" function, asprintf() in the DB utility binary builds; I
278 hope that is portable enough.
280 JH/26 Bug 2311: Fix DANE-TA verification under GnuTLS. Previously it was also
281 requiring a known-CA anchor certificate; make it now rely entirely on the
282 TLSA as an anchor. Checking the name on the leaf cert against the name
283 on the A-record for the host is still done for TA (but not for EE mode).
285 JH/27 Fix logging of proxy address. Previously, a pointless "PRX=[]:0" would be
286 included in delivery lines for non-proxied connections, when compiled with
287 SUPPORT_SOCKS and running with proxy logging enabled.
289 JH/28 Bug 2314: Fire msg:fail:delivery event even when error is being ignored.
290 Developement by Matthias Kurz, tweaked by JH. While in that bit of code,
291 move the existing event to fire before the normal logging of message
292 failure so that custom logging is bracketed by normal logging.
294 JH/29 Bug 2322: A "fail" command in a non-system filter (file) now fires the
295 msg:fail:internal event. Developement by Matthias Kurz.
297 JH/30 Bug 2329: Increase buffer size used for dns lookup from 2k, which was
298 far too small for todays use of crypto signatures stored there. Go all
299 the way to the max DNS message size of 64kB, even though this might be
300 overmuch for IOT constrained device use.
302 JH/31 Fix a bad use of a copy function, which could be used to pointlessly
303 copy a string over itself. The library routine is documented as not
304 supporting overlapping copies, and on MacOS it actually raised a SIGABRT.
306 JH/32 For main options check_spool_space and check_inode_space, where the
307 platform supports 64b integers, support more than the previous 2^31 kB
308 (i.e. more than 2 TB). Accept E, P and T multipliers in addition to
309 the previous G, M, k.
311 JH/33 Bug 2338: Fix the cyrus-sasl authenticator to fill in the
312 $authenticated_fail_id variable on authentication failure. Previously
315 JH/34 Increase RSA keysize of autogen selfsign cert from 1024 to 2048. RHEL 8.0
316 OpenSSL didn't want to use such a weak key. Do for GnuTLS also, and for
317 more-modern GnuTLS move from GNUTLS_SEC_PARAM_LOW to
318 GNUTLS_SEC_PARAM_MEDIUM.
320 JH/35 OpenSSL: fail the handshake when SNI processing hits a problem, server
321 side. Previously we would continue as if no SNI had been received.
323 JH/36 Harden the handling of string-lists. When a list consisted of a sole
324 "<" character, which should be a list-separator specification, we walked
325 off past the nul-terimation.
327 JH/37 Bug 2341: Send "message delayed" warning MDNs (restricted to external
328 causes) even when the retry time is not yet met. Previously they were
329 not, meaning that when (say) an account was over-quota and temp-rejecting,
330 and multiple senders' messages were queued, only one sender would get
331 notified on each configured delay_warning cycle.
333 JH/38 Bug 2351: Log failures to extract envelope addresses from message headers.
335 JH/39 OpenSSL: clear the error stack after an SSL_accept(). With anon-auth
336 cipher-suites, an error can be left on the stack even for a succeeding
337 accept; this results in impossible error messages when a later operation
340 AM/01 Bug 2359: GnuTLS: repeat lowlevel read and write operations while they
341 return error codes indicating retry. Under TLS1.3 this becomes required.
343 JH/40 Fix the feature-cache refresh for EXPERIMENTAL_PIPE_CONNECT. Previously
344 it only wrote the new authenticators, resulting in a lack of tracking of
345 peer changes of ESMTP extensions until the next cache flush.
347 JH/41 Fix the loop reading a message header line to check for integer overflow,
348 and more-often against header_maxsize. Previously a crafted message could
349 induce a crash of the recive process; now the message is cleanly rejected.
351 JH/42 Bug 2366: Fix the behaviour of the dkim_verify_signers option. It had
352 been totally disabled for all of 4.91. Discovery and fix by "Mad Alex".
358 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
359 When redis_servers is set to a list of > 1 element, and the Redis servers
360 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
361 case of MOVED into a DEFER case instead, thus moving the query onto the
362 next server in the list. For a cluster of N elements, all N servers must
363 be defined in redis_servers.
365 GF/02 Catch and remove uninitialized value warning in exiqsumm
366 Check for existence of @ARGV before looking at $ARGV[0]
368 JH/01 Replace the store_release() internal interface with store_newblock(),
369 which internalises the check required to safely use the old one, plus
370 the allocate and data copy operations duplicated in both (!) of the
371 extant use locations.
373 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
374 modifier. This matches the restriction on the commandline.
376 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
377 Previously only the last row was returned.
379 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
380 we assumed that tags in the header were well-formed, and parsed the
381 element content after inspecting only the first char of the tag.
382 Assumptions at that stage could crash the receive process on malformed
385 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
386 While running the DKIM ACL we operate on the Permanent memory pool so that
387 variables created with "set" persist to the DATA ACL. Also (at any time)
388 DNS lookups that fail create cache records using the Permanent pool. But
389 expansions release any allocations made on the current pool - so a dnsdb
390 lookup expansion done in the DKIM ACL releases the memory used for the
391 DNS negative-cache, and bad things result. Solution is to switch to the
392 Main pool for expansions.
393 While we're in that code, add checks on the DNS cache during store_reset,
394 active in the testsuite.
395 Problem spotted, and debugging aided, by Wolfgang Breyha.
397 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
398 When none of the hosts presented to a transport match an already-open
399 connection, close it and proceed with the list. Previously we would
400 queue the message. Spotted by Lena with Yahoo, probably involving
403 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
404 Previously a spurious "250 OK id=" response was appended to the proper
407 JH/08 The "support for" informational output now, which built with Content
408 Scanning support, has a line for the malware scanner interfaces compiled
409 in. Interface can be individually included or not at build time.
411 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
412 by the template makefile "src/EDITME". The "STREAM" support for an older
413 ClamAV interface method is removed.
415 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
416 rows affected is given instead).
418 JH/11 The runtime Berkeley DB library version is now additionally output by
419 "exim -d -bV". Previously only the compile-time version was shown.
421 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
422 SMTP connection. Previously, when one had more receipients than the
423 first, an abortive onward connection was made. Move to full support for
424 multiple onward connections in sequence, handling cutthrough connection
425 for all multi-message initiating connections.
427 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
428 routers. Previously, a multi-recipient message would fail to match the
429 onward-connection opened for the first recipient, and cause its closure.
431 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
432 a timeout on read on a GnuTLS initiating connection, resulting in the
433 initiating connection being dropped. This mattered most when the callout
434 was marked defer_ok. Fix to keep the two timeout-detection methods
437 JH/15 Relax results from ACL control request to enable cutthrough, in
438 unsupported situations, from error to silently (except under debug)
439 ignoring. This covers use with PRDR, frozen messages, queue-only and
442 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
444 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
445 metadata, resulting in a crash in free().
447 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
448 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
449 Broken also in d185889f4, with init system revamp.
451 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
452 Previously we abruptly closed the connection after reading a malware-
453 found indication; now we go on to read the "scan ok" response line,
456 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
457 ACL. Previously, a crash would result.
459 JH/19 Speed up macro lookups during configuration file read, by skipping non-
460 macro text after a replacement (previously it was only once per line) and
461 by skipping builtin macros when searching for an uppercase lead character.
463 JH/20 DANE support moved from Experimental to mainline. The Makefile control
464 for the build is renamed.
466 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
467 was allocated for every new TLS startup, meaning one per message. Fix
468 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
470 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
471 reported the original. Fix to report (as far as possible) the ACL
472 result replacing the original.
474 JH/23 Fix memory leak during multi-message connections using STARTTLS under
475 OpenSSL. Certificate information is loaded for every new TLS startup,
476 and the resources needed to be freed.
478 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
480 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
483 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
484 DATA response info to the (existing) per-recipient response info for
485 the "C=" log element. It can have useful tracking info from the
486 destination system. Patch from Simon Arlott.
488 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
489 length value. Previously this would segfault.
491 HS/02 Support Avast multiline protoocol, this allows passing flags to
492 newer versions of the scanner.
494 JH/28 Ensure that variables possibly set during message acceptance are marked
495 dead before release of memory in the daemon loop. This stops complaints
496 about them when the debug_store option is enabled. Discovered specifically
497 for sender_rate_period, but applies to a whole set of variables.
498 Do the same for the queue-runner and queue-list loops, for variables set
499 from spool message files. Do the same for the SMTP per-message loop, for
500 certain variables indirectly set in ACL operations.
502 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
503 as a multi-recipient message from a mailinglist manager). The coding had
504 an arbitrary cutoff number of characters while checking for more input;
505 enforced by writing a NUL into the buffer. This corrupted long / fast
506 input. The problem was exposed more widely when more pipelineing of SMTP
507 responses was introduced, and one Exim system was feeding another.
508 The symptom is log complaints of SMTP syntax error (NUL chars) on the
509 receiving system, and refused recipients seen by the sending system
510 (propating to people being dropped from mailing lists).
511 Discovered and pinpointed by David Carter.
513 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
514 replaced by the ${authresults } expansion.
516 JH/31 Bug 2257: Fix pipe transport to not use a socket-only syscall.
518 HS/03 Set a handler for SIGTERM and call exit(3) if running as PID 1. This
519 allows proper process termination in container environments.
521 JH/32 Bug 2258: Fix spool_wireformat in combination with LMTP transport.
522 Previously the "final dot" had a newline after it; ensure it is CR,LF.
524 JH/33 SPF: remove support for the "spf" ACL condition outcome values "err_temp"
525 and "err_perm", deprecated since 4.83 when the RFC-defined words
526 "temperror" and "permerror" were introduced.
528 JH/34 Re-introduce enforcement of no cutthrough delivery on transports having
529 transport-filters or DKIM-signing. The restriction was lost in the
530 consolidation of verify-callout and delivery SMTP handling.
531 Extend the restriction to also cover ARC-signing.
533 JH/35 Cutthrough: for a final-dot response timeout (and nonunderstood responses)
534 in defer=pass mode supply a 450 to the initiator. Previously the message
537 PP/02 DANE: add dane_require_tls_ciphers SMTP Transport option; if unset,
538 tls_require_ciphers is used as before.
540 HS/03 Malware Avast: Better match the Avast multiline protocol. Add
541 "pass_unscanned". Only tmpfails from the scanner are written to
542 the paniclog, as they may require admin intervention (permission
543 denied, license issues). Other scanner errors (like decompression
544 bombs) do not cause a paniclog entry.
546 JH/36 Fix reinitialisation of DKIM logging variable between messages.
547 Previously it was possible to log spurious information in receive log
550 JH/37 Bug 2255: Revert the disable of the OpenSSL session caching. This
551 triggered odd behaviour from Outlook Express clients.
553 PP/03 Add util/renew-opendmarc-tlds.sh script for safe renewal of public
556 JH/38 DKIM: accept Ed25519 pubkeys in SubjectPublicKeyInfo-wrapped form,
557 since the IETF WG has not yet settled on that versus the original
558 "bare" representation.
560 JH/39 Fix syslog logging for syslog_timestamp=no and log_selector +millisec.
561 Previously the millisecond value corrupted the output.
562 Fix also for syslog_pid=no and log_selector +pid, for which the pid
563 corrupted the output.
569 JH/01 Rework error string handling in TLS interface so that the caller in
570 more cases is responsible for logging. This permits library-sourced
571 string to be attached to addresses during delivery, and collapses
572 pairs of long lines into single ones.
574 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
575 during configuration. Wildcards are allowed and expanded.
577 JH/02 Rework error string handling in DKIM to pass more info back to callers.
578 This permits better logging.
580 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
581 do not close it down and have the child transport start it up again on
582 the passed-on TCP connection. Instead, proxy the child (and any
583 subsequent ones) for TLS via a unix-domain socket channel. Logging is
584 affected: the continued delivery log lines do not have any DNSSEC, TLS
585 Certificate or OCSP information. TLS cipher information is still logged.
587 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
588 identical IP addresses on different listening ports. Will also affect
591 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
592 add noisy ifdef guards to special-case this sillyness.
593 Patch from Bernd Kuhls.
595 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
596 than 255 are no longer allowed.
598 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
599 Disable the session-cache too, which might reduce our load. Since we
600 currrectly use a new context for every connection, both as server and
601 client, there is no benefit for these.
602 GnuTLS appears to not support tickets server-side by default (we don't
603 call gnutls_session_ticket_enable_server()) but client side is enabled
604 by default on recent versions (3.1.3 +) unless the PFS priority string
607 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
608 <https://reproducible-builds.org/specs/source-date-epoch/>.
610 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
611 the check for any unsuccessful recipients did not notice the limit, and
612 erroneously found still-pending ones.
614 JH/08 Pipeline CHUNKING command and data together, on kernels that support
615 MSG_MORE. Only in-clear (not on TLS connections).
617 JH/09 Avoid using a temporary file during transport using dkim. Unless a
618 transport-filter is involved we can buffer the headers in memory for
619 creating the signature, and read the spool data file once for the
620 signature and again for transmission.
622 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
623 4.77 as the kernel support then wasn't solid, having issues in 64bit
624 mode. Now, it's been long enough. Add support for FreeBSD also.
626 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
627 case where the routing stage had gathered several addresses to send to
628 a host before calling the transport for the first, we previously failed
629 to close down TLS in the old transport process before passing the TCP
630 connection to the new process. The new one sent a STARTTLS command
631 which naturally failed, giving a failed delivery and bloating the retry
632 database. Investigation and fix prototype from Wolfgang Breyha.
634 JH/12 Fix check on SMTP command input synchronisation. Previously there were
635 false-negatives in the check that the sender had not preempted a response
636 or prompt from Exim (running as a server), due to that code's lack of
637 awareness of the SMTP input buffering.
639 PP/04 Add commandline_checks_require_admin option.
640 Exim drops privileges sanely, various checks such as -be aren't a
641 security problem, as long as you trust local users with access to their
642 own account. When invoked by services which pass untrusted data to
643 Exim, this might be an issue. Set this option in main configuration
644 AND make fixes to the calling application, such as using `--` to stop
647 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
648 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
649 responses to those, into a single TLS record each way (this usually means
650 a single packet). As a side issue, smtp_enforce_sync now works on TLS
653 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
654 affects you only if you're dancing at the edge of the param size limits.
655 If you are, and this message makes sense to you, then: raise the
656 configured limit or use OpenSSL 1.1. Nothing we can do for older
659 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
660 cmdline element to get the documented default one. Previously it was
663 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
664 are unsuable for non-message-specific reasons - eg. connection timeout,
665 banner-time rejection.
667 JH/16 Fix logging of delivery remote port, when specified by router, under
670 PP/06 Repair manualroute's ability to take options in any order, even if one
671 is the name of a transport.
674 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
676 JH/17 Change the list-building routines interface to use the expanding-string
677 triplet model, for better allocation and copying behaviour.
679 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
680 Previously it was constructed the first time a possibly-matching string
681 was met in the configuration file input during startup; now it is done
684 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
685 compatible one, to avoid the (poorly documented) possibility of a config
686 file in the working directory redirecting the DB files, possibly correpting
687 some existing file. CVE-2017-10140 assigned for BDB.
689 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
690 cache-hot. Previously, although the result was properly cached, the
691 initial verify call returned a defer.
693 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
694 the main verify for receipient in uncached-mode.
696 JH/22 Retire historical build files to an "unsupported" subdir. These are
697 defined as "ones for which we have no current evidence of testing".
699 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
700 if present. Previously it was ignored.
702 JH/24 Start using specified-initialisers in C structure init coding. This is
703 a C99 feature (it's 2017, so now considered safe).
705 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
706 if was a fixed-sized field and bitmask ops via macros; it is now more
709 PP/07 GitHub PR 56: Apply MariaDB build fix.
710 Patch provided by Jaroslav Škarvada.
712 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
713 during Coverity cleanups [4.87 JH/47]
714 Diagnosis and fix provided by Michael Fischer v. Mollard.
716 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
717 the right size to place the terminating semicolon on its own folded
718 line, the header hash was calculated to an incorrect value thanks to
719 the (relaxed) space the fold became.
721 HS/02 Fix Bug 2130: large writes from the transport subprocess were chunked
722 and confused the parent.
724 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
725 which could crash as a result. This could lead to undeliverable messages.
727 JH/28 Logging: "next input sent too soon" now shows where input was truncated
730 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
731 matters on fast-turnover and PID-randomising systems, which were getting
732 out-of-order delivery.
734 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
735 a possibly-overlapping copy. The symptom was that "Remote host closed
736 connection in response to HELO" was logged instead of the actual 4xx
739 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
740 Previously only that bufferd was discarded, resulting in SYMTP command
743 JH/32 DKIM: when a message has multiple signatures matching an identity given
744 in dkim_verify_signers, run the dkim acl once for each. Previously only
745 one run was done. Bug 2189.
747 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
748 "panic the current process" to "deliberately defer". The panic log is
749 still written with the problem list name; the mail and reject logs now
750 get a temp-reject line for the message that was being handled, saying
751 something like "domains check lookup or other defer". The SMTP 451
752 message is still "Temporary local problem".
754 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
755 A crafted sequence of BDAT commands could result in in-use memory beeing
756 freed. CVE-2017-16943.
758 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
759 from SMTP input. Previously it was always done; now only done for DATA
760 and not BDAT commands. CVE-2017-16944.
762 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
763 to the message (such as an overlong header line). Previously this was
764 not done and we did not exit BDAT mode. Followon from the previous item
765 though a different problem.
771 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
772 than -2003 did; needs libidn2 in addition to libidn.
774 JH/02 The path option on a pipe transport is now expanded before use.
776 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
777 Patch provided by "Björn", documentation fix added too.
779 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
780 missing a wire-to-host endian conversion.
782 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
783 close after a BDAT command line could be taken as a following command,
784 giving a synch failure. Fix by only checking for synch immediately
785 before acknowledging the chunk.
787 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
788 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
789 macro. Patches provided by Josh Soref.
791 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
792 Previously we did not; the RFC seems ambiguous and VRFY is not listed
793 by IANA as a service extension. However, John Klensin suggests that we
796 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
797 the dkim code may be unix-mode line endings rather than smtp wire-format
798 CRLF, so prepend a CR to any bare LF.
800 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
801 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
803 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
804 the first were themselves being wrongly included in the feed into dkim
805 processing; with most chunk sizes in use this resulted in an incorrect
806 body hash calculated value.
808 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
809 DKIM signature block, for verification. Although advised against by
810 standards it is specifically not ruled illegal.
812 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
814 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
815 missing a body hash (the bh= tag).
817 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
818 It seems that HAProxy sends the Proxy Protocol information in clear and
819 only then does a TLS startup, so do the same.
821 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
822 TCP connections (such as for Spamd) unless the daemon successfully set
823 Fast Open mode on its listening sockets. This fixes breakage seen on
824 too-old kernels or those not configured for Fast Open, at the cost of
825 requiring both directions being enabled for TFO, and TFO never being used
826 by non-daemon-related Exim processes.
828 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
829 endings, at least on the first header line. Try to canonify any that get
830 past that check, despite the cost.
832 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
833 now limited to an arbitrary five deep, while parsing addresses with the
834 strip_excess_angle_brackets option enabled.
836 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
837 instead leave the unprompted TLS handshake in socket buffer for the
838 TLS library to consume.
840 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
842 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
844 JH/16 Drop variables when they go out of scope. Memory management drops a whole
845 region in one operation, for speed, and this leaves assigned pointers
846 dangling. Add checks run only under the testsuite which checks all
847 variables at a store-reset and panics on a dangling pointer; add code
848 explicitly nulling out all the variables discovered. Fixes one known
849 bug: a transport crash, where a dangling pointer for $sending_ip_address
850 originally assigned in a verify callout, is re-used.
852 PP/06 Drop '.' from @INC in various Perl scripts.
854 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
856 PP/08 Reduce a number of compilation warnings under clang; building with
857 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
858 should be warning-free.
860 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
862 HS/01 Fix portability problems introduced by PP/08 for platforms where
863 realloc(NULL) is not equivalent to malloc() [SunOS et al].
865 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
866 chunk. This allows us to accept broken chunked messages. We need a more
867 general solution here.
869 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
870 already-broken messages in the queue.
872 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
874 JH/19 Fix reference counting bug in routing-generated-address tracking.
880 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
881 supports it and a size is available (ie. the sending peer gave us one).
883 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
884 years of being deprecated). The replacements are the ACLs
885 acl_smtp_mime and acl_not_smtp_mime.
887 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
888 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
889 or even an in-clear connection were permitted. Now, if the host lookup
890 was dnssec and dane was requested then the host is only used if the
891 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
892 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
893 if one fails this test.
894 This means that a poorly-configured remote DNS will make it incommunicado;
895 but it protects against a DNS-interception attack on it.
897 JH/04 Bug 1810: make continued-use of an open smtp transport connection
898 non-noisy when a race steals the message being considered.
900 JH/05 If main configuration option tls_certificate is unset, generate a
901 self-signed certificate for inbound TLS connections.
903 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
904 in rewrites and routers.
906 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
907 and logged a warning sing 4.83; now they are a configuration file error.
909 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
910 (lacking @domain). Apply the same qualification processing as RCPT.
912 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
914 JH/10 Support ${sha256:} applied to a string (as well as the previous
917 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
918 a cutthrough deliver is pending, as we always want to make a connection.
919 This also avoids re-routing the message when later placing the cutthrough
920 connection after a verify cache hit.
921 Do not update it with the verify result either.
923 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
924 when routing results in more than one destination address.
926 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
927 signing (which inhibits the cutthrough capability). Previously only
928 the presence of an option was tested; now an expansion evaluating as
929 empty is permissible (obviously it should depend only on data available
930 when the cutthrough connection is made).
932 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
933 the relevant preceding SMTP command did not note the pipelining mode.
935 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
936 Previously they were not counted.
938 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
939 as one having no matching records. Previously we deferred the message
940 that needed the lookup.
942 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
943 distinguished as "(=".
945 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
946 for missing MX records. Previously it only worked for missing A records.
948 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
950 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
951 after the data-go-ahead and data-ack. Patch from Jason Betts.
953 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
954 even for a "none" policy. Patch from Tony Meyer.
956 JH/22 Fix continued use of a connection for further deliveries. If a port was
957 specified by a router, it must also match for the delivery to be
960 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
961 When one of the recipients of a message was unsuitable for the connection
962 (has no matching addresses), we lost track of needing to mark it
963 deferred. As a result mail would be lost.
965 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
967 JH/25 Decoding ACL controls is now done using a binary search; the source code
968 takes up less space and should be simpler to maintain. Merge the ACL
969 condition decode tables also, with similar effect.
971 JH/26 Fix problem with one_time used on a redirect router which returned the
972 parent address unchanged. A retry would see the parent address marked as
973 delivered, so not attempt the (identical) child. As a result mail would
976 JH/27 Fix a possible security hole, wherein a process operating with the Exim
977 UID can gain a root shell. Credit to http://www.halfdog.net/ for
978 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
981 JH/28 Enable {spool,log} filesystem space and inode checks as default.
982 Main config options check_{log,spool}_{inodes,space} are now
983 100 inodes, 10MB unless set otherwise in the configuration.
985 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
986 Previously it only applied to the main-section connection policy
989 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
991 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
992 by me. Added RFC7919 DH primes as an alternative.
994 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
995 are not in the system include path.
997 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
998 GnuTLS, when a session startup failed (eg because the client disconnected)
999 Exim did stdio operations after fclose. This was exposed by a recent
1000 change which nulled out the file handle after the fclose.
1002 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
1003 signed directly by the cert-signing cert, rather than an intermediate
1004 OCSP-signing cert. This is the model used by LetsEncrypt.
1006 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
1008 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
1009 an incoming connection.
1011 HS/02 Bug 1802: Do not half-close the connection after sending a request
1014 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
1015 fallback to "prime256v1".
1017 JH/34 SECURITY: Use proper copy of DATA command in error message.
1018 Could leak key material. Remotely exploitable. CVE-2016-9963.
1024 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
1025 and 3.4.4 - once the server is enabled to respond to an OCSP request
1026 it does even when not requested, resulting in a stapling non-aware
1027 client dropping the TLS connection.
1029 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
1030 support variable-length bit vectors. No functional change.
1032 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
1033 The I= interface field on outgoing lines is now after the H= remote
1034 host field, same as incoming lines. There is a separate
1035 outgoing_interface log selector which allows you to disable the
1038 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
1039 If not running log_selector +smtp_connection the mainlog would be held
1040 open indefinitely after a "too many connections" event, including to a
1041 deleted file after a log rotate. Leave the per net connection logging
1042 leaving it open for efficiency as that will be quickly detected by the
1043 check on the next write.
1045 HS/01 Bug 1671: Fix post transport crash.
1046 Processing the wait-<transport> messages could crash the delivery
1047 process if the message IDs didn't exist for some reason. When
1048 using 'split_spool_directory=yes' the construction of the spool
1049 file name failed already, exposing the same netto behaviour.
1051 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
1052 mime_regex ACL conditions.
1054 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
1055 to DSN fail messages (bounces): remote IP, remote greeting, remote response
1056 to HELO, local diagnostic string.
1058 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
1059 log line to debug. Even when configured with a tls authenticator many
1060 client connections are expected to not authenticate in this way, so
1061 an authenticate fail is not an error.
1063 HS/02 Add the Exim version string to the process info. This way exiwhat
1064 gives some more detail about the running daemon.
1066 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
1067 matter for fast-change records such as DNSBLs.
1069 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
1070 retry record, even if constant. There may be multiple transports with
1071 different interface settings and the retry behaviour needs to be kept
1074 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
1076 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
1078 JH/10 Bug 840: fix log_defer_output option of pipe transport
1080 JH/11 Bug 830: use same host for all RCPTS of a message, even under
1081 hosts_randomize. This matters a lot when combined with mua_wrapper.
1083 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
1084 ${quote_pgsql:<string>} operator.
1086 JH/13 Bug 1708: avoid misaligned access in cached lookup.
1088 JH/14 Change header file name for freeradius-client. Relevant if compiling
1089 with Radius support; from the Gentoo tree and checked under Fedora.
1091 JH/15 Bug 1712: Introduce $prdr_requested flag variable
1093 JH/16 Bug 1714: Permit an empty string as expansion result for transport
1094 option transport_filter, meaning no filtering.
1096 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
1098 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
1099 defaults to "*" (all hosts). The variable is now available when not built
1100 with TLS, default unset, mainly to enable keeping the testsuite sane.
1101 If a server certificate is not supplied (via tls_certificate) an error is
1102 logged, and clients will find TLS connections fail on startup. Presumably
1103 they will retry in-clear.
1104 Packagers of Exim are strongly encouraged to create a server certificate
1105 at installation time.
1107 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
1108 with the $config_file variable.
1110 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
1111 in transport context, after the attempt, and per-recipient. The latter type
1112 is per host attempted. The event data is the error message, and the errno
1113 information encodes the lookup type (A vs. MX) used for the (first) host,
1114 and the trailing two digits of the smtp 4xx response.
1116 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
1117 to write to mainlog (or rejectlog, paniclog) in the window between file
1118 creation and permissions/ownership being changed. Particularly affects
1119 installations where exicyclog is run as root, rather than exim user;
1120 result is that the running daemon panics and dies.
1122 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
1124 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
1125 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
1126 "pri" and "weight". Note that the previous implicit priority given by the
1127 list order is no longer honoured.
1129 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
1130 for DKIM processing.
1132 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
1133 by defining SUPPORT_SOCKS.
1135 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
1136 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
1137 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
1138 variables are renamed to proxy_{local,external}_{address,port}.
1140 JH/27 Move Internationalisation support from Experimental to mainline, enabled
1141 for a build by defining SUPPORT_I18N
1143 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
1144 of the query string, and make ${quote_redis:} do that quoting.
1146 JH/29 Move Events support from Experimental to mainline, enabled by default
1147 and removable for a build by defining DISABLE_EVENT.
1149 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
1151 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
1152 cached by the daemon.
1154 JH/32 Move Redis support from Experimental to mainline, enabled for a build
1155 by defining LOOKUP_REDIS. The libhiredis library is required.
1157 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
1158 keys are given for lookup.
1160 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
1161 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
1162 only supported when built with TLS support. The PolarSSL SHA routines
1163 are still used when the TLS library is too old for convenient support.
1165 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
1166 openssl_options), for security. OpenSSL forces this from version 1.1.0
1167 server-side so match that on older versions.
1169 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
1170 allocation for $value could be released as the expansion processing
1171 concluded, but leaving the global pointer active for it.
1173 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
1174 and to use the domains and local_parts ACL conditions.
1176 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
1177 incorrectly not doubled on cutthrough transmission, hence seen as a
1178 body-termination at the receiving system - resulting in truncated mails.
1179 Commonly the sender saw a TCP-level error, and retransmitted the message
1180 via the normal store-and-forward channel. This could result in duplicates
1181 received - but deduplicating mailstores were liable to retain only the
1182 initial truncated version.
1184 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
1186 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
1188 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
1189 we're in there, support oversigning also; bug 1309.
1191 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
1193 HS/04 Add support for keep_environment and add_environment options.
1195 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
1196 either intentional arithmetic overflow during PRNG, or testing config-
1199 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
1200 delivery resulted in actual delivery. Cancel cutthrough before DATA
1203 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
1204 rejects a recipient: pass the reject to the originator.
1206 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
1207 Many were false-positives and ignorable, but it's worth fixing the
1210 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
1211 for the new environment-manipulation done at startup. Move the routines
1212 from being local to tls.c to being global via the os.c file.
1214 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
1215 an extract embedded as result-arg for a map, the first arg for extract
1216 is unavailable so we cannot tell if this is a numbered or keyed
1217 extraction. Accept either.
1223 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
1226 JH/02 The smtp transport option "multi_domain" is now expanded.
1228 JH/03 The smtp transport now requests PRDR by default, if the server offers
1231 JH/04 Certificate name checking on server certificates, when exim is a client,
1232 is now done by default. The transport option tls_verify_cert_hostnames
1233 can be used to disable this per-host. The build option
1234 EXPERIMENTAL_CERTNAMES is withdrawn.
1236 JH/05 The value of the tls_verify_certificates smtp transport and main options
1237 default to the word "system" to access the system default CA bundle.
1238 For GnuTLS, only version 3.0.20 or later.
1240 JH/06 Verification of the server certificate for a TLS connection is now tried
1241 (but not required) by default. The verification status is now logged by
1242 default, for both outbound TLS and client-certificate supplying inbound
1245 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
1248 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
1249 Status Notification (bounce) messages are now MIME format per RFC 3464.
1250 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
1251 under the control of the dsn_advertise_hosts option, and routers may
1252 have a dsn_lasthop option.
1254 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
1255 default, modifiable by a malware= option. The list separator for
1256 the options can now be changed in the usual way. Bug 68.
1258 JH/10 The smtp_receive_timeout main option is now expanded before use.
1260 JH/11 The incoming_interface log option now also enables logging of the
1261 local interface on delivery outgoing connections.
1263 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
1264 if the interface and destination host and port all match.
1266 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
1269 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
1270 Patch from Andrew Lewis.
1272 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
1273 now supports optional time-restrictions, weighting, and priority
1274 modifiers per server. Patch originally by <rommer@active.by>.
1276 JH/16 The spamd_address main option now supports a mixed list of local
1277 and remote servers. Remote servers can be IPv6 addresses, and
1278 specify a port-range.
1280 JH/17 Bug 68: The spamd_address main option now supports an optional
1281 timeout value per server.
1283 JH/18 Bug 1581: Router and transport options headers_add/remove can
1284 now have the list separator specified.
1286 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
1289 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
1292 JH/21 Support for the A6 type of dns record is withdrawn.
1294 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
1295 rather than the verbs used.
1297 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
1298 from 255 to 1024 chars.
1300 JH/24 Verification callouts now attempt to use TLS by default.
1302 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
1303 are generic router options now. The defaults didn't change.
1305 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
1306 Original patch from Alexander Shikoff, worked over by JH.
1308 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
1309 files if ZCAT_COMMAND is not executable.
1311 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
1313 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
1315 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
1316 Normally benign, it bites when the pair was led to by a CNAME;
1317 modern usage is to not canonicalize the domain to a CNAME target
1318 (and we were inconsistent anyway for A-only vs AAAA+A).
1320 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
1322 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
1323 when evaluating $sender_host_dnssec.
1325 JH/31 Check the HELO verification lookup for DNSSEC, adding new
1326 $sender_helo_dnssec variable.
1328 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
1330 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
1332 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
1334 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
1335 documented as working, but never had. Support all but $spam_report.
1337 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
1338 added for tls authenticator.
1340 HS/03 Add perl_taintmode main config option
1346 TL/01 When running the test suite, the README says that variables such as
1347 no_msglog_check are global and can be placed anywhere in a specific
1348 test's script, however it was observed that placement needed to be near
1349 the beginning for it to behave that way. Changed the runtest perl
1350 script to read through the entire script once to detect and set these
1351 variables, reset to the beginning of the script, and then run through
1352 the script parsing/test process like normal.
1354 TL/02 The BSD's have an arc4random API. One of the functions to induce
1355 adding randomness was arc4random_stir(), but it has been removed in
1356 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
1357 function when detected.
1359 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
1360 cause callback expansion.
1362 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
1363 syntax errors in an expansion can be treated as a string instead of
1364 logging or causing an error, due to the internal use of bool_lax
1365 instead of bool when processing it.
1367 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
1368 server certificates when making smtp deliveries.
1370 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
1372 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
1374 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
1376 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
1377 Merged patch from Sebastian Wiedenroth.
1379 JH/05 Fix results-pipe from transport process. Several recipients, combined
1380 with certificate use, exposed issues where response data items split
1381 over buffer boundaries were not parsed properly. This eventually
1382 resulted in duplicates being sent. This issue only became common enough
1383 to notice due to the introduction of connection certificate information,
1384 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1386 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1387 size buffer was used, resulting in syntax errors when an expansion
1390 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1391 version 3.3.6 or later.
1393 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1394 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1395 both become "event_action", the variables become $event_name, $event_data
1396 and $event_defer_errno. There is a new variable $verify_mode, usable in
1397 routers, transports and related events. The tls:cert event is now also
1398 raised for inbound connections, if the main configuration event_action
1401 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1402 early OCSP support, but no stapling (appears to be less than 1.0.0).
1404 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1405 server certificate names available under the smtp transport option
1406 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1409 JH/10 Time-related extraction expansions from certificates now use the main
1410 option "timezone" setting for output formatting, and are consistent
1411 between OpenSSL and GnuTLS compilations. Bug 1541.
1413 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1414 encoded parameter in the incoming message. Bug 1558.
1416 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1417 include certificate info, eximon was claiming there were spoolfile
1420 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1422 JH/14 Log delivery-related information more consistently, using the sequence
1423 "H=<name> [<ip>]" wherever possible.
1425 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1426 are problematic for Debian distribution, omit them from the release
1429 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1431 JH/16 Fix string representation of time values on 64bit time_t architectures.
1434 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1435 output list separator was used.
1440 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1441 checkers that were complaining about end of non-void function with no
1444 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1445 This was a regression introduced in 4.83 by another bugfix.
1447 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1449 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1450 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1456 TF/01 Correctly close the server side of TLS when forking for delivery.
1458 When a message was received over SMTP with TLS, Exim failed to clear up
1459 the incoming connection properly after forking off the child process to
1460 deliver the message. In some situations the subsequent outgoing
1461 delivery connection happened to have the same fd number as the incoming
1462 connection previously had. Exim would try to use TLS and fail, logging
1463 a "Bad file descriptor" error.
1465 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1466 utilities have not been installed.
1468 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1469 temporary space as the ACL may create new global variables.
1471 TL/01 LDAP support uses per connection or global context settings, depending
1472 upon the detected version of the libraries at build time.
1474 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1475 to extract and use the src ip:port in logging and expansions as if it
1476 were a direct connection from the outside internet. PPv2 support was
1477 updated based on HAProxy spec change in May 2014.
1479 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1481 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1482 Properly escape header and check for NULL return.
1484 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1487 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1489 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1490 characters in header names, implemented as a verify condition.
1491 Contributed by Michael Fischer v. Mollard.
1493 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1494 results permerror and temperror. Previous values are deprecated but
1495 still accepted. In a future release, err_perm and err_temp will be
1496 completely removed, which will be a backward incompatibility if the
1497 ACL tests for either of these two old results. Patch contributed by
1498 user bes-internal on the mailing list.
1500 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1502 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1503 selectors, in both main and reject logs.
1505 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1508 JH/07 Add malware type "sock" for talking to simple daemon.
1510 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1512 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1513 routers/transports under cutthrough routing.
1515 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1516 numbers. Touch up "bool" conditional to keep the same definition.
1518 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1520 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1522 JH/12 Expand items in router/transport headers_add or headers_remove lists
1523 individually rather than the list as a whole. Bug 1452.
1525 Required for reasonable handling of multiple headers_ options when
1526 they may be empty; requires that headers_remove items with embedded
1527 colons must have them doubled (or the list-separator changed).
1529 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1530 view the policy declared in the DMARC record. Currently, $dmarc_status
1531 is a combined value of both the record presence and the result of the
1534 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1536 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1537 dnslookup router and the smtp transport (applying to the forward
1540 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1541 of ldap servers used for a specific lookup. Patch provided by Heiko
1544 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1545 New variable $lookup_dnssec_authenticated for observability.
1547 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1548 Patch submitted by Lars Timman.
1550 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1552 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1553 Requires trusted mode and valid format message id, aborts otherwise.
1554 Patch contributed by Heiko Schlichting.
1556 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1557 certextract with support for various fields. Bug 1358.
1559 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1560 is requested by default, modifiable by smtp transport option
1563 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1564 operate on certificate variables to give certificate fingerprints
1565 Also new ${sha256:cert_variable}.
1567 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1569 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1570 Christian Aistleitner.
1572 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1574 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1575 file. Patch from Wolfgang Breyha.
1577 JH/25 Expand the coverage of the delivery $host and $host_address to
1578 client authenticators run in verify callout. Bug 1476.
1580 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1581 align with daemon_smtp_ports. Bug 72.
1583 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1584 support and error reporting did not work properly.
1586 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1587 and is readable. Patch from Andrew Colin Kissa.
1589 TL/14 Enhance documentation of ${run expansion and how it parses the
1590 commandline after expansion, particularly in the case when an
1591 unquoted variable expansion results in an empty value.
1593 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1595 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1596 by renaming away from it. Addresses GH issue 15, reported by
1599 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1601 TL/15 SECURITY: prevent double expansion in math comparison functions
1602 (can expand unsanitized data). Not remotely exploitable.
1609 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1611 PP/02 Make -n do something, by making it not do something.
1612 When combined with -bP, the name of an option is not output.
1614 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1617 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1618 $sender_host_name and config options to manage this, and basic check
1621 PP/05 DSCP support for outbound connections and control modifier for inbound.
1623 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1624 (Only plugin which currently uses this is kerberos4, which nobody should
1625 be using, but we should make it available and other future plugins might
1626 conceivably use it, even though it would break NAT; stuff *should* be
1627 using channel bindings instead).
1629 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1630 name; added for Sendmail compatibility; requires admin caller.
1631 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1632 just ignore it); requires trusted caller.
1633 Also parse but ignore: -Ac -Am -X<logfile>
1636 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1638 TL/02 Add +smtp_confirmation as a default logging option.
1640 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1641 Patch by Magnus Holmgren from 2007-02-20.
1643 TL/04 Bugzilla 1281 - Spec typo.
1644 Bugzilla 1283 - Spec typo.
1645 Bugzilla 1290 - Spec grammar fixes.
1647 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1649 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1651 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1652 mailing list by by Dmitry Isaikin.
1654 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1656 JH/02 Support "G" suffix to numbers in ${if comparisons.
1658 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1660 NM/01 Bugzilla 1197 - Spec typo
1661 Bugzilla 1196 - Spec examples corrections
1663 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1665 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1666 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1669 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1670 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1672 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1673 "acl {{name}{arg}...}", and optional args on acl condition
1676 JH/05 Permit multiple router/transport headers_add/remove lines.
1678 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1680 JH/07 Avoid using a waiting database for a single-message-only transport.
1681 Performance patch from Paul Fisher. Bugzilla 1262.
1683 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1686 JH/09 Add $headers_added variable, with content from use of ACL modifier
1687 add_header (but not yet added to the message). Bugzilla 199.
1689 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1690 Pulled from Bugzilla 817 by Wolfgang Breyha.
1692 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1694 (nb: this is the same fix as in Exim 4.80.1)
1696 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1699 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1702 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1704 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1705 advertise SMTP AUTH mechanism to us, instead of a generic
1706 protocol violation error. Also, make Exim more robust to bad
1707 data from the Dovecot auth socket.
1709 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1711 When a queue runner is handling a message, Exim first routes the
1712 recipient addresses, during which it prunes them based on the retry
1713 hints database. After that it attempts to deliver the message to
1714 any remaining recipients. It then updates the hints database using
1717 So if a recipient address works intermittently, it can get repeatedly
1718 deferred at routing time. The retry hints record remains fresh so the
1719 address never reaches the final cutoff time.
1721 This is a fairly common occurrence when a user is bumping up against
1722 their storage quota. Exim had some logic in its local delivery code
1723 to deal with this. However it did not apply to per-recipient defers
1724 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1726 This change adds a proper retry rule check during routing so that the
1727 final cutoff time is checked against the message's age. We only do
1728 this check if there is an address retry record and there is not a
1729 domain retry record; this implies that previous attempts to handle
1730 the address had the retry_use_local_parts option turned on. We use
1731 this as an approximation for the destination being like a local
1732 delivery, as in LMTP.
1734 I suspect this new check makes the old local delivery cutoff check
1735 redundant, but I have not verified this so I left the code in place.
1737 TF/02 Correct gecos expansion when From: is a prefix of the username.
1739 Test 0254 submits a message to Exim with the header
1743 When I ran the test suite under the user fanf2, Exim expanded
1744 the header to contain my full name, whereas it should have added
1745 a Resent-Sender: header. It erroneously treats any prefix of the
1746 username as equal to the username.
1748 This change corrects that bug.
1750 GF/01 DCC debug and logging tidyup
1751 Error conditions log to paniclog rather than rejectlog.
1752 Debug lines prefixed by "DCC: " to remove any ambiguity.
1754 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1756 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1757 Bug spotted by Jeremy Harris; was flawed since initial commit.
1758 Would have resulted in OCSP responses post-SNI triggering an Exim
1759 NULL dereference and crash.
1761 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1763 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1764 Bug detection, analysis and fix by Samuel Thibault.
1765 Bugzilla 1331, Debian bug #698092.
1767 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1769 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1770 Server implementation by Todd Lyons, client by JH.
1771 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1772 config variable "prdr_enable" controls whether the server
1773 advertises the facility. If the client requests PRDR a new
1774 acl_data_smtp_prdr ACL is called once for each recipient, after
1775 the body content is received and before the acl_smtp_data ACL.
1776 The client is controlled by both of: a hosts_try_prdr option
1777 on the smtp transport, and the server advertisement.
1778 Default client logging of deliveries and rejections involving
1779 PRDR are flagged with the string "PRDR".
1781 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1782 fclose(). Diagnosis by Todd Lyons.
1784 PP/17 Update configure.default to handle IPv6 localhost better.
1785 Patch by Alain Williams (plus minor tweaks).
1788 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1789 This is now consistent with GnuTLS, and is now documented: the
1790 previous undocumented portable approach to treating the option as
1791 unset was to force an expansion failure. That still works, and
1792 an empty string is now equivalent.
1794 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1795 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1796 not performing validation itself.
1798 PP/20 Added force_command boolean option to pipe transport.
1799 Patch from Nick Koston, of cPanel Inc.
1801 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1804 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1806 PP/21 Fix eximon continuous updating with timestamped log-files.
1807 Broken in a format-string cleanup in 4.80, missed when I repaired the
1808 other false fix of the same issue.
1809 Report and fix from Heiko Schlichting.
1812 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1813 Report from Prashanth Katuri.
1815 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1816 It's SecureTransport, so affects any MacOS clients which use the
1817 system-integrated TLS libraries, including email clients.
1819 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1820 using a MIME ACL for non-SMTP local injection.
1821 Report and assistance in diagnosis by Warren Baker.
1823 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1825 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1827 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1828 last id that failed so it may be referenced in subsequent ACL's.
1830 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1833 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1834 ldap library initialization, allowing self-signed CA's to be
1835 used. Also properly sets require_cert option later in code by
1836 using NULL (global ldap config) instead of ldap handle (per
1837 session). Bug diagnosis and testing by alxgomz.
1839 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1840 the src/util/ subdirectory.
1842 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1843 renamed to Transport Post Delivery Action by Jeremy Harris, as
1846 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1847 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1848 redis_servers = needs to be configured which will be used by the redis
1849 lookup. Patch from Warren Baker, of The Packet Hub.
1851 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1853 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1854 hostname or reverse DNS when processing a host list. Used suggestions
1855 from multiple comments on this bug.
1857 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1859 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1860 Missed a few lines, added it to make the runtest require no keyboard
1863 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1864 contains upper case chars. Make router use caseful_local_part.
1866 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1867 support when GnuTLS has been built with p11-kit.
1873 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1875 This, or similar/improved, will also be change PP/11 of 4.82.
1881 PP/01 Handle short writes when writing local log-files.
1882 In practice, only affects FreeBSD (8 onwards).
1883 Bugzilla 1053, with thanks to Dmitry Isaikin.
1885 NM/01 Bugzilla 949 - Documentation tweak
1887 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1890 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1892 PP/02 Implemented gsasl authenticator.
1894 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1896 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1897 `pkg-config foo` for cflags/libs.
1899 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1900 with rest of GSASL and with heimdal_gssapi.
1902 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1903 `pkg-config foo` for cflags/libs for the TLS implementation.
1905 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1906 properties get this fed in as external SSF. A number of robustness
1907 and debugging improvements to the cyrus_sasl authenticator.
1909 PP/08 cyrus_sasl server now expands the server_realm option.
1911 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1912 Patch by Jeremy Harris.
1914 PP/10 Added dbmjz lookup type.
1916 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1918 PP/12 MAIL args handles TAB as well as SP, for better interop with
1919 non-compliant senders.
1920 Analysis and variant patch by Todd Lyons.
1922 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1923 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1924 Patch from Dirk Mueller <dmueller@suse.com>
1926 PP/13 tls_peerdn now print-escaped for spool files.
1927 Observed some $tls_peerdn in wild which contained \n, which resulted
1928 in spool file corruption.
1930 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1931 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1932 or write after TLS renegotiation, which otherwise led to messages
1935 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1936 as a tracking header (ie: a signed header comes before the signature).
1937 Patch from Wolfgang Breyha.
1939 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1940 comma-sep list; embedded commas doubled.
1942 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1944 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1946 Report and patch from Dmitry Banschikov.
1948 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1949 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1950 failures. We appear to now support TLS1.1+ with Exim.
1952 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1953 lets Exim select keys and certificates based upon TLS SNI from client.
1954 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1955 before an outbound SMTP session. New log_selector, +tls_sni.
1957 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1958 NULL dereference. Report and patch from Alun Jones.
1960 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1961 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1962 Not seeing resolver debug output on NetBSD, but suspect this is a
1963 resolver implementation change.
1965 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1966 Left warnings. Added "eximon gdb" invocation mode.
1968 PP/21 Defaulting "accept_8bitmime" to true, not false.
1970 PP/22 Added -bw for inetd wait mode support.
1972 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1973 locate the relevant includes and libraries. Made this the default.
1975 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1976 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1978 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1979 This may cause build issues on older platforms.
1981 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1982 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1983 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1984 Added SNI support via GnuTLS too.
1985 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1987 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1989 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1990 Applied second patch from Janne, implementing suggestion to default
1991 multiple-strings-in-record handling to match SPF spec.
1993 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1995 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1996 read-only, out of scope).
1997 Patch from Wolfgang Breyha, report from Stuart Northfield.
1999 PP/29 Fix three issues highlighted by clang analyser static analysis.
2000 Only crash-plausible issue would require the Cambridge-specific
2001 iplookup router and a misconfiguration.
2002 Report from Marcin Mirosław.
2004 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
2006 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
2007 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
2008 As part of this, removing so much warning spew let me fix some minor
2009 real issues in debug logging.
2011 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
2012 assignment on my part. Fixed.
2014 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
2015 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
2016 Janne Snabb (who went above and beyond: thank you).
2018 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
2019 string otherwise requires a connection and a bunch more work and it's
2020 relatively easy to get wrong. Should also expose TLS library linkage
2023 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
2024 64-bit ${eval} (JH/03).
2026 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
2027 GNU libc to support some of the 64-bit stuff, should not lead to
2028 conflicts. Defined before os.h is pulled in, so if a given platform
2029 needs to override this, it can.
2031 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
2032 protection layer was required, which is not implemented.
2033 Bugzilla 1254, patch from Wolfgang Breyha.
2035 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
2036 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
2037 tls_dhparam take prime identifiers. Also unbreak combination of
2038 OpenSSL+DH_params+TLSSNI.
2040 PP/39 Disable SSLv2 by default in OpenSSL support.
2046 PP/01 Solaris build fix for Oracle's LDAP libraries.
2047 Bugzilla 1109, patch from Stephen Usher.
2049 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
2051 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
2054 TF/02 Fix a couple more cases where we did not log the error message
2055 when unlink() failed. See also change 4.74-TF/03.
2057 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
2058 lock up or crash if it happened to be inside a call to libc when it
2059 got a SIGUSR1 from exiwhat.
2061 The SIGUSR1 handler appends the current process status to the process
2062 log which is later printed by exiwhat. It used to use the general
2063 purpose logging code to do this, but several functions it calls are
2064 not safe for signals.
2066 The new output code in the SIGUSR1 handler is specific to the process
2067 log, and simple enough that it's easy to inspect for signal safety.
2068 Removing some special cases also simplifies the general logging code.
2069 Removing the spurious timestamps from the process log simplifies
2072 TF/04 Improved ratelimit ACL condition.
2074 The /noupdate option has been deprecated in favour of /readonly which
2075 has clearer semantics. The /leaky, /strict, and /readonly update modes
2076 are mutually exclusive. The update mode is no longer included in the
2077 database key; it just determines when the database is updated. (This
2078 means that when you upgrade Exim will forget old rate measurements.)
2080 Exim now checks that the per_* options are used with an update mode that
2081 makes sense for the current ACL. For example, when Exim is processing a
2082 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
2083 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
2084 must specify per_mail/readonly. If you omit the update mode it defaults to
2085 /leaky where that makes sense (as before) or /readonly where required.
2087 The /noupdate option is now undocumented but still supported for
2088 backwards compatibility. It is equivalent to /readonly except that in
2089 ACLs where /readonly is required you may specify /leaky/noupdate or
2090 /strict/noupdate which are treated the same as /readonly.
2092 A useful new feature is the /count= option. This is a generalization
2093 of the per_byte option, so that you can measure the throughput of other
2094 aggregate values. For example, the per_byte option is now equivalent
2095 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
2097 The per_rcpt option has been generalized using the /count= mechanism
2098 (though it's more complicated than the per_byte equivalence). When it is
2099 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
2100 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
2101 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
2102 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
2103 Note that using per_rcpt with a non-readonly update mode in more than
2104 one ACL will cause the recipients to be double-counted. (The per_mail
2105 and per_byte options don't have this problem.)
2107 The handling of very low rates has changed slightly. If the computed rate
2108 is less than the event's count (usually one) then this event is the first
2109 after a long gap. In this case the rate is set to the same as this event's
2110 count, so that the first message of a spam run is counted properly.
2112 The major new feature is a mechanism for counting the rate of unique
2113 events. The new per_addr option counts the number of different
2114 recipients that someone has sent messages to in the last time period. It
2115 behaves like per_rcpt if all the recipient addresses are different, but
2116 duplicate recipient addresses do not increase the measured rate. Like
2117 the /count= option this is a general mechanism, so the per_addr option
2118 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
2119 example, measure the rate that a client uses different sender addresses
2120 with the options per_mail/unique=$sender_address. There are further
2121 details in the main documentation.
2123 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
2125 TF/06 Removed a few PCRE remnants.
2127 TF/07 Automatically extract Exim's version number from tags in the git
2128 repository when doing development or release builds.
2130 PP/02 Raise smtp_cmd_buffer_size to 16kB.
2131 Bugzilla 879. Patch from Paul Fisher.
2133 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
2134 Heavily based on revision 40f9a89a from Simon Arlott's tree.
2137 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
2139 PP/05 Variable $av_failed, true if the AV scanner deferred.
2140 Bugzilla 1078. Patch from John Horne.
2142 PP/06 Stop make process more reliably on build failure.
2143 Bugzilla 1087. Patch from Heiko Schlittermann.
2145 PP/07 Make maildir_use_size_file an _expandable_ boolean.
2146 Bugzilla 1089. Patch from Heiko Schlittermann.
2148 PP/08 Handle ${run} returning more data than OS pipe buffer size.
2149 Bugzilla 1131. Patch from Holger Weiß.
2151 PP/09 Handle IPv6 addresses with SPF.
2152 Bugzilla 860. Patch from Wolfgang Breyha.
2154 PP/10 GnuTLS: support TLS 1.2 & 1.1.
2156 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
2159 PP/11 match_* no longer expand right-hand-side by default.
2160 New compile-time build option, EXPAND_LISTMATCH_RHS.
2161 New expansion conditions, "inlist", "inlisti".
2163 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
2165 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
2167 PP/14 fix log_write() format string regression from TF/03.
2168 Bugzilla 1152. Patch from Dmitry Isaikin.
2174 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
2176 PP/02 Harmonised TLS library version reporting; only show if debugging.
2177 Layout now matches that introduced for other libraries in 4.74 PP/03.
2179 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
2181 PP/04 New "dns_use_edns0" global option.
2183 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
2186 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
2187 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
2189 TK/01 Updated PolarSSL code to 0.14.2.
2190 Bugzilla 1097. Patch from Andreas Metzler.
2192 PP/07 Catch divide-by-zero in ${eval:...}.
2193 Fixes bugzilla 1102.
2195 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
2198 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
2199 format-string attack -- SECURITY: remote arbitrary code execution.
2201 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
2202 time unintentionally subject to list matching rules, letting the header
2203 cause arbitrary Exim lookups (of items which can occur in lists, *not*
2204 arbitrary string expansion). This allowed for information disclosure.
2206 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
2207 INT_MIN/-1 -- value coerced to INT_MAX.
2213 NM/01 Workaround for PCRE version dependency in version reporting
2216 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
2217 This fixes portability to compilers other than gcc, notably
2218 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
2220 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
2221 makefiles for portability to HP-UX and POSIX correctness.
2223 PP/01 Permit LOOKUP_foo enabling on the make command-line.
2224 Also via indented variable definition in the Makefile.
2225 (Debugging by Oliver Heesakkers).
2227 PP/02 Restore caching of spamd results with expanded spamd_address.
2228 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
2230 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
2231 Improves build reliability. Fix from: Frank Elsner
2233 NM/02 Fix wide character breakage in the rfc2047 coding
2234 Fixes bug 1064. Patch from Andrey N. Oktyabrski
2236 NM/03 Allow underscore in dnslist lookups
2237 Fixes bug 1026. Patch from Graeme Fowler
2239 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
2240 Code patches from Adam Ciarcinski of NetBSD.
2242 NM/04 Fixed exiqgrep to cope with mailq missing size issue
2245 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
2246 is logged, to avoid truncation. Patch from John Horne.
2248 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
2249 Patch from Jakob Hirsch.
2251 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
2252 SQL string expansion failure details.
2253 Patch from Andrey Oktyabrski.
2255 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
2256 Patch from Simon Arlott.
2258 PP/09 New lookups functionality failed to compile on old gcc which rejects
2259 extern declarations in function scope.
2260 Patch from Oliver Fleischmann
2262 PP/10 Use sig_atomic_t for flags set from signal handlers.
2263 Check getgroups() return and improve debugging.
2264 Fixed developed for diagnosis in bug 927 (which turned out to be
2267 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
2268 Patch from Mark Zealey.
2270 PP/12 Bugzilla 1056: Improved spamd server selection.
2271 Patch from Mark Zealey.
2273 PP/13 Bugzilla 1086: Deal with maildir quota file races.
2274 Based on patch from Heiko Schlittermann.
2276 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
2277 Patch from Uwe Doering, sign-off by Michael Haardt.
2279 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
2280 variable declaration deep within a block. Bug and patch from
2283 PP/15 lookups-Makefile IRIX compatibility coercion.
2285 PP/16 Make DISABLE_DKIM build knob functional.
2287 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
2288 Patch by Simon Arlott
2290 TF/03 Fix valgrind.h portability to C89 compilers that do not support
2291 variable argument macros. Our copy now differs from upstream.
2297 TF/01 Failure to get a lock on a hints database can have serious
2298 consequences so log it to the panic log.
2300 TF/02 Log LMTP confirmation messages in the same way as SMTP,
2301 controlled using the smtp_confirmation log selector.
2303 TF/03 Include the error message when we fail to unlink a spool file.
2305 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
2306 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
2307 for maintaining out-of-tree patches for some time.
2309 PP/01 Bugzilla 139: Documentation and portability issues.
2310 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
2311 Handle per-OS dynamic-module compilation flags.
2313 PP/02 Let /dev/null have normal permissions.
2314 The 4.73 fixes were a little too stringent and complained about the
2315 permissions on /dev/null. Exempt it from some checks.
2316 Reported by Andreas M. Kirchwitz.
2318 PP/03 Report version information for many libraries, including
2319 Exim version information for dynamically loaded libraries. Created
2320 version.h, now support a version extension string for distributors
2321 who patch heavily. Dynamic module ABI change.
2323 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
2324 privilege escalation vulnerability whereby the Exim run-time user
2325 can cause root to append content of the attacker's choosing to
2328 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
2331 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
2332 If dropping privileges for untrusted macros, we disabled normal logging
2333 on the basis that it would fail; for the Exim run-time user, this is not
2334 the case, and it resulted in successful deliveries going unlogged.
2335 Fixed. Reported by Andreas Metzler.
2341 PP/01 Date: & Message-Id: revert to normally being appended to a message,
2342 only prepend for the Resent-* case. Fixes regression introduced in
2343 Exim 4.70 by NM/22 for Bugzilla 607.
2345 PP/02 Include check_rfc2047_length in configure.default because we're seeing
2346 increasing numbers of administrators be bitten by this.
2348 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
2350 PP/03 Bugzilla 994: added openssl_options main configuration option.
2352 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
2354 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
2356 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
2358 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
2359 without a peer certificate, leading to a segfault because of an
2360 assumption that peers always have certificates. Be a little more
2361 paranoid. Problem reported by Martin Tscholak.
2363 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
2364 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
2365 NB: ClamAV planning to remove STREAM in "middle of 2010".
2366 CL also introduces -bmalware, various -d+acl logging additions and
2367 more caution in buffer sizes.
2369 PP/09 Implemented reverse_ip expansion operator.
2371 PP/10 Bugzilla 937: provide a "debug" ACL control.
2373 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
2375 PP/12 Bugzilla 973: Implement --version.
2377 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2379 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2381 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2383 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2384 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2385 ignore trailing whitespace.
2387 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2389 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2392 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2393 Notification from Dr Andrew Aitchison.
2395 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2396 ExtendedDetectionInfo response format.
2397 Notification from John Horne.
2399 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2402 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2403 XSL and documented dependency on system catalogs, with examples of how
2406 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2409 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2410 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2411 configuration file which is writeable by the Exim user or group.
2413 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2414 of configuration files to cover files specified with the -C option if
2415 they are going to be used with root privileges, not just the default
2418 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2419 option (effectively making it always true).
2421 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2422 files to be used while preserving root privileges.
2424 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2425 that rogue child processes cannot use them.
2427 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2428 run-time user, instead of root.
2430 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2431 Exim run-time user without dropping privileges.
2433 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2434 result string, instead of calling string_vformat() twice with the same
2437 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2438 for other users. Others should always drop root privileges if they use
2439 -C on the command line, even for a whitelisted configure file.
2441 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2443 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2449 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2450 $data_path, and $header_path variables; fixed documentation bugs and
2453 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2454 exipick to access non-standard spools, including the "frozen" queue
2457 NM/01 Bugzilla 965: Support mysql stored procedures.
2458 Patch from Alain Williams
2460 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2462 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2463 Patch from Andreas Metzler
2465 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2466 Patch from Kirill Miazine
2468 NM/05 Bugzilla 671: Added umask to procmail example.
2470 JJ/03 installed exipick 20100323.0, fixing doc bug
2472 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2473 directory. Notification and patch from Dan Rosenberg.
2475 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2477 TK/02 Improve log output when DKIM signing operation fails.
2479 MH/01 Treat the transport option dkim_domain as a colon separated
2480 list, not as a single string, and sign the message with each element,
2481 omitting multiple occurences of the same signer.
2483 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2484 Bugzilla 985, 986. Patch by Simon Arlott
2486 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2487 Patch by Simon Arlott
2489 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2490 MBX locking. Notification from Dan Rosenberg.
2496 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2498 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2500 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2502 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2504 NM/04 Bugzilla 915: Flag broken perl installation during build.
2510 TK/01 Added patch by Johannes Berg that expands the main option
2511 "spamd_address" if it starts with a dollar sign.
2513 TK/02 Write list of recipients to X-Envelope-Sender header when building
2514 the mbox-format spool file for content scanning (suggested by Jakob
2517 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2518 (http://www.dcc-servers.net/) support via dccifd. Activated by
2519 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2521 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2522 by Mark Daniel Reidel <mr@df.eu>.
2524 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2525 When building exim an external PCRE library is now needed -
2526 PCRE is a system library on the majority of modern systems.
2527 See entry on PCRE_LIBS in EDITME file.
2529 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2530 conversation. Added nologin parameter to request.
2531 Patch contributed by Kirill Miazine.
2533 TF/01 Do not log submission mode rewrites if they do not change the address.
2535 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2537 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2538 log files in place. Contributed by Roberto Lima.
2540 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2542 TF/03 Bugzilla 615: When checking the local_parts router precondition
2543 after a local_part_suffix or local_part_prefix option, Exim now
2544 does not use the address's named list lookup cache, since this
2545 contains cached lookups for the whole local part.
2547 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2548 Robert Millan. Documentation is in experimental-spec.txt.
2550 TF/04 Bugzilla 668: Fix parallel build (make -j).
2552 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2554 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2555 Patch provided by Jan Srzednicki.
2557 TF/05 Leading white space used to be stripped from $spam_report which
2558 wrecked the formatting. Now it is preserved.
2560 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2561 that they are available at delivery time.
2563 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2565 TF/08 TLS error reporting now respects the incoming_interface and
2566 incoming_port log selectors.
2568 TF/09 Produce a more useful error message if an SMTP transport's hosts
2569 setting expands to an empty string.
2571 NM/06 Bugzilla 744: EXPN did not work under TLS.
2572 Patch provided by Phil Pennock.
2574 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2575 Patch provided by Richard Godbee.
2577 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2578 acl_smtp_notquit, added index entry.
2580 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2581 Patch provided by Eugene Bujak.
2583 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2584 accept(). Patch provided by Maxim Dounin.
2586 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2587 Patch provided by Phil Pennock.
2589 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2591 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2592 Patch provided by Brad "anomie" Jorsch.
2594 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2595 Patch provided by Dean Brooks.
2597 TK/05 Add native DKIM support (does not depend on external libraries).
2599 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2600 Patch provided by Graeme Fowler.
2602 NM/16 Bugzilla 851: Documentation example syntax fix.
2604 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2606 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2609 NM/19 Bugzilla 745: TLS version reporting.
2610 Patch provided by Phil Pennock.
2612 NM/20 Bugzilla 167: bool: condition support.
2613 Patch provided by Phil Pennock.
2615 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2616 clients. Patch provided by Phil Pennock.
2618 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2619 Patch provided by Brad "anomie" Jorsch.
2621 NM/23 Bugzilla 687: Fix misparses in eximstats.
2622 Patch provided by Heiko Schlittermann.
2624 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2625 Patch provided by Heiko Schlittermann.
2627 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2628 plus update to original patch.
2630 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2632 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2633 Patch provided by David Brownlee.
2635 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2637 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2639 NM/30 Bugzilla 888: TLS documentation bugfixes.
2641 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2643 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2644 Unlike the original bugzilla I have changed all shell scripts in src tree.
2646 NM/33 Bugzilla 898: Transport filter timeout fix.
2647 Patch by Todd Rinaldo.
2649 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2650 Patch by Serge Demonchaux.
2652 NM/35 Bugzilla 39: Base64 decode bug fixes.
2653 Patch by Jakob Hirsch.
2655 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2657 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2659 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2661 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2667 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2668 ALT-N's libdkim that I have put here:
2669 http://duncanthrax.net/exim-experimental/
2671 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2672 were called 'true' and it seems that C99 defines that as a reserved
2673 keyword to be used with 'bool' variable types. That means you could
2674 not include C99-style headers which use bools without triggering
2675 build errors in sieve.c.
2677 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2678 as mailq or other aliases. Changed the --help handling significantly
2679 to do whats expected. exim_usage() emits usage/help information.
2681 SC/01 Added the -bylocaldomain option to eximstats.
2683 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2685 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2687 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2693 PH/01 Another patch from the Sieve maintainer.
2695 PH/02 When an IPv6 address is converted to a string for single-key lookup
2696 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2697 dots are used instead of colons so that keys in lsearch files need not
2698 contain colons. This was done some time before quoting was made available
2699 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2700 (notated using the quote facility) so as to distinguish them from IPv4
2701 keys. This meant that lookups for IP addresses in host lists did not work
2702 for iplsearch lookups.
2704 This has been fixed by arranging for IPv6 addresses to be expressed with
2705 colons if the lookup type is iplsearch. This is not incompatible, because
2706 previously such lookups could never work.
2708 The situation is now rather anomalous, since one *can* have colons in
2709 ordinary lsearch keys. However, making the change in all cases is
2710 incompatible and would probably break a number of configurations.
2712 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2715 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2716 conversion specification without a maximum field width, thereby enabling
2717 a rogue spamd server to cause a buffer overflow. While nobody in their
2718 right mind would setup Exim to query an untrusted spamd server, an
2719 attacker that gains access to a server running spamd could potentially
2720 exploit this vulnerability to run arbitrary code as the Exim user.
2722 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2723 $primary_hostname instead of what libspf2 thinks the hosts name is.
2725 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2726 a directory entry by the name of the lookup key. Previously, if a
2727 symlink pointed to a non-existing file or a file in a directory that
2728 Exim lacked permissions to read, a lookup for a key matching that
2729 symlink would fail. Now it is enough that a matching directory entry
2730 exists, symlink or not. (Bugzilla 503.)
2732 PH/03 The body_linecount and body_zerocount variables are now exported in the
2735 PH/04 Added the $dnslist_matched variable.
2737 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2738 This means they are set thereafter only if the connection becomes
2741 PH/06 Added the client_condition to authenticators so that some can be skipped
2742 by clients under certain conditions.
2744 PH/07 The error message for a badly-placed control=no_multiline_responses left
2745 "_responses" off the end of the name.
2747 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2749 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2750 (without spaces) instead of just copying the configuration text.
2752 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2754 PH/11 Added $max_received_linelength.
2756 PH/12 Added +ignore_defer and +include_defer to host lists.
2758 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2759 way in which PCRE > 7.0 is built.
2761 PH/14 Implemented queue_only_load_latch.
2763 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2764 MAIL command. The effect was to mangle the value on 64-bit systems.
2766 PH/16 Another patch from the Sieve maintainer.
2768 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2770 PH/18 If a system quota error occurred while trying to create the file for
2771 a maildir delivery, the message "Mailbox is full" was not appended to the
2772 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2773 only to a quota excession during the actual writing of the file.
2775 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2776 characters?) which causes problems in log lines. The DN values are now
2777 passed through string_printing() before being added to log lines.
2779 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2780 and InterBase are left for another time.)
2782 PH/21 Added message_body_newlines option.
2784 PH/22 Guard against possible overflow in moan_check_errorcopy().
2786 PH/23 POSIX allows open() to be a macro; guard against that.
2788 PH/24 If the recipient of an error message contained an @ in the local part
2789 (suitably quoted, of course), incorrect values were put in $domain and
2790 $local_part during the evaluation of errors_copy.
2796 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2797 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2800 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2801 issue a MAIL command.
2803 PH/02 In an ACL statement such as
2805 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2807 if a client was not listed at all, or was listed with a value other than
2808 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2809 the condition was not true (as it should be), so access was not denied.
2810 The bug was that the ! inversion was incorrectly passed on to the second
2811 item. This has been fixed.
2813 PH/03 Added additional dnslists conditions == and =& which are different from
2814 = and & when the dns lookup returns more than one IP address.
2816 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2817 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2819 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2820 FSYNC, which compiles an option called disable_fsync that allows for
2821 bypassing fsync(). The documentation is heavily laced with warnings.
2823 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2825 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2826 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2827 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2828 including adding "make clean"; (3) Added -fPIC when compiling the test
2829 dynamically loaded module, to get rid of a warning.
2831 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2832 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2833 = 0s. The bug is otherwise harmless.
2835 PH/07 There was a bug in the dovecot authenticator such that the value of
2836 $auth1 could be overwritten, and so not correctly preserved, after a
2837 successful authentication. This usually meant that the value preserved by
2838 the server_setid option was incorrect.
2840 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2842 PH/09 Installed PCRE release 7.0.
2844 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2845 run for batched SMTP input. It is now run at the start of every message
2846 in the batch. While fixing this I discovered that the process information
2847 (output by running exiwhat) was not always getting set for -bs and -bS
2848 input. This is fixed, and it now also says "batched" for BSMTP.
2850 PH/11 Added control=no_pipelining.
2852 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2853 patch, slightly modified), and move the expansion of helo_data till after
2854 the connection is made in the smtp transport (so it can use these
2857 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2859 PH/14 Added log_selector = +pid.
2861 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2863 PH/16 Add ${if forany and ${if forall.
2865 PH/17 Added dsn_from option to vary the From: line in DSNs.
2867 PH/18 Flush SMTP output before performing a callout, unless control =
2868 no_callout_flush is set.
2870 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2871 was true (the default) a successful delivery failed to delete the retry
2872 item, thus causing premature timeout of the address. The bug is now
2875 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2877 PH/21 Long custom messages for fakedefer and fakereject are now split up
2878 into multiline responses in the same way that messages for "deny" and
2879 other ACL rejections are.
2881 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2882 with slight modification.
2884 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2885 draft, changing the syntax and factoring some duplicate code".
2887 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2888 for deliveries of the second and subsequent messages over the same SMTP
2891 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2892 ${reduce, with only minor "tidies".
2894 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2896 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2897 expansion side effects.
2899 PH/27 When a message times out after an over-quota error from an Exim-imposed
2900 quota, the bounce message says "mailbox is full". This message was not
2901 being given when it was a system quota that was exceeded. It now should
2904 MH/03 Made $recipients available in local_scan(). local_scan() already has
2905 better access to the recipient list through recipients_list[], but
2906 $recipients can be useful in postmaster-provided expansion strings.
2908 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2909 in the case of a MAIL command with additional options following the
2910 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2911 were accidentally chopped off.
2913 PH/29 SMTP synchronization checks are implemented when a command is read -
2914 there is a check that no more input is waiting when there shouldn't be
2915 any. However, for some commands, a delay in an ACL can mean that it is
2916 some time before the response is written. In this time, more input might
2917 arrive, invalidly. So now there are extra checks after an ACL has run for
2918 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2919 pipelining has not been advertised.
2921 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2923 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2924 returned in the SMTP response when smtp_return_error_details was false.
2925 This has been fixed.
2927 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2928 instead of the C I/O that was originally supplied, because problems were
2929 reported on Solaris.
2931 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2932 Exim which did not show up earlier: it was assuming that a call to
2933 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2934 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2935 was a macro that became an assignment, so it seemed to work. This has
2936 changed to a proper function call with a void return, hence the compile
2937 error. Exim's code has been fixed.
2939 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2942 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2944 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2946 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2947 because it turns out to be too big or there is a timeout) there is no
2948 "Completed" line in the log. When some messages of this type were
2949 selected by exigrep, they were listed as "not completed". Others were
2950 picked up by some special patterns. I have improved the selection
2951 criteria to be more general.
2953 PH/38 The host_find_failed option in the manualroute router can now be set
2954 to "ignore", to completely ignore a host whose IP address cannot be
2955 found. If all hosts are ignored, the behaviour is controlled by the new
2956 host_all_ignored option.
2958 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2959 homing or because of multiple MX records with /mx) generated more than
2960 one IP address, and the following item turned out to be the local host,
2961 all the secondary addresses of the first item were incorrectly removed
2962 from the list, along with the local host and any following hosts (which
2963 is what is supposed to happen).
2965 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2966 whoever called Exim into the -H file. In the case of the daemon it was
2967 behaving confusingly. When first started, it used values for whoever
2968 started the daemon, but after a SIGHUP it used the Exim user (because it
2969 calls itself on a restart). I have changed the code so that it now always
2972 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2973 message are rejected with the same error (e.g. no authentication or bad
2974 sender address), and a DATA command is nevertheless sent (as can happen
2975 with PIPELINING or a stupid MUA), the error message that was given to the
2976 RCPT commands is included in the rejection of the DATA command. This is
2977 intended to be helpful for MUAs that show only the final error to their
2980 PH/42 Another patch from the Sieve maintainer.
2982 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2983 Eximstats - Fixed some broken HTML links and added missing column headers
2985 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2986 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2988 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2990 PH/43 Yet another patch from the Sieve maintainer.
2992 PH/44 I found a way to check for a TCP/IP connection going away before sending
2993 the response to the final '.' that terminates a message, but only in the
2994 case where the client has not sent further data following the '.'
2995 (unfortunately, this is allowed). However, in many cases there won't be
2996 any further data because there won't be any more messages to send. A call
2997 to select() can be used: if it shows that the input is "ready", there is
2998 either input waiting, or the socket has been closed. An attempt to read
2999 the next input character can distinguish the two cases. Previously, Exim
3000 would have sent an OK response which the client would never have see.
3001 This could lead to message repetition. This fix should cure that, at
3002 least in a lot of common cases.
3004 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
3005 advertised in response to EHLO.
3011 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
3012 fixed by 4.65/MH/01 (is this a record?) are fixed:
3014 (i) An empty string was always treated as zero by the numeric comparison
3015 operators. This behaviour has been restored.
3017 (ii) It is documented that the numeric comparison operators always treat
3018 their arguments as decimal numbers. This was broken in that numbers
3019 starting with 0 were being interpreted as octal.
3021 While fixing these problems I realized that there was another issue that
3022 hadn't been noticed. Values of message_size_limit (both the global option
3023 and the transport option) were treated as octal if they started with 0.
3024 The documentation was vague. These values are now always treated as
3025 decimal, and I will make that clear in the documentation.
3031 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
3032 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
3035 MH/01 Don't check that the operands of numeric comparison operators are
3036 integers when their expansion is in "skipping" mode (fixes bug
3037 introduced by 4.64-PH/07).
3039 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
3040 child addresses, Exim now panics and dies. Previously, because the count
3041 is held in a short int, deliveries were likely to be lost. As such a
3042 large number of recipients for a single message is ridiculous
3043 (performance will be very, very poor), I have chosen to impose a limit
3044 rather than extend the field.
3050 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
3051 leftover -K file (the existence of which was triggered by #402).
3052 While we were at it, introduced process PID as part of the -K
3053 filename. This should rule out race conditions when creating
3056 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
3057 processing considerably. Previous code took too long for large mails,
3058 triggering a timeout which in turn triggers #401.
3060 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
3061 in the DK code in transports.c. sendfile() is not really portable,
3062 hence the _LINUX specificness.
3064 TF/01 In the add_headers option to the mail command in an Exim filter,
3065 there was a bug that Exim would claim a syntax error in any
3066 header after the first one which had an odd number of characters
3069 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
3070 callout verification, Exim cached a "reject" for the entire domain. This
3071 is correct for most verifications, but it is not correct for a recipient
3072 verification with use_sender or use_postmaster set, because in that case
3073 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
3074 case of MAIL FROM:<> rejection from other early rejections (e.g.
3075 rejection of HELO). When verifying a recipient using a non-null MAIL
3076 address, the cache is ignored if it shows MAIL FROM:<> rejection.
3077 Whatever the result of the callout, the value of the domain cache is
3078 left unchanged (for any other kind of callout, getting as far as trying
3079 RCPT means that the domain itself is ok).
3081 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
3084 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
3085 manifest itself as EPIPE rather than ECONNECT. When tidying away a
3086 session, the daemon ignores ECONNECT errors and logs others; it now
3087 ignores EPIPE as well.
3089 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
3090 (quoted-printable decoding).
3092 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
3093 later the small subsequent patch to fix an introduced bug.
3095 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
3097 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
3099 PH/08 An error is now given if message_size_limit is specified negative.
3101 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
3102 to be given (somewhat) arbitrary names.
3104 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
3107 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
3108 miscellaneous code fixes
3110 PH/10 Added the log_reject_target ACL modifier to specify where to log
3113 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
3114 hostname. This is wrong, because it relates to the incoming message (and
3115 probably the interface on which it is arriving) and not to the outgoing
3116 callout (which could be using a different interface). This has been
3117 changed to use the value of the helo_data option from the smtp transport
3118 instead - this is what is used when a message is actually being sent. If
3119 there is no remote transport (possible with a router that sets up host
3120 addresses), $smtp_active_hostname is used.
3122 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
3123 tweaks were necessary in order to get it to work (see also 21 below):
3124 (a) The code assumed that strncpy() returns a negative number on buffer
3125 overflow, which isn't the case. Replaced with Exim's string_format()
3127 (b) There were several signed/unsigned issues. I just did the minimum
3128 hacking in of casts. There is scope for a larger refactoring.
3129 (c) The code used strcasecmp() which is not a standard C function.
3130 Replaced with Exim's strcmpic() function.
3131 (d) The code set only $1; it now sets $auth1 as well.
3132 (e) A simple test gave the error "authentication client didn't specify
3133 service in request". It would seem that Dovecot has changed its
3134 interface. Fortunately there's a specification; I followed it and
3135 changed what the client sends and it appears to be working now.
3137 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
3140 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
3141 address A is aliased to B and C, where B exists and C does not. Without
3142 -v the output is "A verified" because verification stops after a
3143 successful redirection if more than one address is generated. However,
3144 with -v the child addresses are also verified. Exim was outputting "A
3145 failed to verify" and then showing the successful verification for C,
3146 with its parentage. It now outputs "B failed to verify", showing B's
3147 parentage before showing the successful verification of C.
3149 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
3150 look up a TXT record in a specific list after matching in a combined
3153 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
3154 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
3155 they consult the DNS. I had assumed they would set it the way they
3156 wanted; and indeed my experiments on Linux seem to show that in some
3157 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
3158 To be on the safe side, however, I have now made the interface to
3159 host_find_byname() similar to host_find_bydns(), with an argument
3160 containing the DNS resolver options. The host_find_byname() function now
3161 sets these options at its start, just as host_find_bydns() does. The smtp
3162 transport options dns_qualify_single and dns_search_parents are passed to
3163 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
3164 of host_find_byname() use the default settings of RES_DEFNAMES
3165 (qualify_single) but not RES_DNSRCH (search_parents).
3167 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
3168 spool_read_header() do less string testing, by means of a preliminary
3169 switch on the second character of optional "-foo" lines. (This is
3170 overdue, caused by the large number of possibilities that now exist.
3171 Originally there were few.) While I was there, I also converted the
3172 str(n)cmp tests so they don't re-test the leading "-" and the first
3173 character, in the hope this might squeeze out yet more improvement.
3175 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
3176 flag allowing group syntax was set by the header_syntax check but not
3177 turned off, possible causing trouble later; (2) The flag was not being
3178 set at all for the header_verify test, causing "group"-style headers to
3179 be rejected. I have now set it in this case, and also caused header_
3180 verify to ignore an empty address taken from a group. While doing this, I
3181 came across some other cases where the code for allowing group syntax
3182 while scanning a header line wasn't quite right (mostly, not resetting
3183 the flag correctly in the right place). These bugs could have caused
3184 trouble for malformed header lines. I hope it is now all correct.
3186 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
3187 with the "reply" argument non-NULL. The code, however (which originally
3188 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
3189 but it didn't always do it. This confused somebody who was copying the
3190 code for some other use. I have removed all the tests.
3192 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
3193 feature that was used to support insecure browsers during the U.S. crypto
3194 embargo. It requires special client support, and Exim is probably the
3195 only MTA that supported it -- and would never use it because real RSA is
3196 always available. This code has been removed, because it had the bad
3197 effect of slowing Exim down by computing (never used) parameters for the
3198 RSA_EXPORT functionality.
3200 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
3201 authenticator to fail if there's a tab character in the incoming data
3202 (there should never be unless someone is messing about, as it's supposed
3203 to be base64-encoded). Also added, on Timo's advice, the "secured" option
3204 if the connection is using TLS or if the remote IP is the same as the
3205 local IP, and the "valid-client-cert option" if a client certificate has
3208 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
3209 authenticators. This can be used for authorization after authentication
3210 succeeds. (In the case of plaintext, it servers for both authentication
3213 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
3214 if any retry times were supplied.
3216 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
3217 connection, where there is no client IP address to check. In this
3218 situation, the verify now always succeeds.
3220 PH/25 Applied John Jetmore's -Mset patch.
3222 PH/26 Added -bem to be like -Mset, but loading a message from a file.
3224 PH/27 In a string expansion for a processed (not raw) header when multiple
3225 headers of the same name were present, leading whitespace was being
3226 removed from all of them, but trailing whitespace was being removed only
3227 from the last one. Now trailing whitespace is removed from each header
3228 before concatenation. Completely empty headers in a concatenation (as
3229 before) are ignored.
3231 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
3232 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
3234 PH/29 [Removed. This was a change that I later backed out, and forgot to
3235 correct the ChangeLog entry (that I had efficiently created) before
3236 committing the later change.]
3238 PH/30 Exim was sometimes attempting to deliver messages that had suffered
3239 address errors (4xx response to RCPT) over the same connection as other
3240 messages routed to the same hosts. Such deliveries are always "forced",
3241 so retry times are not inspected. This resulted in far too many retries
3242 for the affected addresses. The effect occurred only when there were more
3243 hosts than the hosts_max_try setting in the smtp transport when it had
3244 the 4xx errors. Those hosts that it had tried were not added to the list
3245 of hosts for which the message was waiting, so if all were tried, there
3246 was no problem. Two fixes have been applied:
3248 (i) If there are any address or message errors in an SMTP delivery, none
3249 of the hosts (tried or untried) are now added to the list of hosts
3250 for which the message is waiting, so the message should not be a
3251 candidate for sending over the same connection that was used for a
3252 successful delivery of some other message. This seems entirely
3253 reasonable: after all the message is NOT "waiting for some host".
3254 This is so "obvious" that I'm not sure why it wasn't done
3255 previously. Hope I haven't missed anything, but it can't do any
3256 harm, as the worst effect is to miss an optimization.
3258 (ii) If, despite (i), such a delivery is accidentally attempted, the
3259 routing retry time is respected, so at least it doesn't keep
3260 hammering the server.
3262 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
3263 in ${readsocket because some servers need this prod.
3265 PH/32 Added some extra debug output when updating a wait-xxx database.
3267 PH/33 The hint "could be header name not terminated by colon", which has been
3268 given for certain expansion errors for a long time, was not being given
3269 for the ${if def:h_colon_omitted{... case.
3271 PH/34 The spec says: "With one important exception, whenever a domain list is
3272 being scanned, $domain contains the subject domain." There was at least
3273 one case where this was not true.
3275 PH/35 The error "getsockname() failed: connection reset by peer" was being
3276 written to the panic log as well as the main log, but it isn't really
3277 panic-worthy as it just means the connection died rather early on. I have
3278 removed the panic log writing for the ECONNRESET error when getsockname()
3281 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
3282 runs only) independently of the message's sender address. This meant
3283 that, if the 4xx error was in fact related to the sender, a different
3284 message to the same recipient with a different sender could confuse
3285 things. In particular, this can happen when sending to a greylisting
3286 server, but other circumstances could also provoke similar problems.
3287 I have changed the default so that the retry time for these errors is now
3288 based a combination of the sender and recipient addresses. This change
3289 can be overridden by setting address_retry_include_sender=false in the
3292 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
3293 remote server are returned as part of bounce messages. This was not
3294 happening for LMTP over a pipe (the lmtp transport), but now it is the
3295 same for both kinds of LMTP.
3297 PH/38 Despite being documented as not happening, Exim was rewriting addresses
3298 in header lines that were in fact CNAMEs. This is no longer the case.
3300 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
3301 and queue runs started by the daemon processed all messages. This has
3302 been fixed so that -R and -S can now usefully be given with -q<time>.
3304 PH/40 Import PCRE release 6.7 (fixes some bugs).
3306 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
3308 PH/42 Give an error if -q is specified more than once.
3310 PH/43 Renamed the variables $interface_address and $interface_port as
3311 $received_ip_address and $received_port, to make it clear that these
3312 values apply to message reception, and not to the outgoing interface when
3313 a message is delivered. (The old names remain recognized, of course.)
3315 PH/44 There was no timeout on the connect() call when using a Unix domain
3316 socket in the ${readsocket expansion. There now is.
3318 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
3319 be meaningful with "accept".
3321 SC/01 Eximstats V1.43
3322 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
3324 SC/02 Eximstats V1.44
3325 Use a glob alias rather than an array ref in the generated
3326 parser. This improves both readability and performance.
3328 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
3329 Collect SpamAssassin and rejection statistics.
3330 Don't display local sender or destination tables unless
3331 there is data to show.
3332 Added average volumes into the top table text output.
3334 SC/04 Eximstats V1.46
3335 Collect data on the number of addresses (recipients)
3336 as well as the number of messages.
3338 SC/05 Eximstats V1.47
3339 Added 'Message too big' to the list of mail rejection
3340 reasons (thanks to Marco Gaiarin).
3342 SC/06 Eximstats V1.48
3343 Mainlog lines which have GMT offsets and are too short to
3344 have a flag are now skipped.
3346 SC/07 Eximstats V1.49 (Alain Williams)
3347 Added the -emptyok flag.
3349 SC/08 Eximstats V1.50
3350 Fixes for obtaining the IP address from reject messages.
3352 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
3353 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
3354 whitespace changes from 4.64-PH/27
3356 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
3359 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
3360 are found, allow negative numbers in numeric criteria)
3362 JJ/06 exipick.20061117.2, added new $message_body_missing variable
3364 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
3365 to match changes made in 4.64-PH/43
3367 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
3369 PH/47 Put in an explicit test for a DNS lookup of an address record where the
3370 "domain" is actually an IP address, and force a failure. This locks out
3371 those revolvers/nameservers that support "A-for-A" lookups, in
3372 contravention of the specifications.
3374 PH/48 When a host name was looked up from an IP address, and the subsequent
3375 forward lookup of the name timed out, the host name was left in
3376 $sender_host_name, contrary to the specification.
3378 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3379 restricted to single-key lookups, Exim was not diagnosing an error if
3380 * or *@ was used with a query-style lookup.
3382 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3384 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3385 long ago, but noone interested enough thought of it. Let's just say that
3386 the "1.1" means that there are some new functions that weren't there at
3387 some point in the past.
3389 PH/51 Error processing for expansion failure of helo_data from an smtp
3390 transport during callout processing was broken.
3392 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3393 tested/used via the -bh/-bhc/-bs options.
3395 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3396 bug, fixed in subsequent PCRE releases).
3398 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3399 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3401 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3407 SC/01 Use a glob alias rather than an array ref in eximstats generated
3408 parser. This improves both readability and performance.
3410 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3411 Don't display local sender or destination tables in eximstats unless
3412 there is data to show.
3413 Added average volumes into the eximstats top table text output.
3415 SC/03 Collect data on the number of addresses (recipients) as well
3416 as the number of messages in eximstats.
3418 TF/01 Correct an error in the documentation for the redirect router. Exim
3419 does (usually) call initgroups() when daemonizing.
3421 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3422 with consistent privilege compared to when running as a daemon.
3424 TF/03 Note in the spec that $authenticated_id is not set for local
3425 submissions from trusted users.
3427 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3428 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3430 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3431 by adding some example configuration directives to the default
3432 configuration file. A little bit of work is required to uncomment the
3433 directives and define how usernames and passwords are checked, but
3434 there is now a framework to start from.
3436 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3437 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3438 without this. I don't know how relevant this is to other LDAP libraries.
3440 PH/02 Add the verb name to the "unknown ACL verb" error.
3442 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3444 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3446 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3447 directory not expanded when it should be if an expanded home directory
3448 was set for the address (which is overridden by the transport).
3450 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3453 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3454 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3455 because it is too late at that time, and has no effect.
3457 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3458 security issue with \' (bugzilla #107). I could not use the
3459 PQescapeStringConn() function, because it needs a PGconn value as one of
3462 PH/08 When testing addresses using -bt, indicate those final addresses that
3463 are duplicates that would not cause an additional delivery. At least one
3464 person was confused, thinking that -bt output corresponded to deliveries.
3465 (Suppressing duplicates isn't a good idea as you lose the information
3466 about possibly different redirections that led to the duplicates.)
3468 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3469 systems where poll() doesn't work, in particular OS X.
3471 PH/10 Added more information to debugging output for retry time not reached.
3473 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3474 operations in malware.c.
3476 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3479 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3480 syslog_duplication set false, log lines that would normally be written
3481 both the the main log and to the reject log were not written to syslog at
3484 PH/14 In the default configuration, change the use of "message" in ACL warn
3485 statements to "add_header".
3487 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3488 not followed by a command (e.g. "seen endif").
3490 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3491 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3494 PH/17 Added extra conditions to the default value of delay_warning_condition
3498 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3499 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3500 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3503 The Auto-Submitted: and various List- headers are standardised, whereas I
3504 don't think Precedence: ever was.
3506 PH/18 Refactored debugging code in route_finduser() to show more information,
3507 in particular, the error code if getpwnam() issues one.
3509 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3510 This is apparently needed in addition to the PH/07 change above to avoid
3511 any possible encoding problems.
3513 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3514 but not after initializing Perl.
3516 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3517 output them only if debugging. By default they are written stderr,
3518 apparently, which is not desirable.
3520 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3523 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3526 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3528 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3529 authenticated or an ident call has been made. Suppress the default
3530 values for $authenticated_id and $authenticated_sender (but permit -oMai
3531 and -oMas) when testing with -bh.
3533 PH/24 Re-jigged the order of the tests in the default configuration so that the
3534 tests for valid domains and recipients precede the DNS black list and CSA
3535 tests, on the grounds that those ones are more expensive.
3537 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3538 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3539 This bug exists in every version of Exim that I still have, right back to
3542 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3543 However, an attempt to turn on case-sensitivity in a regex key by
3544 including (?-i) didn't work because the subject string was already
3545 lowercased, and the effects were non-intuitive. It turns out that a
3546 one-line patch can be used to allow (?-i) to work as expected.
3552 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3553 other effects) broke the use of negated acl sub-conditions.
3555 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3558 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3559 "Deny" causes Exim to reject the incoming connection with a 554 error.
3560 Unfortunately, if there is a major crisis, such as a disk failure,
3561 tcp-wrappers gives "deny", whereas what one would like would be some
3562 kind of temporary error. A kludge has been added to help with this.
3563 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3564 554 error is used if errno is still zero or contains ENOENT (which occurs
3565 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3568 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3570 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3571 errors. Otherwise a message that provokes a temporary error (when other
3572 messages do not) can cause a whole host to time out.
3574 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3575 addresses were routed directly to files or pipes from a redirect router.
3576 File deliveries just didn't batch; pipe deliveries might have suffered
3579 PH/06 A failure to get a lock for a hints database would erroneously always say
3580 "Failed to get write lock", even when it was really a read lock.
3582 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3583 of 0600. This has been changed to use the value of the lockfile_mode
3584 option (which defaults to 0600).
3586 PH/08 Applied small patch from the Sieve maintainer.
3588 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3589 folder from quota calculations, a direct delivery into this folder messed
3590 up the contents of the maildirsize file. This was because the regex was
3591 used only to exclude .Trash (or whatever) when the size of the mailbox
3592 was calculated. There was no check that a delivery was happening into an
3593 excluded directory. This bug has been fixed by ignoring all quota
3594 processing for deliveries into excluded directories.
3596 PH/10 Added the maildirfolder_create_regex option to appendfile.
3602 PH/01 The code for finding all the local interface addresses on a FreeBSD
3603 system running IPv6 was broken. This may well have applied to all BSD
3604 systems, as well as to others that have similar system calls. The broken
3605 code found IPv4 interfaces correctly, but gave incorrect values for the
3606 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3607 that it would not match correctly against @[] and not recognize the IPv6
3610 PH/02 The ipliteral router was not recognizing addresses of the form user@
3611 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3613 PH/03 Added disable_ipv6.
3615 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3616 decoded form, because it is most often used to construct To: headers
3617 lines in autoreplies, and the decoded form may well be syntactically
3618 invalid. However, $reply_address has leading white space removed, and all
3619 newlines turned into spaces so that the autoreply transport does not
3622 PH/05 If group was specified without a user on a router, and no group or user
3623 was specified on a transport, the group from the router was ignored.
3625 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3626 for visible compile-time settings that can be used to change these
3627 numbers, for those that want even more. Backwards compatibility with old
3628 spool files has been maintained. However, going back to a previous Exim
3629 release will lost any variables that are in spool files.
3631 PH/07 Two small changes when running in the test harness: increase delay when
3632 passing a TCP/IP connection to a new process, in case the original
3633 process has to generate a bounce, and remove special handling of
3634 127.0.0.2 (sic), which is no longer necessary.
3636 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3637 be the same on different OS.
3639 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3642 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3643 whether --show-vars was specified or not
3645 JJ/02 exipick: Added support for new ACL variable spool format introduced
3648 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3649 syntactically invalid From: or Reply-to: line, and a filter used this to
3650 generate an autoreply, and therefore failed to obtain an address for the
3651 autoreply, Exim could try to deliver to a non-existent relative file
3652 name, causing unrelated and misleading errors. What now happens is that
3653 it logs this as a hard delivery error, but does not attempt to create a
3656 PH/11 The exinext utility has a -C option for testing purposes, but although
3657 the given file was scanned by exinext itself; it wasn't being passed on
3658 when Exim was called.
3660 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3661 an end-of-file indication when reading a command response.
3663 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3664 compiled. In many other places in Exim, IPv6 addresses are always
3665 recognized, so I have changed this. It also means that IPv4 domain
3666 literals of the form [IPV4:n.n.n.n] are now always recognized.
3668 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3669 used if the router is not running as root, for example, when verifying at
3670 ACL time, or when using -bh. The debugging output from this situation was
3671 non-existent - all you got was a failure to exec. I have made two
3674 (a) Failures to set uid/gid, the current directory, or a process leader
3675 in a subprocess such as that created by queryprogram now generate
3676 suitable debugging output when -d is set.
3678 (b) The queryprogram router detects when it is not running as root,
3679 outputs suitable debugging information if -d is set, and then runs
3680 the subprocess without attempting to change uid/gid.
3682 PH/15 Minor change to Makefile for building test_host (undocumented testing
3685 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3686 additional section of a DNS packet that returns MX or SRV records.
3687 Instead, it always explicitly searches for A/AAAA records. This avoids
3688 major problems that occur when a DNS server includes only records of one
3689 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3690 fixed another bug: if SRV records were looked up and the corresponding
3691 address records were *not* found in the additional section, the port
3692 values from the SRV records were lost.
3694 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3695 using the correct key (the original address) when searching the retry
3696 rules in order to find which one to use for generating the retry hint.
3698 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3699 adding the default one. Similarly, if it contains a Reply-To: header, the
3700 errors_reply_to option, if set, is not used.
3702 PH/19 When calculating a retry time, Exim used to measure the "time since
3703 failure" by looking at the "first failed" field in the retry record. Now
3704 it does not use this if it is later than than the arrival time of the
3705 message. Instead it uses the arrival time. This makes for better
3706 behaviour in cases where some deliveries succeed, thus re-setting the
3707 "first failed" field. An example is a quota failure for a huge message
3708 when small messages continue to be delivered. Without this change, the
3709 "time since failure" will always be short, possible causing more frequent
3710 delivery attempts for the huge message than are intended.
3711 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3713 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3714 $1, $2, $3) because the numerical variables can be reset during some
3715 expansion items (e.g. "match"), thereby losing the authentication data.
3717 PH/21 Make -bV show the size of off_t variables so that the test suite can
3718 decide whether to run tests for quotas > 2G.
3720 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3721 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3722 filecount value is greater than 2G or if a quota value is greater than 2G
3723 on a system where the size of off_t is not greater than 4, a panic error
3726 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3727 never match. The debug and -bh output now contains an explicit error
3728 message indicating a malformed IPv4 address or mask.
3730 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3731 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3732 PH/23 above applies.
3734 PH/25 Do not write to syslog when running in the test harness. The only
3735 occasion when this arises is a failure to open the main or panic logs
3736 (for which there is an explicit test).
3738 PH/26 Added the /no_tell option to "control=freeze".
3740 PH/27 If a host name lookup failed very early in a connection, for example, if
3741 the IP address matched host_lookup and the reverse lookup yielded a name
3742 that did not have a forward lookup, an error message of the form "no IP
3743 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3744 could be logged. Now it outputs the IP address instead of "NULL".
3746 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3747 allows the sender and the authenticated sender to be set when
3748 submitting a message from within Exim. Since child_open_exim() is
3749 documented for local_scan(), the new function should be too.
3751 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3752 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3753 results in an empty string is now treated as unset.
3755 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3757 PH/31 Added sender_verify_fail logging option.
3759 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3760 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3761 tidied the source and removed it altogether.
3763 PH/33 When a queue run was abandoned because the load average was too high, a
3764 log line was always written; now it is written only if the queue_run log
3765 selector is set. In addition, the log line for abandonment now contains
3766 information about the queue run such as the pid. This is always present
3767 in "start" and "stop" lines but was omitted from the "abandon" line.
3769 PH/34 Omit spaces between a header name and the colon in the error message that
3770 is given when verify = headers_syntax fails (if there are lots of them,
3771 the message gets confusing).
3773 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3774 names, as there are now some PTR records that contain slashes. This check
3775 is only to protect against broken name servers that fall over on strange
3776 characters, so the fact that it applies to all lookups doesn't matter.
3778 PH/36 Now that the new test suite is complete, we can remove some of the
3779 special code in Exim that was needed for the old test suite. For example,
3780 sorting DNS records because real resolvers return them in an arbitrary
3781 order. The new test suite's fake resolver always returns records in the
3784 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3785 bounces) except when queue_only is set, to avoid logging races between
3786 the different processes.
3788 PH/38 Panic-die if .include specifies a non-absolute path.
3790 PH/39 A tweak to the "H" retry rule from its user.
3792 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3793 a label. They prevented compilation on older perls.
3795 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3796 a warning to be raised on newish perls.
3798 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3799 on queue. Changes to match documented behaviour of showing count of
3800 messages matching specified criteria.
3802 PH/40 Changed the default ident timeout from 30s to 5s.
3804 PH/41 Added support for the use of login_cap features, on those BSD systems
3805 that have them, for controlling the resources used by pipe deliveries.
3807 PH/42 The content-scanning code uses fopen() to create files in which to put
3808 message data. Previously it was not paying any attention to the mode of
3809 the files. Exim runs with umask(0) because the rest of the code creates
3810 files with open(), and sets the required mode explicitly. Thus, these
3811 files were ending up world-writeable. This was not a big issue, because,
3812 being within the spool directory, they were not world-accessible. I have
3813 created a function called modefopen, which takes an additional mode
3814 argument. It sets umask(777), creates the file, chmods it to the required
3815 mode, then resets the umask. All the relevant calls to fopen() in the
3816 content scanning code have been changed to use this function.
3818 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3819 to 24 hours. This avoids potential overflow problems when processing G
3820 and H retry rules. I suspect nobody ever tinkers with this value.
3822 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3824 PH/45 When the plaintext authenticator is running as a client, the server's
3825 challenges are checked to ensure they are valid base64 strings. By
3826 default, the authentication attempt is cancelled if an invalid string is
3827 received. Setting client_ignore_invalid_base64 true ignores these errors.
3828 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3829 they are received. Thus, the responses can be made to depend on the
3830 challenges. If an invalid string is ignored, an empty string is placed in
3833 PH/46 Messages that are created by the autoreply transport now contains a
3834 References: header, in accordance with RFCs 2822 and 3834.
3836 PH/47 Added authenticated_sender_force to the smtp transport.
3838 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3840 PH/49 Installed latest patch from the Sieve maintainer.
3842 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3843 was also set, the appendfile transport was unnecessarily scanning a
3844 directory of message files (e.g. for maildir delivery) to find the count
3845 of files (along with the size), even though it did not need this
3846 information. It now does the scan only if it needs to find either the
3847 size of the count of files.
3849 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3851 PH/52 Two bugs concerned with error handling when the smtp transport is
3854 (i) Exim was not creating retry information for temporary errors given
3855 for individual recipients after the DATA command when the smtp transport
3856 was used in LMTP mode. This meant that they could be retried too
3857 frequently, and not timed out correctly.
3859 (ii) Exim was setting the flag that allows error details to be returned
3860 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3861 recipients that were returned after the DATA command.
3863 PH/53 This is related to PH/52, but is more general: for any failing address,
3864 when detailed error information was permitted to be returned to the
3865 sender, but the error was temporary, then after the final timeout, only
3866 "retry timeout exceeded" was returned. Now it returns the full error as
3867 well as "retry timeout exceeded".
3869 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3870 do this, and (what is worse) MTAs that accept it.
3872 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3873 will now be deprecated.
3875 PH/56 New os.c-cygwin from the Cygwin maintainer.
3877 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3878 formats (previously only available in exim formats via -bpr, -bpru,
3879 and -bpra. Now also available in native and exiqgrep formats)
3881 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3882 with very large, slow to parse queues
3884 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3886 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3888 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3889 responses to authentication challenges, though it was showing the
3890 challenges; (ii) I've removed the CR characters from the debug output for
3893 PH/58 Allow for the insertion of a newline as well as a space when a string
3894 is turned into more than one encoded-word during RFC 2047 encoding. The
3895 Sieve code now uses this.
3897 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3898 data_4xx, lost_connection, tls_required.
3900 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3901 message was being sent as an SMTP response.
3903 PH/61 Add -l and -k options to exicyclog.
3905 PH/62 When verifying, if an address was redirected to one new address, so that
3906 verification continued, and the new address failed or deferred after
3907 having set something in $address_data, the value of $address_data was not
3908 passed back to the ACL. This was different to the case when no
3909 redirection occurred. The value is now passed back in both cases.
3911 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3912 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3913 use login_cap.h, so on its own it isn't the distinguishing feature. The
3914 new name refers directly to the setclassresources() function.
3916 PH/65 Added configuration files for NetBSD3.
3918 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3920 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3921 is preferred over IPv4.
3923 PH/68 The bounce_return_message and bounce_return_body options were not being
3924 honoured for bounces generated during the reception of non-SMTP messages.
3925 In particular, this applied to messages rejected by the ACL. This bug has
3926 been fixed. However, if bounce_return_message is true and bounce_return_
3927 body is false, the headers that are returned for a non-SMTP message
3928 include only those that have been read before the error was detected.
3929 (In the case of an ACL rejection, they have all been read.)
3931 PH/69 The HTML version of the specification is now built in a directory called
3932 spec_html instead of spec.html, because the latter looks like a path with
3933 a MIME-type, and this confuses some software.
3935 PH/70 Catch two compiler warnings in sieve.c.
3937 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3938 function verify_get_ident() calls ip_connect() to connect a socket, but
3939 if the "connect()" function timed out, ip_connect() used to close the
3940 socket. However, verify_get_ident() also closes the socket later, and in
3941 between Exim writes to the log, which may get opened at this point. When
3942 the socket was closed in ip_connect(), the log could get the same file
3943 descriptor number as the socket. This naturally causes chaos. The fix is
3944 not to close the socket in ip_connect(); the socket should be closed by
3945 the function that creates it. There was only one place in the code where
3946 this was missing, in the iplookup router, which I don't think anybody now
3947 uses, but I've fixed it anyway.
3949 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3950 well as to direct DNS lookups. Otherwise the handling of names in host
3951 lists is inconsistent and therefore confusing.
3957 PH/01 Two changes to the default runtime configuration:
3959 (1) Move the checks for relay_from_hosts and authenticated clients from
3960 after to before the (commented out) DNS black list checks.
3962 (2) Add control=submission to the relay_from_hosts and authenticated
3963 clients checks, on the grounds that messages accepted by these
3964 statements are most likely to be submissions.
3966 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3968 (1) Generate an error if the third argument for the ${prvs expansion is
3971 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3974 (3) Reset the variables that are obtained from the first argument of
3975 ${prvscheck and used in the second argument before leaving the code,
3976 because their memory is reclaimed, so using them afterwards may do
3979 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3980 one (it's much easier than Tom thought :-).
3982 (5) Because of (4), we can now allow for the use of $prvscheck_result
3983 inside the third argument.
3985 PH/03 For some reason, the default setting of PATH when running a command from
3986 a pipe transport was just "/usr/bin". I have changed it to
3989 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3990 anything to be listed in the output from -bV.
3992 PH/05 When a filter generated an autoreply, the entire To: header line was
3993 quoted in the delivery log line, like this:
3995 => >A.N.Other <ano@some.domain> <original@ddress> ...
3997 This has been changed so that it extracts the operative address. There
3998 may be more than one such address. If so, they are comma-separated, like
4001 => >ano@some.domain,ona@other.domain <original@ddress> ...
4003 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
4004 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
4005 not being looked up in the rDNS to get a host name, Exim was showing the
4006 IP address twice in Received: lines, even though the IP addresses were
4007 identical. For example:
4009 Received: from [1.2.3.4] (helo=[1.2.3.4])
4011 However, if the real host name was known, it was omitting the HELO data
4012 if it matched the actual IP address. This has been tidied up so that it
4013 doesn't show the same IP address twice.
4015 PH/07 When both +timestamp and +memory debugging was on, the value given by
4016 $tod_xxx expansions could be wrong, because the tod_stamp() function was
4017 called by the debug printing, thereby overwriting the timestamp buffer.
4018 Debugging no longer uses the tod_stamp() function when +timestamp is set.
4020 PH/08 When the original message was included in an autoreply transport, it
4021 always said "this is a copy of the message, including all the headers",
4022 even if body_only or headers_only was set. It now gives an appropriate
4025 PH/09 Applied a patch from the Sieve maintainer which:
4027 o fixes some comments
4028 o adds the (disabled) notify extension core
4029 o adds some debug output for the result of if/elsif tests
4030 o points to the current vacation draft in the documentation
4031 and documents the missing references header update
4035 o fixes a bug in processing the envelope test (when testing
4036 multiple envelope elements, the last element determined the
4039 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
4040 Electronic Mail") by including:
4042 Auto-submitted: auto-generated
4044 in the messages that it generates (bounce messages and others, such as
4045 warnings). In the case of bounce messages for non-SMTP messages, there was
4046 also a typo: it was using "Auto_submitted" (underscore instead of
4047 hyphen). Since every message generated by Exim is necessarily in response
4048 to another message, thes have all been changed to:
4050 Auto-Submitted: auto-replied
4052 in accordance with these statements in the RFC:
4054 The auto-replied keyword:
4056 - SHOULD be used on messages sent in direct response to another
4057 message by an automatic process,
4059 - MUST NOT be used on manually-generated messages,
4061 - MAY be used on Delivery Status Notifications (DSNs) and Message
4062 Disposition Notifications (MDNs),
4064 - MUST NOT be used on messages generated by automatic or periodic
4065 processes, except for messages which are automatic responses to
4068 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
4069 to the default Received: header definition.
4071 PH/12 Added log selector acl_warn_skipped (default on).
4073 PH/13 After a successful wildlsearch lookup, discard the values of numeric
4074 variables because (a) they are in the wrong storage pool and (b) even if
4075 they were copied, it wouldn't work properly because of the caching.
4077 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
4078 checking when decoding. Apparently there are clients that generate
4079 overlong encoded strings. Why am I not surprised?
4081 PH/15 If the first argument of "${if match_address" was not empty, but did not
4082 contain an "@" character, Exim crashed. Now it writes a panic log message
4083 and treats the condition as false.
4085 PH/16 In autoreply, treat an empty string for "once" the same as unset.
4087 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
4088 extension "envelope-auth". The code is finished and in agreement with
4089 other implementations, but there is no documentation so far and in fact,
4090 nobody wrote the draft yet. This extension is currently #undef'ed, thus
4091 not changing the active code.
4093 Print executed "if" and "elsif" statements when debugging is used. This
4094 helps a great deal to understand what a filter does.
4096 Document more things not specified clearly in RFC3028. I had all this
4097 sorted out, when out of a sudden new issues came to my mind. Oops."
4099 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
4102 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
4103 canonical form (as documented). However, after a host name lookup from
4104 the IP address, check_host() was doing a simple string comparison with
4105 addresses acquired from the DNS when checking that the found name did
4106 have the original IP as one of its addresses. Since any found IPv6
4107 addresses are likely to be in abbreviated form, the comparison could
4108 fail. Luckily, there already exists a function for doing the comparison
4109 by converting both addresses to binary, so now that is used instead of
4110 the text comparison.
4112 PH/20 There was another similar case to PH/19, when a complete host name was
4113 given in a host list; looking up its IP address could give an abbreviated
4114 form, whereas the current host's name might or might not be abbreviated.
4115 The same fix has been applied.
4121 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
4122 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
4125 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
4126 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
4128 PH/03 Typo: missing ".o" in src/pcre/Makefile.
4130 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
4131 header line, restrict the check to what is listed in RFCs 2369 and 2929.
4132 Also, for "Auto-Submitted", treat anything other than "no" as
4133 non-personal, in accordance with RFC 3834. (Previously it treated
4134 anything starting "auto-" as non-personal.)
4136 TF/01 The control=submission/name=... option had a problem with syntax
4137 errors if the name included a slash character. The /name= option
4138 now slurps the rest of the string, so it can include any characters
4139 but it must come last in the list of options (after /sender_retain
4142 PH/05 Some modifications to the interface to the fake nameserver for the new
4150 TK/01 Added the "success_on_redirect" address verification option. See
4151 NewStuff for rationale and an example.
4153 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
4155 PH/02 Patch to exigrep to allow it to work on syslog lines.
4157 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
4158 fread() to skip over the body file's header line, because in Cygwin the
4159 header line is locked and is inaccessible.
4161 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
4162 co-exist for some time) to make it clear that it is the Exim ID that is
4163 referenced, not the Message-ID: header line.
4165 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
4166 string_format() function, because snprintf() does not exist on all
4169 PH/06 The use of forbid_filter_existstest now also locks out the use of the
4170 ${stat: expansion item.
4172 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
4173 protocol synchronization error", to keep the pedants happy.
4175 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
4176 well as for IRIX systems, when gcc is being used. See the host.c source
4179 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
4181 PH/10 Named domain lists were not working if used in a queue_smtp_domains
4184 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
4185 transport and to the smtp transport in LMTP mode.
4187 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
4189 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
4190 run a filter in a subprocess. This could lead to confusion in subsequent
4191 lookups in the parent process. There should also be a search_tidyup() at
4192 the end of the subprocess.
4194 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
4195 only if the host matched helo_try_verify_hosts, which caused the
4196 verification to occur when the EHLO/HELO command was issued. The ACL just
4197 tested the remembered result. Now, if a previous verification attempt has
4198 not happened, "verify = helo" does it there and then.
4200 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
4202 TK/03 Fix log output including CR from clamd.
4204 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
4205 exist provoked a memory error which could cause a segfault.
4207 PH/15 Installed PCRE 6.2
4209 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
4211 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
4212 of the problem. Specifically, suggested +O2 rather than +O1 for the
4215 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
4217 PH/20 If a delivery was routed to a non-standard port by means of an SRV
4218 record, the port was not correctly logged when the outgoing_port log
4219 selector was set (it logged the transort's default port).
4221 PH/21 Added support for host-specific ports to manualroute, queryprogram,
4222 fallback_hosts, and "hosts" in the smtp transport.
4224 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
4225 host errors such as "Connection refused".
4227 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
4228 authentication with radiusclient 0.4.9:
4230 - Error returned from rc_read_config was caught wrongly
4231 - Username/password not passed on to radius server due to wrong length.
4233 The presumption is that some radiusclient API changes for 4.51/PH/17
4234 were not taken care of correctly. The code is still untested by me (my
4235 Linux distribution still has 0.3.2 of radiusclient), but it was
4236 contributed by a Radius user.
4238 PH/24 When doing a callout, the value of $domain wasn't set correctly when
4239 expanding the "port" option of the smtp transport.
4241 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
4242 while reading a MIME header. Thanks to Tom Hughes for a patch.
4244 PH/24 Include config.h inside local_scan.h so that configuration settings are
4247 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
4248 that in an ACL for RCPT (for example), you can examine exactly what was
4251 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
4252 commands, but it was not correctly comparing the address with the actual
4253 client host address. Thus, it would show the EHLO address in Received:
4254 header lines when this was not necessary.
4256 PH/27 Added the % operator to ${eval:}.
4258 PH/28 Exim tries to create and chdir to its spool directory when it starts;
4259 it should be ignoring failures (because with -C, for example, it has lost
4260 privilege). It wasn't ignoring creation failures other than "already
4263 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
4266 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
4267 because an input line was too long, either on its own, or by virtue of
4268 too many continuations, the temporary file was not being removed, and the
4269 return code was incorrect.
4271 PH/31 Missing "BOOL" in function definition in filtertest.c.
4273 PH/32 Applied Sieve patches from the maintainer.
4275 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
4277 PH/33 Added "verify = not_blind".
4279 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
4280 Local/Makefile (with some defaults set). These are used in built scripts
4281 such as exicyclog, but they have never been used in the exim_install
4282 script (though there are many overriding facilities there). I have
4283 arranged that the exim_install script now takes note of these two
4286 PH/35 Installed configuration files for Dragonfly.
4288 PH/36 When a locally submitted message by a trusted user did not contain a
4289 From: header, and the sender address was obtained from -f or from an SMTP
4290 MAIL command, and the trusted user did not use -F to supply a sender
4291 name, $originator_name was incorrectly used when constructing a From:
4292 header. Furthermore, $originator_name was used for submission mode
4293 messages from external hosts without From: headers in a similar way,
4294 which is clearly wrong.
4296 PH/37 Added control=suppress_local_fixups.
4298 PH/38 When log_selector = +received_sender was set, and the addition of the
4299 sender made the log line's construction buffer exactly full, or one byte
4300 less than full, an overflow happened when the terminating "\n" was
4303 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
4304 when the result of a list match is failure because a DNS lookup failed.
4306 PH/40 RM_COMMAND is now used in the building process.
4308 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
4309 the "build-* directories that it finds.
4311 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
4312 address in a domain literal was a prefix of an interface address.
4314 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
4315 when verifying a sender address, unless rewrite_headers is false.
4317 PH/44 Wrote a long comment about why errors_to addresses are verified as
4318 recipients, not senders.
4320 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
4321 the ratelimit ACL was added.
4323 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
4325 PH/46 Added extra information about PostgreSQL errors to the error string.
4327 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
4328 avoiding the need to install special zones in a real server. This is
4329 backwards compatible; if it can't find the fake resolver, it drops back.
4330 Thus, both old and new test suites can be run.
4332 TF/02 Added util/ratelimit.pl
4334 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
4335 clock is set back in time.
4337 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
4338 Candler <B.Candler@pobox.com>.
4340 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
4341 OK for addresses that are the result of redirections.
4343 PH/48 A number of further additions for the benefit of the new test suite,
4344 including a fake gethostbyname() that interfaces to the fake DNS resolver
4347 TF/06 The fix for widen_domains has also been applied to qualify_single and
4348 search_parents which are the other dnslookup options that can cause
4351 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
4354 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
4356 TF/07 Exim produced the error message "an SRV record indicated no SMTP
4357 service" if it encountered an MX record with an empty target hostname.
4358 The message is now "an MX or SRV record indicated no SMTP service".
4360 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
4361 if the DNS of the sending site is misconfigured. This is quite a
4362 common situation. This change restores the behaviour of treating a
4363 helo verification defer as a failure.
4365 PH/51 If self=fail was set on a router, the bounce message did not include the
4366 actual error message.
4372 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
4374 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
4375 command itself ended in error, the underlying message about the transport
4376 filter timeout was being overwritten with the pipe command error. Now the
4377 underlying error message should be appended to the second error message.
4379 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4381 PH/02 Reduce the amount of output that "make" produces by default. Full output
4382 can still be requested.
4384 PH/03 The warning log line about a condition test deferring for a "warn" verb
4385 was being output only once per connection, rather than after each
4386 occurrence (because it was using the same function as for successful
4387 "warn" verbs). This seems wrong, so I have changed it.
4389 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4390 it should not have, which might have caused a crash in the right
4391 circumstances, but probably never did.
4393 PH/04 Installed a modified version of Tony Finch's patch to make submission
4394 mode fix the return path as well as the Sender: header line, and to
4395 add a /name= option so that you can make the user's friendly name appear
4398 TF/03 Added the control = fakedefer ACL modifier.
4400 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4401 Mark Lowes for thorough testing.
4403 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4405 TK/03 Merged latest SRS patch from Miles Wilton.
4407 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4408 with the definition in sysexits.h (which is #included earlier).
4409 Fortunately, Exim does not actually use EX_OK. The code used to try to
4410 preserve the sysexits.h value, by assuming that macro definitions were
4411 scanned for macro replacements. I have been disabused of this notion,
4412 so now the code just undefines EX_OK before #including unistd.h.
4414 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4415 in the smtp transport. When a block could not be written in a single
4416 write() function, the timeout was being re-applied to each part-write.
4417 This seems wrong - if the receiver was accepting one byte at a time it
4418 would take for ever. The timeout is now adjusted when this happens. It
4419 doesn't have to be particularly precise.
4421 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4422 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4424 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4425 without a domain if the check to <postmaster@domain> fails.
4427 SC/01 Eximstats: added -xls and the ability to specify output files
4428 (patch written by Frank Heydlauf).
4430 SC/02 Eximstats: use FileHandles for outputting results.
4432 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4434 SC/04 Eximstats: fixed display of large numbers with -nvr option
4436 SC/05 Eximstats: fixed merging of reports with empty tables.
4438 SC/06 Eximstats: added the -include_original_destination flag
4440 SC/07 Eximstats: removed tabs and trailing whitespace.
4442 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4444 TK/06 MBOX spool code: Add real "From " MBOX separator line
4445 so the .eml file is really in mbox format (even though
4446 most programs do not really care). Patch from Alex Miller.
4448 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4449 The latter is generated from $received_to and is only set if the
4450 message has one envelope recipient. SA can use these headers,
4451 obviously out-of-the-box. Patch from Alex Miller.
4453 PH/08 The ${def test on a variable was returning false if the variable's
4454 value was "0", contrary to what the specification has always said!
4455 The result should be true unless the variable is empty.
4457 PH/09 The syntax error of a character other than { following "${if
4458 def:variable_name" (after optional whitespace) was not being diagnosed.
4459 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4460 accidental colon was present, for example, could give incorrect results.
4462 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4463 result is used (not including appendfile, where other changes are about
4466 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4467 This involved changing a lot of size variables from int to off_t. It
4468 should work with maildirs and everything.
4470 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4471 spamd dying while we are connected to it.
4473 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4476 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4477 function for BDB 4.3.
4479 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4481 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4482 the sources for PCRE have been re-arranged and more files are now
4485 PH/15 The code I had for printing potentially long long variables in PH/11
4486 above was not the best (it lost precision). The length of off_t variables
4487 is now inspected at build time, and an appropriate printing format (%ld
4488 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4489 to be "long long int" or "long int". This is needed for the internal
4490 formatting function string_vformat().
4492 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4493 the configuration file to be ":syslog", then the script "guesses" where
4494 the logs files are, rather than using the compiled in default. In our
4495 case the guess is not the same as the compiled default, so the script
4496 suddenly stopped working when I started to use syslog. The patch checks
4497 to see if log_file_path is "". If so, it attempts to read it from exim
4498 with no configuration file to get the compiled in version, before it
4499 falls back to the previous guessing code."
4501 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4502 implementing BATV in an Exim configuration. See NewStuff for the gory
4505 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4506 Makefile that are specific to HP-UX.
4508 PH/18 If the "use_postmaster" option was set for a recipient callout together
4509 with the "random" option, the postmaster address was used as the MAIL
4510 FROM address for the random test, but not for the subsequent recipient
4511 test. It is now used for both.
4513 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4514 patch removes a few documentation additions to RFC 3028, because the
4515 latest draft now contains them. It adds the new en;ascii-case comparator
4516 and a new error check for 8bit text in MIME parts. Comparator and
4517 require names are now matched exactly. I enabled the subaddress
4518 extension, but it is not well tested yet (read: it works for me)."
4520 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4521 rework some of the code of TK/09 above to avoid the hardwired use of
4522 "%lld" and "long long". Replaced the call to snprintf() with a call to
4525 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4526 records point to non-existent hosts", "retry timeout exceeded", and
4527 "retry time not reached for any host after a long failure period".
4529 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4530 experimental DomainKeys support:
4532 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4533 (2) On an error such as an illegally used "control", the wrong name for
4534 the control was given.
4536 These problems did NOT occur unless DomainKeys support was compiled.
4538 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4540 PH/24 Added ${if match_ip condition.
4542 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4543 databases so that it will be absolutely obvious if a crash occurs in the
4544 DB library. This is a regular occurrence (often caused by mis-matched
4547 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4548 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4549 picked up on a user's system that detects such things. There doesn't seem
4550 to be a gcc warning option for this - only an attribute that has to be
4551 put on the function's prototype. It seems that in Fedora Core 4 they have
4552 set this on a number of new functions. No doubt there will be more in due
4555 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4556 specify a transport. However, if an address that was verified by such a
4557 router was the subject of a callout, Exim crashed because it tried to
4558 read the rcpt_include_affixes from the non-existent transport. Now it
4559 just assumes that the setting of that option is false. This bug was
4560 introduced by 4.51/PH/31.
4562 PH/28 Changed -d+all to exclude +memory, because that information is very
4563 rarely of interest, but it makes the output a lot bigger. People tend to
4564 do -d+all out of habit.
4566 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4567 code in os-type was giving problems when libc.so lives in lib64, like on
4570 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4571 aren't the modern standard, and it seems that some systems' include files
4572 don't always have them. Exim was already checking for some of the newer
4573 ones like T_AAAA, and defining it itself. I've added checks for all the
4574 record types that Exim uses.
4576 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4577 not automatically generating a new one, as it is supposed to. This
4578 prevented TLS from working. If the file did exist, but contained invalid
4579 data, a new version was generated, as expected. It was only the case of a
4580 non-existent file that was broken.
4582 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4583 with a change in libdomainkeys > 0.64.
4585 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4586 from DNS. If the selector record carries the flag, it now has
4587 precedence over the domain-wide flag.
4589 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4591 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4592 the use of a port name that isn't defined in /etc/services) occurred, the
4593 message was deferred as in a normal delivery, and thus remained on the
4594 spool, instead of being failed because of the mua_wrapper setting. This
4595 is now fixed, and I tidied up some of the mua_wrapper messages at the
4598 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4599 the messages in a hash of arrays rather than using individual hashes.
4600 This is a bit cleaner and results in dramatic memory savings, albeit
4601 at a slight CPU cost.
4603 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4604 as requested by Marc Sherman.
4606 SC/10 Eximstats: added histograms for user specified patterns as requested
4609 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4611 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4612 fopen() in the content-scanning modules that did not already have it.
4618 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4619 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4621 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4623 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4625 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4626 to test Sieve filters that use "vacation".
4628 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4629 that changes the way the GnuTLS parameters are stored in the cache file.
4630 The new format can be generated externally. For backward compatibility,
4631 if the data in the cache doesn't make sense, Exim assumes it has read an
4632 old-format file, and it generates new data and writes a new file. This
4633 means that you can't go back to an older release without removing the
4636 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4637 work if there are any delivery delays because "one_time" forces the
4638 parent to be marked "delivered", so its unseen clone is never tried
4639 again. For this reason, Exim now forbids the simultaneous setting of
4642 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4643 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4644 entry if you want to know the details. The fix, however, neglected to
4645 consider the case when local delivery batching is involved. The test for
4646 "previously delivered" was not happening when checking to see if an
4647 address could be batched with a previous (undelivered) one; under
4648 certain circumstances this could lead to multiple deliveries to the same
4651 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4652 in its include files, and this causes problems building Exim.
4654 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4655 header_syntax) but Exim was just ignoring anything given after a slash.
4656 In particular, this caused confusion with an attempt to use "verify =
4657 reverse_host_lookup/defer_ok". An error is now given when options are
4658 supplied for verify items that do not have them. (Maybe reverse_host_
4659 lookup should have a defer_ok option, but that's a different point.)
4661 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4662 defined by RFC 821) to 2048, because there were problems with some AUTH
4663 commands, and RFC 1869 says the size should be increased for extended
4664 SMTP commands that take arguments.
4666 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4669 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4670 "unknown" error; now it says that the functionality isn't in the binary.
4672 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4673 an address' error message when a string expansion fails (syntax or
4674 whatever). Otherwise the password may appear in the log. Following change
4675 PH/42 below, there is no longer a chance of it appearing in a bounce
4678 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4680 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4681 of its addresses was ever tried. (Bugzilla bug #2.)
4683 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4684 the result incorrectly in the debug output. (It correctly added a newline
4685 to what was transported.)
4687 TF/01 Added $received_time.
4689 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4690 commented out examples of how to interface to a virus scanner and to
4691 SpamAssassin. Also added commented examples of av_scanner and
4692 spamd_address settings.
4694 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4695 and controls are allowed in which ACLs. There were a couple of minor
4696 errors. Some of the entries in the conditions table (which is a table of
4697 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4698 negation of where the condition IS allowed.
4700 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4702 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4703 header file does not have a version number, so I've had to invent a new
4704 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4705 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4706 radiusclient), but it was contributed by a Radius user.
4708 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4709 files or directories, for OpenSSL.
4711 PH/19 When an Exim process that is running as root has to create an Exim log
4712 file, it does so in a subprocess that runs as exim:exim so as to get the
4713 ownership right at creation (otherwise, other Exim processes might see
4714 the file with the wrong ownership). There was no test for failure of this
4715 fork() call, which would lead to the process getting stuck as it waited
4716 for a non-existent subprocess. Forks do occasionally fail when resources
4717 run out. I reviewed all the other calls to fork(); they all seem to check
4720 PH/20 When checking for unexpected SMTP input at connect time (before writing
4721 the banner), Exim was not dealing correctly with a non-positive return
4722 from the read() function. If the client had disconnected by this time,
4723 the result was a log entry for a synchronization error with an empty
4724 string after "input=" when read() returned zero. If read() returned -1
4725 (an event I could not check), uninitialized data bytes were printed.
4726 There were reports of junk text (parts of files, etc) appearing after
4729 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4731 PH/22 Added support for macro redefinition, and (re)definition in between
4732 driver and ACL definitions.
4734 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4735 forgetting to use the resulting value; it was using the unexpanded value.
4737 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4738 hadn't been configured. The fix is from Juergen Kreileder, who
4739 understands it better than I do:
4741 "Here's what I see happening with three configured cyrus_sasl
4742 authenticators configured (plain, login, cram-md5):
4744 On startup auth_cyrus_sasl_init() gets called for each of these.
4745 This means three calls to sasl_listmech() without a specified mech_list.
4746 => SASL tests which mechs of all available mechs actually work
4747 => three warnings about OTP not working
4748 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4750 With the patch, sasl_listmech() also gets called three times. But now
4751 SASL's mech_list option is set to the server_mech specified in the the
4752 authenticator. Or in other words, the answer from sasl_listmech()
4753 gets limited to just the mech you're testing for (which is different
4755 => the return list contains just 'plain' or 'login', 'cram-md5' or
4756 nothing depending on the value of ob->server_mech.
4758 I've just tested the patch: Authentication still works fine,
4759 unavailable mechs specified in the exim configuration are still
4760 caught, and the auth.log warnings about OTP are gone."
4762 PH/25 When debugging is enabled, the contents of the command line are added
4763 to the debugging output, even when log_selector=+arguments is not
4766 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4767 answer is "GNU", and only if the return is "GNU/something" is the answer
4770 PH/27 $acl_verify_message is now set immediately after the failure of a
4771 verification in an ACL, and so is available in subsequent modifiers. In
4772 particular, the message can be preserved by coding like this:
4774 warn !verify = sender
4775 set acl_m0 = $acl_verify_message
4777 Previously, $acl_verify_message was set only while expanding "message"
4778 and "log_message" when a very denied access.
4780 PH/28 Modified OS/os.c-Linux with
4782 -#ifndef OS_LOAD_AVERAGE
4783 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4785 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4786 nomenclature these days.)
4788 PH/29 Installed patch from the Sieve maintainer that adds the options
4789 sieve_useraddress and sieve_subaddress to the redirect router.
4791 PH/30 In these circumstances:
4792 . Two addresses routed to the same list of hosts;
4793 . First host does not offer TLS;
4794 . First host accepts first address;
4795 . First host gives temporary error to second address;
4796 . Second host offers TLS and a TLS session is established;
4797 . Second host accepts second address.
4798 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4799 and peerdn, if requested) that were in fact used only for the second
4802 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4803 attention to any local part prefix or suffix that was matched by the
4804 router that accepted the address. It now behaves in the same way as it
4805 does for delivery: the affixes are removed from the local part unless
4806 rcpt_include_affixes is set on the transport.
4808 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4809 timeout during the DATA phase of an incoming message.
4811 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4812 applied a patch sanctioned by the Sieve maintainer.
4814 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4815 the uid or gid is negative. A case of a negative gid caused this to be
4816 noticed. The fix allows for either to be negative.
4818 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4819 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4822 PH/36 The change PH/12 above was broken. Fixed it.
4824 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4825 the grounds that routing the same address twice would always produce the
4826 same answer. This might have been true once, but it is certainly no
4827 longer true now. Routing a child address may depend on the previous
4828 routing that produced that child. Some complicated redirection strategies
4829 went wrong when messages had multiple recipients, and made Exim's
4830 behaviour dependent on the order in which the addresses were given.
4832 I have moved the duplicate checking until after the routing is complete.
4833 Exim scans the addresses that are assigned to local and remote
4834 transports, and removes any duplicates. This means that more work will be
4835 done, as duplicates will always all be routed, but duplicates are
4836 presumably rare, so I don't expect this is of any significance.
4838 For deliveries to pipes, files, and autoreplies, the duplicate checking
4839 still happens during the routing process, since they are not going to be
4842 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4843 It corrects a timeout issue with spamd. This is Ian's comment: "The
4844 background is that sometimes spamd either never reads data from a
4845 connection it has accepted, or it never writes response data. The exiscan
4846 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4847 blindly assumes that writes won't block so it may never time out."
4849 PH/39 Allow G after quota size as well as K and M.
4851 PH/40 The value set for $authenticated_id in an authenticator may not contain
4852 binary zeroes or newlines because the value is written to log lines and
4853 to spool files. There was no check on this. Now the value is run through
4854 the string_printing() function so that such characters are converted to
4855 printable escape sequences.
4857 PH/41 $message_linecount is a new variable that contains the total number of
4858 lines in the message. Compare $body_linecount, which is the count for the
4861 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4862 bounce and delay warning messages, except in certain special cases, which
4865 (a) An SMTP error message from a remote host;
4866 (b) A message specified in a :fail: redirection;
4867 (c) A message specified in a "fail" command in a system filter;
4868 (d) A message specified in a FAIL return from the queryprogram router;
4869 (e) A message specified by the cannot_route_message router option.
4871 In these cases only, Exim does include the error details in bounce and
4872 warning messages. There are also a few cases where bland messages such
4873 as "unrouteable address" or "local delivery error" are given.
4875 PH/43 $value is now also set for the "else" part of a ${run expansion.
4877 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4878 being worked on, but at least Exim now implements the latest version to
4881 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4882 process to complete was treated as a delivery failure, a timeout while
4883 writing the message to the pipe was logged, but erroneously treated as a
4884 successful delivery. Such timeouts include transport filter timeouts. For
4885 consistency with the overall process timeout, these timeouts are now
4886 treated as errors, giving rise to delivery failures by default. However,
4887 there is now a new Boolean option for the pipe transport called
4888 timeout_defer, which, if set TRUE, converts the failures into defers for
4889 both kinds of timeout. A transport filter timeout is now identified in
4892 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4893 systems where "make" and "gmake" are different, calling "gmake" at top
4894 level broke things. I've arranged for the value of $(MAKE) to be passed
4895 from the Makefile to this script so that it can call the same version of
4899 A note about Exim versions 4.44 and 4.50
4900 ----------------------------------------
4902 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4903 changes of various kinds. As a consequence, a big documentation update was
4904 needed. This delayed the release for rather longer than seemed good, especially
4905 in the light of a couple of (minor) security issues. Therefore, the changes
4906 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4907 release. So 4.44 and 4.50 are in effect two different branches that both start
4910 I have left the 4.50 change log unchanged; it contains all the changes since
4911 4.43. The change log for 4.44 is below; many of its items are identical to
4912 those for 4.50. This seems to be the most sensible way to preserve the
4913 historical information.
4919 1. Minor wording change to the doc/README.SIEVE file.
4921 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4922 computation of the current number of files was incorrect.
4924 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4925 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4928 4. Give more explanation in the error message when the command for a transport
4929 filter fails to execute.
4931 5. There are several places where Exim runs a non-Exim command in a
4932 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4933 was being done only for the command run by the queryprogram router. It is
4934 now done for all such subprocesses. The other cases are: ${run, transport
4935 filters, and the commands run by the lmtp and pipe transports.
4937 6. Added CONFIGURE_GROUP build-time option.
4939 7. Some older OS have a limit of 256 on the maximum number of file
4940 descriptors. Exim was using setrlimit() to set 1000 as a large value
4941 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4942 systems. I've change it so that if it can't get 1000, it tries for 256.
4944 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4945 was an oversight, and furthermore, ever since the addition of extra
4946 controls (e.g. 4.43/32), the checks on when to allow different forms of
4947 "control" were broken. There should now be diagnostics for all cases when a
4948 control that does not make sense is encountered.
4950 9. Added the /retain_sender option to "control=submission".
4952 10. $recipients is now available in the predata ACL (oversight).
4954 11. Tidy the search cache before the fork to do a delivery from a message
4955 received from the command line. Otherwise the child will trigger a lookup
4956 failure and thereby defer the delivery if it tries to use (for example) a
4957 cached ldap connection that the parent has called unbind on.
4959 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4960 of $address_data from the recipient verification was clobbered by the
4961 sender verification.
4963 13. The value of address_data from a sender verification is now available in
4964 $sender_address_data in subsequent conditions in the ACL statement.
4966 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4968 15. Added a new option "connect=<time>" to callout options, to set a different
4971 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4972 was its contents. (It was OK if the option was not defined at all.)
4974 17. A "Completed" log line is now written for messages that are removed from
4975 the spool by the -Mrm option.
4977 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4978 information about exactly what failed.
4980 19. Added -dd to debug only the daemon process.
4982 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4983 handles timeouts, both on the server side and network timeouts. Renamed the
4984 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4986 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4987 It is now set to "smtps".
4989 22. $host_address is now set to the target address during the checking of
4990 ignore_target_hosts.
4992 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4993 being passed; this would have caused $sender_host_name to have been used if
4994 matching the list had actually called for a host name (not very likely,
4995 since this list is usually IP addresses). A host name is now passed as
4998 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4999 code that specifies a non-restarting handler (typically sigaction() in
5000 modern systems) in an attempt to fix a rare and obscure crash bug.
5002 25. Narrowed the window for a race in the daemon that could cause it to ignore
5003 SIGCHLD signals. This is not a major problem, because they are used only to
5004 wake it up if nothing else does.
5006 26. A malformed maildirsize file could cause Exim to calculate negative values
5007 for the mailbox size or file count. Odd effects could occur as a result.
5008 The maildirsize information is now recalculated if the size or filecount
5011 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5012 support for a long time. Removed HAVE_SYS_VFS_H.
5014 28. Installed the latest version of exipick from John Jetmore.
5016 29. In an address list, if the pattern was not a regular expression, an empty
5017 subject address (from a bounce message) matched only if the pattern was an
5018 empty string. Non-empty patterns were not even tested. This was the wrong
5019 because it is perfectly reasonable to use an empty address as part of a
5020 database query. An empty address is now tested by patterns that are
5021 lookups. However, all the other forms of pattern expect the subject to
5022 contain a local part and a domain, and therefore, for them, an empty
5023 address still always fails if the pattern is not itself empty.
5025 30. Exim went into a mad DNS loop when attempting to do a callout where the
5026 host was specified on an smtp transport, and looking it up yielded more
5027 than one IP address.
5029 31. Re-factored the code for checking spool and log partition space into a
5030 function that finds that data and another that does the check. The former
5031 is then used to implement four new variables: $spool_space, $log_space,
5032 $spool_inodes, and $log_inodes.
5034 32. The RFC2047 encoding function was originally intended for short strings
5035 such as real names; it was not keeping to the 75-character limit for
5036 encoded words that the RFC imposes. It now respects the limit, and
5037 generates multiple encoded words if necessary. To be on the safe side, I
5038 have increased the buffer size for the ${rfc2047: expansion operator from
5041 33. It is now permitted to omit both strings after an "if" condition; if the
5042 condition is true, the result is "true". As before, when the second string
5043 is omitted, a false condition yields an empty string. This makes it less
5044 cumbersome to write custom ACL and router conditions.
5046 34. Failure to deliver a bounce message always caused it to be frozen, even if
5047 there was an errors_to setting on the router. The errors_to setting is now
5050 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5051 canonical form (fully expanded) before being placed in
5052 $sender_host_address.
5054 36. The table in the code that translates DNS record types into text (T_A to
5055 "A" for instance) was missing entries for NS and CNAME. It is just possible
5056 that this could have caused confusion if both these types were looked up
5057 for the same domain, because the text type is used as part of Exim's
5058 per-process caching. But the chance of anyone hitting this buglet seems
5061 37. The dnsdb lookup has been extended in a number of ways.
5063 (1) There is a new type, "zns", which walks up the domain tree until it
5064 finds some nameserver records. It should be used with care.
5066 (2) There is a new type, "mxh", which is like "mx" except that it returns
5067 just the host names, not the priorities.
5069 (3) It is now possible to give a list of domains (or IP addresses) to be
5070 looked up. The behaviour when one of the lookups defers can be
5071 controlled by a keyword.
5073 (4) It is now possible to specify the separator character for use when
5074 multiple records are returned.
5076 38. The dnslists ACL condition has been extended: it is now possible to supply
5077 a list of IP addresses and/or domains to be looked up in a particular DNS
5080 39. Added log_selector=+queue_time_overall.
5082 40. When running the queue in the test harness, wait just a tad after forking a
5083 delivery process, to get repeatability of debugging output.
5085 41. Include certificate and key file names in error message when GnuTLS fails
5086 to set them up, because the GnuTLS error message doesn't include the name
5087 of the failing file when there is a problem reading it.
5089 42. Allow both -bf and -bF in the same test run.
5091 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
5093 44. The "Exiscan patch" is now merged into the mainline Exim source.
5095 45. Sometimes the final signoff response after QUIT could fail to get
5096 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5097 < 0 before doing a fflush(). This bug looks as though it goes back to the
5098 introduction of TLS in release 3.20, but "sometimes" must have been rare
5099 because the tests only now provoked it.
5101 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
5102 (this can affect the format of dates).
5104 47. exim_tidydb, when checking for the continued existence of a message for
5105 which it has found a message-specific retry record, was not finding
5106 messages that were in split spool directories. Consequently, it was
5107 deleting retry records that should have stayed in existence.
5109 48. Steve fixed some bugs in eximstats.
5111 49. The SPA authentication driver was not abandoning authentication and moving
5112 on to the next authenticator when an expansion was forced to fail,
5113 contradicting the general specification for all authenticators. Instead it
5114 was generating a temporary error. It now behaves as specified.
5116 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
5117 (the order specifies the preference for clients). The order is now AES256,
5118 AES128, 3DES, ARCFOUR128.
5120 51. Small patch to Sieve code - explicitly set From: when generating an
5123 52. Exim crashed if a remote delivery caused a very long error message to be
5124 recorded - for instance if somebody sent an entire SpamAssassin report back
5125 as a large number of 550 error lines. This bug was coincidentally fixed by
5126 increasing the size of one of Exim's internal buffers (big_buffer) that
5127 happened as part of the Exiscan merge. However, to be on the safe side, I
5128 have made the code more robust (and fixed the comments that describe what
5131 53. Now that there can be additional text after "Completed" in log lines (if
5132 the queue_time_overall log selector is set), a one-byte patch to exigrep
5133 was needed to allow it to recognize "Completed" as not the last thing in
5136 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
5137 patch that reportedly fixes this has been added. I am not expert enough to
5138 create a test for it. This is what the patch creator wrote:
5140 "I found a little strange behaviour of ldap code when working with
5141 Windows 2003 AD Domain, where users was placed in more than one
5142 Organization Units. When I tried to give exim partial DN, the exit code
5143 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
5144 But simultaneously result of request was absolutely normal ldap result,
5145 so I produce this patch..."
5147 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
5148 REFERENCE, so I have modified the code to exclude the patch when that macro
5151 55. Some experimental protocols are using DNS PTR records for new purposes. The
5152 keys for these records are domain names, not reversed IP addresses. The
5153 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5154 leaves it alone. Component reversal etc. now happens only for IP addresses.
5157 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
5159 57. Double the size of the debug message buffer (to 2048) so that more of very
5160 long debug lines gets shown.
5162 58. The exicyclog utility now does better if the number of log files to keep
5163 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
5165 59. Two changes related to the smtp_active_hostname option:
5167 (1) $smtp_active_hostname is now available as a variable.
5168 (2) The default for smtp_banner uses $smtp_active_hostname instead
5169 of $primary_hostname.
5171 60. The host_aton() function is supposed to be passed a string that is known
5172 to be a valid IP address. However, in the case of IPv6 addresses, it was
5173 not checking this. This is a hostage to fortune. Exim now panics and dies
5174 if the condition is not met. A case was found where this could be provoked
5175 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5176 components; fortuitously, this particular loophole had already been fixed
5177 by change 4.50/55 above.
5179 If there are any other similar loopholes, the new check in host_aton()
5180 itself should stop them being exploited. The report I received stated that
5181 data on the command line could provoke the exploit when Exim was running as
5182 exim, but did not say which command line option was involved. All I could
5183 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5184 running as the user.
5187 61. There was a buffer overflow vulnerability in the SPA authentication code
5188 (which came originally from the Samba project). I have added a test to the
5189 spa_base64_to_bits() function which I hope fixes it.
5192 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
5193 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
5195 63. The daemon start-up calls getloadavg() while still root for those OS that
5196 need the first call to be done as root, but it missed one case: when
5197 deliver_queue_load_max is set with deliver_drop_privilege. This is
5198 necessary for the benefit of the queue runner, because there is no re-exec
5199 when deliver_drop_privilege is set.
5201 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
5202 This has been fixed.
5204 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5205 was in use, was not putting the data itself into the right store pool;
5206 consequently, it could be overwritten for a subsequent message in the same
5207 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5210 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
5212 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
5213 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
5214 the function were treating the return as a boolean value, which happened to
5215 work because 0=false and not-0=true, but is not correct code.
5217 68. The host_aton() function was not handling scoped IPv6 addresses (those
5218 with, for example, "%eth0" on the end) correctly.
5220 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
5221 negated items (that is, ~something) in unsigned ints. Some compilers
5222 apparently mutter when there is no cast.
5224 70. If an address verification called from an ACL failed, and did not produce a
5225 user-specific message (i.e. there was only a "system" message), nothing was
5226 put in $acl_verify_message. In this situation, it now puts the system
5229 71. Change 4.23/11 added synchronization checking at the start of an SMTP
5230 session; change 4.31/43 added the unwanted input to the log line - except
5231 that it did not do this in the start of session case. It now does.
5233 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
5234 This is wrong and can cause the other end to generate a synchronization
5235 error if it is another Exim or anything else that does the synchronization
5236 check. A QUIT command is no longer sent after a timeout.
5238 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
5239 during host lookups.
5241 74. The defer_ok option of callout verification was not working if it was used
5242 when verifying addresses in header lines, that is, for this case:
5244 verify = header_sender/callout=defer_ok
5246 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
5247 those file descriptors could be used for SMTP connections. If anything
5248 wrote to stderr (the example that came up was "warn" in embedded Perl), it
5249 could be sent to the SMTP client, causing chaos. The daemon now opens
5250 stdin, stdout, and stderr to /dev/null when it puts itself into the
5253 76. Arrange for output from Perl's "warn" command to be written to Exim's main
5254 log by default. The user can override this with suitable Perl magic.
5256 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
5257 the log message when discard triggers, was not working for the DATA ACL or
5258 for the non-SMTP ACL.
5260 78. Error message wording change in sieve.c.
5262 79. If smtp_accept_max_per_host was set, the number of connections could be
5263 restricted to fewer than expected, because the daemon was trying to set up
5264 a new connection before checking whether the processes handling previous
5265 connections had finished. The check for completed processes is now done
5266 earlier. On busy systems, this bug wouldn't be noticed because something
5267 else would have woken the daemon, and it would have reaped the completed
5270 80. If a message was submitted locally by a user whose login name contained one
5271 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
5272 It caused a spool format error. I have fixed the spool reading code. A
5273 related problem was that the "from" clause in the Received: line became
5274 illegal because of the space(s). It is now covered by ${quote_local_part.
5276 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
5279 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
5281 83. Patch from Sieve maintainer to fix unterminated string problem in
5282 "vacation" handling.
5284 84. Some minor changes to the Linux configuration files to help with other
5285 OS variants using glibc.
5287 85. One more patch for Sieve to update vacation handling to latest spec.
5290 ----------------------------------------------------
5291 See the note above about the 4.44 and 4.50 releases.
5292 ----------------------------------------------------
5298 1. Change 4.43/35 introduced a bug that caused file counts to be
5299 incorrectly computed when quota_filecount was set in an appendfile
5302 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5303 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5306 3. Give more explanation in the error message when the command for a transport
5307 filter fails to execute.
5309 4. There are several places where Exim runs a non-Exim command in a
5310 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5311 was being done only for the command run by the queryprogram router. It is
5312 now done for all such subprocesses. The other cases are: ${run, transport
5313 filters, and the commands run by the lmtp and pipe transports.
5315 5. Some older OS have a limit of 256 on the maximum number of file
5316 descriptors. Exim was using setrlimit() to set 1000 as a large value
5317 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5318 systems. I've change it so that if it can't get 1000, it tries for 256.
5320 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5321 was an oversight, and furthermore, ever since the addition of extra
5322 controls (e.g. 4.43/32), the checks on when to allow different forms of
5323 "control" were broken. There should now be diagnostics for all cases when a
5324 control that does not make sense is encountered.
5326 7. $recipients is now available in the predata ACL (oversight).
5328 8. Tidy the search cache before the fork to do a delivery from a message
5329 received from the command line. Otherwise the child will trigger a lookup
5330 failure and thereby defer the delivery if it tries to use (for example) a
5331 cached ldap connection that the parent has called unbind on.
5333 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5334 of $address_data from the recipient verification was clobbered by the
5335 sender verification.
5337 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5338 was its contents. (It was OK if the option was not defined at all.)
5340 11. A "Completed" log line is now written for messages that are removed from
5341 the spool by the -Mrm option.
5343 12. $host_address is now set to the target address during the checking of
5344 ignore_target_hosts.
5346 13. When checking ignore_target_hosts for an ipliteral router, no host name was
5347 being passed; this would have caused $sender_host_name to have been used if
5348 matching the list had actually called for a host name (not very likely,
5349 since this list is usually IP addresses). A host name is now passed as
5352 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5353 code that specifies a non-restarting handler (typically sigaction() in
5354 modern systems) in an attempt to fix a rare and obscure crash bug.
5356 15. Narrowed the window for a race in the daemon that could cause it to ignore
5357 SIGCHLD signals. This is not a major problem, because they are used only to
5358 wake it up if nothing else does.
5360 16. A malformed maildirsize file could cause Exim to calculate negative values
5361 for the mailbox size or file count. Odd effects could occur as a result.
5362 The maildirsize information is now recalculated if the size or filecount
5365 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5366 support for a long time. Removed HAVE_SYS_VFS_H.
5368 18. Updated exipick to current release from John Jetmore.
5370 19. Allow an empty sender to be matched against a lookup in an address list.
5371 Previously the only cases considered were a regular expression, or an
5374 20. Exim went into a mad DNS lookup loop when doing a callout where the
5375 host was specified on the transport, if the DNS lookup yielded more than
5378 21. The RFC2047 encoding function was originally intended for short strings
5379 such as real names; it was not keeping to the 75-character limit for
5380 encoded words that the RFC imposes. It now respects the limit, and
5381 generates multiple encoded words if necessary. To be on the safe side, I
5382 have increased the buffer size for the ${rfc2047: expansion operator from
5385 22. Failure to deliver a bounce message always caused it to be frozen, even if
5386 there was an errors_to setting on the router. The errors_to setting is now
5389 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5390 canonical form (fully expanded) before being placed in
5391 $sender_host_address.
5393 24. Updated eximstats to version 1.33
5395 25. Include certificate and key file names in error message when GnuTLS fails
5396 to set them up, because the GnuTLS error message doesn't include the name
5397 of the failing file when there is a problem reading it.
5399 26. Expand error message when OpenSSL has problems setting up cert/key files.
5402 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5403 (this can affect the format of dates).
5405 28. exim_tidydb, when checking for the continued existence of a message for
5406 which it has found a message-specific retry record, was not finding
5407 messages that were in split spool directories. Consequently, it was
5408 deleting retry records that should have stayed in existence.
5410 29. eximstats updated to version 1.35
5411 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5412 1.35 - bugfix such that pie charts by volume are generated correctly
5414 30. The SPA authentication driver was not abandoning authentication and moving
5415 on to the next authenticator when an expansion was forced to fail,
5416 contradicting the general specification for all authenticators. Instead it
5417 was generating a temporary error. It now behaves as specified.
5419 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5420 (the order specifies the preference for clients). The order is now AES256,
5421 AES128, 3DES, ARCFOUR128.
5423 31. Small patch to Sieve code - explicitly set From: when generating an
5426 32. Exim crashed if a remote delivery caused a very long error message to be
5427 recorded - for instance if somebody sent an entire SpamAssassin report back
5428 as a large number of 550 error lines. This bug was coincidentally fixed by
5429 increasing the size of one of Exim's internal buffers (big_buffer) that
5430 happened as part of the Exiscan merge. However, to be on the safe side, I
5431 have made the code more robust (and fixed the comments that describe what
5434 33. Some experimental protocols are using DNS PTR records for new purposes. The
5435 keys for these records are domain names, not reversed IP addresses. The
5436 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5437 leaves it alone. Component reversal etc. now happens only for IP addresses.
5440 34. The host_aton() function is supposed to be passed a string that is known
5441 to be a valid IP address. However, in the case of IPv6 addresses, it was
5442 not checking this. This is a hostage to fortune. Exim now panics and dies
5443 if the condition is not met. A case was found where this could be provoked
5444 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5445 components; fortuitously, this particular loophole had already been fixed
5446 by change 4.50/55 or 4.44/33 above.
5448 If there are any other similar loopholes, the new check in host_aton()
5449 itself should stop them being exploited. The report I received stated that
5450 data on the command line could provoke the exploit when Exim was running as
5451 exim, but did not say which command line option was involved. All I could
5452 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5453 running as the user.
5456 35. There was a buffer overflow vulnerability in the SPA authentication code
5457 (which came originally from the Samba project). I have added a test to the
5458 spa_base64_to_bits() function which I hope fixes it.
5461 36. The daemon start-up calls getloadavg() while still root for those OS that
5462 need the first call to be done as root, but it missed one case: when
5463 deliver_queue_load_max is set with deliver_drop_privilege. This is
5464 necessary for the benefit of the queue runner, because there is no re-exec
5465 when deliver_drop_privilege is set.
5467 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5468 was in use, was not putting the data itself into the right store pool;
5469 consequently, it could be overwritten for a subsequent message in the same
5470 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5473 38. Sometimes the final signoff response after QUIT could fail to get
5474 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5475 < 0 before doing a fflush(). This bug looks as though it goes back to the
5476 introduction of TLS in release 3.20, but "sometimes" must have been rare
5477 because the tests only now provoked it.
5483 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5484 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5485 Now it returns an integer. A number of calls were still expecting a T/F
5486 return. Fortuitously, in all cases, the tests worked in OK situations,
5487 which is the norm. However, things would have gone wrong on any write
5488 failures on the smtp file descriptor. This function is used when sending
5489 messages over SMTP and also when doing verify callouts.
5491 2. When Exim is called to do synchronous delivery of a locally submitted
5492 message (the -odf or -odi options), it no longer closes stderr before doing
5495 3. Implemented the mua_wrapper option.
5497 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5499 5. Implemented the functions header_remove(), header_testname(),
5500 header_add_at_position(), and receive_remove_recipient(), and exported them
5503 6. If an ACL "warn" statement specified the addition of headers, Exim already
5504 inserted X-ACL-Warn: at the start if there was no header name. However, it
5505 was not making this test for the second and subsequent header lines if
5506 there were newlines in the string. This meant that an invalid header could
5507 be inserted if Exim was badly configured.
5509 7. Allow an ACL "warn" statement to add header lines at the start or after all
5510 the Received: headers, as well as at the end.
5512 8. Added the rcpt_4xx retry error code.
5514 9. Added postmaster_mailfrom=xxx to callout verification option.
5516 10. Added mailfrom=xxxx to the callout verification option, for verify=
5519 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5520 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5522 12. Inserted some casts to stop certain compilers warning when using pointer
5523 differences as field lengths or precisions in printf-type calls (mostly
5524 affecting debugging statements).
5526 13. Added optional readline() support for -be (dynamically loaded).
5528 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5529 same clock tick as a message's arrival, so that its received time was the
5530 same as the "first fail" time on the retry record, and that message
5531 remained on the queue past the ultimate address timeout, every queue runner
5532 would try a delivery (because it was past the ultimate address timeout) but
5533 after another failure, the ultimate address timeout, which should have then
5534 bounced the address, did not kick in. This was a "< instead of <=" error;
5535 in most cases the first failure would have been in the next clock tick
5536 after the received time, and all would be well.
5538 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5539 being recognized when the domain list was tested by the match_domain
5540 condition in an expansion string.
5542 16. Added the ${str2b64: operator.
5544 17. Exim was always calling setrlimit() to set a large limit for the number of
5545 processes, without checking whether the existing limit was already
5546 adequate. (It did check for the limit on file descriptors.) Furthermore,
5547 errors from getrlimit() and setrlimit() were being ignored. Now they are
5548 logged to the main and panic logs, but Exim does carry on, to try to do its
5549 job under whatever limits there are.
5551 18. Imported PCRE 5.0.
5553 19. Trivial typo in log message " temporarily refused connection" (the leading
5556 20. If the log selector return_path_on_delivery was set and an address was
5557 redirected to /dev/null, the delivery process crashed because it assumed
5558 that a return path would always be set for a "successful" delivery. In this
5559 case, the whole delivery is bypassed as an optimization, and therefore no
5562 21. Internal re-arrangement: the function for sending a challenge and reading
5563 a response while authentication was assuming a zero-terminated challenge
5564 string. It's now changed to take a pointer and a length, to allow for
5565 binary data in such strings.
5567 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5569 23. Exim was not respecting finduser_retries when seeking the login of the
5570 uid under which it was called; it was always trying 10 times. (The default
5571 setting of finduser_retries is zero.) Also, it was sleeping after the final
5572 failure, which is pointless.
5574 24. Implemented tls_on_connect_ports.
5576 25. Implemented acl_smtp_predata.
5578 26. If the domain in control=submission is set empty, Exim assumes that the
5579 authenticated id is a complete email address when it generates From: or
5580 Sender: header lines.
5582 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5583 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5584 chown and chgrp in /bin and hostname in /usr/bin.
5586 28. Exim was keeping the "process log" file open after each use, just as it
5587 does for the main log. This opens the possibility of it remaining open for
5588 long periods when the USR1 signal hits a daemon. Occasional processlog
5589 errors were reported, that could have been caused by this. Anyway, it seems
5590 much more sensible not to leave this file open at all, so that is what now
5593 29. The long-running daemon process does not normally write to the log once it
5594 has entered its main loop, and it closes the log before doing so. This is
5595 so that log files can straightforwardly be renamed and moved. However,
5596 there are a couple of unusual error situations where the daemon does write
5597 log entries, and I had neglected to close the log afterwards.
5599 30. The text of an SMTP error response that was received during a remote
5600 delivery was being truncated at 512 bytes. This is too short for some of
5601 the long messages that one sometimes sees. I've increased the limit to
5604 31. It is now possible to make retry rules that apply only when a message has a
5605 specific sender, in particular, an empty sender.
5607 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5608 it possible to be selective about when SMTP synchronization is enforced.
5610 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5612 32. Implemented hosts_connection_nolog.
5614 33. Added an ACL for QUIT.
5616 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5619 35. Added mailbox_size and mailbox_filecount to appendfile.
5621 36. Added control = no_multiline_responses to ACLs.
5623 37. There was a bug in the logic of the code that waits for the clock to tick
5624 in the case where the clock went backwards by a substantial amount such
5625 that the microsecond fraction of "now" was more than the microsecond
5626 fraction of "then" (but the whole seconds number was less).
5628 38. Added support for the libradius Radius client library this is found on
5629 FreeBSD (previously only the radiusclient library was supported).
5635 1. When certain lookups returned multiple values in the form name=value, the
5636 quoting of the values was not always being done properly. Specifically:
5637 (a) If the value started with a double quote, but contained no whitespace,
5639 (b) If the value contained whitespace other than a space character (i.e.
5640 tabs or newlines or carriage returns) it was not quoted.
5641 This fix has been applied to the mysql and pgsql lookups by writing a
5642 separate quoting function and calling it from the lookup code. The fix
5643 should probably also be applied to nisplus, ibase and oracle lookups, but
5644 since I cannot test any of those, I have not disturbed their existing code.
5646 2. A hit in the callout cache for a specific address caused a log line with no
5647 reason for rejecting RCPT. Now it says "Previous (cached) callout
5648 verification failure".
5650 3. There was an off-by-one bug in the queryprogram router. An over-long
5651 return line was truncated at 256 instead of 255 characters, thereby
5652 overflowing its buffer with the terminating zero. As well as fixing this, I
5653 have increased the buffer size to 1024 (and made a note to document this).
5655 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5656 when Exim is waiting for an SMTP response from a remote server, Exim
5657 restarts its select() call on the socket, thereby resetting its timeout.
5658 This is not a problem when such interrupts are rare. Somebody set up a cron
5659 job to run exiwhat every 2 minutes, which is less than the normal select()
5660 timeout (5 or 10 minutes). This meant that the select() timeout never
5661 kicked in because it was always reset. I have fixed this by comparing the
5662 time when an interrupt arrives with the time at the start of the first call
5663 to select(). If more time than the timeout has elapsed, the interrupt is
5664 treated as a timeout.
5666 5. Some internal re-factoring in preparation for the addition of Sieve
5667 extensions (by MH). In particular, the "personal" test is moved to a
5668 separate function, and given an option for scanning Cc: and Bcc: (which is
5669 not set for Exim filters).
5671 6. When Exim created an email address using the login of the caller as the
5672 local part (e.g. when creating a From: or Sender: header line), it was not
5673 quoting the local part when it contained special characters such as @.
5675 7. Installed new OpenBSD configuration files.
5677 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5678 try to make them clearer.
5680 9. Callout options, other than the timeout value, were being ignored when
5681 verifying sender addresses in header lines. For example, when using
5683 verify = header_sender/callout=no_cache
5685 the cache was (incorrectly) being used.
5687 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5688 only the Cygwin environment.
5690 11. When return_path_on_delivery was set as a log selector, if different remote
5691 addresses in the same message used different return paths and parallel
5692 remote delivery occurred, the wrong values would sometimes be logged.
5693 (Whenever a remote delivery process finished, the return path value from
5694 the most recently started remote delivery process was logged.)
5696 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5697 lines when AUTH and/or TLS are in use. This is the "received protocol"
5698 field. Exim used to use "asmtp" for authenticated SMTP, without any
5699 indication (in the protocol name) for TLS use. Now it follows the RFC and
5700 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5701 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5702 names appear in log lines as well as in Received: header lines.
5704 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5705 extensions, and comparison tests, and to fix some bugs.
5707 14. Changes to the "personal" filter test:
5709 (1) The test was buggy in that it was just doing the equivalent of
5710 "contains" tests on header lines. For example, if a user's address was
5711 anne@some.where, the "personal" test would incorrectly be true for
5713 To: susanne@some.where
5715 This test is now done by extracting each address from the header in turn,
5716 and checking the entire address. Other tests that are part of "personal"
5717 are now done using regular expressions (for example, to check local parts
5718 of addresses in From: header lines).
5720 (2) The list of non-personal local parts in From: addresses has been
5721 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5722 taken from the Sieve specification recommendations.
5724 (3) If the message contains any header line starting with "List-" it is
5725 treated as non-personal.
5727 (4) The test for "circular" in the Subject: header line has been removed
5728 because it now seems ill-conceived.
5730 15. Minor typos in src/EDITME comments corrected.
5732 16. Installed latest exipick from John Jetmore.
5734 17. If headers_add on a router specified a text string that was too long for
5735 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5736 of string_sprintf() is now avoided.
5738 18. $message_body_size was not set (it was always zero) when running the DATA
5739 ACL and the local_scan() function.
5741 19. For the "mail" command in an Exim filter, no default was being set for
5742 the once_repeat time, causing a random time value to be used if "once" was
5743 specified. (If the value happened to be <= 0, no repeat happened.) The
5744 default is now 0s, meaning "never repeat". The "vacation" command was OK
5745 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5746 (I found it when inspecting the code).
5748 20. There is now an overall timeout for performing a callout verification. It
5749 defaults to 4 times the callout timeout, which applies to individual SMTP
5750 commands during the callout. The overall timeout applies when there is more
5751 than one host that can be tried. The timeout is checked before trying the
5752 next host. This prevents very long delays if there are a large number of
5753 hosts and all are timing out (e.g. when the network connections are timing
5754 out). The value of the overall timeout can be changed by specifying an
5755 additional sub-option for "callout", called "maxwait". For example:
5757 verify = sender/callout=5s,maxwait=20s
5759 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5760 to the end before writing, but this should make it even safer).
5762 22. Exim was forgetting that it had advertised PIPELINING for the second and
5763 subsequent messages on an SMTP connection. It was also not resetting its
5764 memory on STARTTLS and an internal HELO.
5766 23. When Exim logs an SMTP synchronization error within a session, it now
5767 records whether PIPELINING has been advertised or not.
5769 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5770 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5771 rather than long int.
5773 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5775 26. Added the never_mail option to autoreply.
5781 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5782 crash if the getsockname() call failed; this can happen if a connection is
5783 closed very soon after it is established. The problem was simply in the
5784 order in which certain operations were done, causing Exim to try to write
5785 to the SMTP stream before it had set up the file descriptor. The bug has
5786 been fixed by making things happen in the correct order.
5792 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5793 before the connection was closed, thus losing the rejection response.
5795 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5796 some early Solaris releases, but causes trouble in current releases where
5797 socklen_t is defined.
5799 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5802 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5805 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5806 that releasing the top of it at the end releases what was used for sub-
5807 expansions (unless the block got too big). However, discard this block if
5808 the first thing is a variable or header, so that we can use its block when
5809 it is dynamic (useful for very large $message_headers, for example).
5811 6. Lookups now cache *every* query, not just the most recent. A new, separate
5812 store pool is used for this. It can be recovered when all lookup caches are
5813 flushed. Lookups now release memory at the end of their result strings.
5814 This has involved some general refactoring of the lookup sources.
5816 7. Some code has been added to the store_xxx() functions to reduce the amount
5817 of flapping under certain conditions.
5819 8. log_incoming_interface used to affect only the <= reception log lines. Now
5820 it causes the local interface and port to be added to several more SMTP log
5821 lines, for example "SMTP connection from", and rejection lines.
5823 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5825 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5827 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5828 could be overwritten at the end of the current message (or the start of a
5829 new message if it was set in a HELO ACL). The value is now preserved for
5830 the duration of the SMTP connection.
5832 12. If a transport had a headers_rewrite setting, and a matching header line
5833 contained an unqualified address, that address was qualified, even if it
5834 did not match any rewriting rules. The underlying bug was that the values
5835 of the flags that permit the existence of unqualified sender and recipient
5836 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5837 non-local messages, and by -bnq for local messages) were not being
5838 preserved with the message after it was received.
5840 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5841 "next input=" as part of the text comprising the host identity instead of
5842 the correct text. The code was using the same buffer for two different
5843 strings. However, depending on which order the printing function evaluated
5844 its arguments, the bug did not always show up. Under Linux, for example, my
5845 test suite worked just fine.
5847 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5848 doesn't work with some older versions of Perl. It has been changed to "my",
5849 which in any case is probably the better facility to use.
5851 15. A really picky compiler found some instances of statements for creating
5852 error messages that either had too many or two few arguments for the format
5855 16. The size of the buffer for calls to the DNS resolver has been increased
5856 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5857 for addresses that have a lot of PTR records. This alleviates a problem; it
5858 does not fully solve it.
5860 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5861 buffer now truncates the list and logs the incident, which is the same
5862 action as happens when Exim is looking up a host name and its aliases.
5863 Previously in this situation something unpredictable would happen;
5864 sometimes it was "internal error: store_reset failed".
5866 18. If a server dropped the connection unexpectedly when an Exim client was
5867 using GnuTLS and trying to read a response, the client delivery process
5868 crashed while trying to generate an error log message.
5870 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5871 string, for example:
5873 warn message = H1: something\nH2: something
5875 the text was added as a single header line from Exim's point of view
5876 though it ended up OK in the delivered message. However, searching for the
5877 second and subsequent header lines using $h_h2: did not work. This has been
5878 fixed. Similarly, if a system filter added multiple headers in this way,
5879 the routers could not see them.
5881 20. Expanded the error message when iplsearch is called with an invalid key to
5882 suggest using net-iplsearch in a host list.
5884 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5885 ACLs are no longer actually imposed (and a message to that effect is
5888 22. If a "gecos" field in a passwd entry contained escaped characters, in
5889 particular, if it contained a \" sequence, Exim got it wrong when building
5890 a From: or a Sender: header from that name. A second bug also caused
5891 incorrect handling when an unquoted " was present following a character
5892 that needed quoting.
5894 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5895 was not being matched caselessly.
5897 24. Arranged for all hyphens in the exim.8 source to be escaped with
5900 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5901 properly. Recipient callout cache records were still being keyed to include
5902 the sender, even when use_sender was set false. This led to far more
5903 callouts that were necessary. The sender is no longer included in the key
5904 when use_sender is false.
5906 26. Added "control = submission" modifier to ACLs.
5908 27. Added the ${base62d: operator to decode base 62 numbers.
5910 28. dnsdb lookups can now access SRV records.
5912 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5913 the configuration file.
5915 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5916 (-v) mode. This makes the output for a verbose queue run more intelligible.
5918 31. Added a use_postmaster feature to recipient callouts.
5920 32. Added the $body_zerocount variable, containing the number of binary zero
5921 bytes in the message body.
5923 33. The time of last modification of the "new" subdirectory is now used as the
5924 "mailbox time last read" when there is a quota error for a maildir
5927 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5929 35. Added +ignore_unknown as a special item in host lists.
5931 36. Code for decoding IPv6 addresses in host lists is now included, even if
5932 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5933 address was recognized as an IP address, but was then not correctly decoded
5934 into binary, causing unexpected and incorrect effects when compared with
5941 1. Very minor rewording of debugging text in manualroute to say "list of
5942 hosts" instead of "hostlist".
5944 2. If verify=header_syntax was set, and a header line with an unqualified
5945 address (no domain) and a large number of spaces between the end of the
5946 name and the colon was received, the reception process suffered a buffer
5947 overflow, and (when I tested it) crashed. This was caused by some obsolete
5948 code that should have been removed. The fix is to remove it!
5950 3. When running in the test harness, delay a bit after writing a bounce
5951 message to get a bit more predictability in the log output.
5953 4. Added a call to search_tidyup() just before forking a reception process. In
5954 theory, someone could use a lookup in the expansion of smtp_accept_max_
5955 per_host which, without the tidyup, could leave open a database connection.
5957 5. Added the variables $recipient_data and $sender_data which get set from a
5958 lookup success in an ACL "recipients" or "senders" condition, or a router
5959 "senders" option, similar to $domain_data and $local_part_data.
5961 6. Moved the writing of debug_print from before to after the "senders" test
5964 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5965 problems for message scanning, either using a data ACL, or using
5966 local_scan() because the Received: header was not generated till after they
5967 were called (in order to set the time as the time of reception completion).
5968 I have revised the way this works. The header is now generated after the
5969 body is received, but before the ACL or local_scan() are called. After they
5970 are run, the timestamp in the header is updated.
5976 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5977 before starting a queue runner without re-exec. This happened only when
5978 deliver_drop_privilege was set or when the Exim user was set to root. The
5979 effect of the bug was that timeouts during subsequent deliveries caused
5980 crashes instead of being properly handled. The handler is now left at its
5981 default (and expected) setting.
5983 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5984 message, again when deliver_drop_privilege is set or Exim is run as root.
5985 The bug described in (1) was not present in this case, but the tidying up
5986 of the other signals was missing. I have made the two cases consistent.
5988 3. The ignore_target_hosts setting on a manualroute router was being ignored
5989 for hosts that were looked up using the /MX notation.
5991 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5994 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5995 operated on the sender address. After changing the $sender_address to <>
5996 for the sender address verify, Exim was re-instated it as the original
5997 (before rewriting) address, but remembering that it had rewritten it, so it
5998 wasn't rewriting it again. This bug also had the effect of breaking the
5999 sender address verification caching when the sender address was rewritten.
6001 6. The ignore_target_hosts option was being ignored by the ipliteral router.
6002 This has been changed so that if the ip literal address matches
6003 ignore_target_hosts, the router declines.
6005 7. Added expansion conditions match_domain, match_address, and match_local_
6006 part (NOT match_host).
6008 8. The placeholder for the Received: header didn't have a length field set.
6010 9. Added code to Exim itself and to exim_lock to test for a specific race
6011 condition that could lead to file corruption when using MBX delivery. The
6012 issue is with the lockfile that is created in /tmp. If this file is removed
6013 after a process has opened it but before that process has acquired a lock,
6014 there is the potential for a second process to recreate the file and also
6015 acquire a lock. This could lead to two Exim processes writing to the file
6016 at the same time. The added code performs the same test as UW imapd; it
6017 checks after acquiring the lock that its file descriptor still refers to
6018 the same named file.
6020 10. The buffer for building added header lines was of fixed size, 8192 bytes.
6021 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
6024 11. Added the smtp_active_hostname option. If used, this will typically be made
6025 to depend on the incoming interface address. Because $interface_address is
6026 not set up until the daemon has forked a reception process, error responses
6027 that can happen earlier (such as "too many connections") no longer contain
6030 12. If an expansion in a condition on a "warn" statement fails because a lookup
6031 defers, the "warn" statement is abandoned, and the next ACL statement is
6032 processed. Previously this caused the whole ACL to be aborted.
6034 13. Added the iplsearch lookup type.
6036 14. Added ident_timeout as a log selector.
6038 15. Added tls_certificate_verified as a log selector.
6040 16. Added a global option tls_require_ciphers (compare the smtp transport
6041 option of the same name). This controls incoming TLS connections.
6043 17. I finally figured out how to make tls_require_ciphers do a similar thing
6044 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
6045 before starting the TLS session.
6047 18. Tabs are now shown as \t in -bP output.
6049 19. If the log selector return_path_on_delivery was set, Exim crashed when
6050 bouncing a message because it had too many Received: header lines.
6052 20. If two routers both had headers_remove settings, and the first one included
6053 a superfluous trailing colon, the final name in the first list and the
6054 first name in the second list were incorrectly joined into one item (with a
6055 colon in the middle).
6061 1. Added -C and -D options to the exinext utility, mainly to make it easier
6062 to include in the automated testing, but these could be helpful when
6063 multiple configurations are in use.
6065 2. The exinext utility was not formatting the output nicely when there was
6066 an alternate port involved in the retry record key, nor when there was a
6067 message id as well (for retries that were specific to a specific message
6068 and a specific host). It was also confused by IPv6 addresses, because of
6069 the additional colons they contain. I have fixed the IPv4 problem, and
6070 patched it up to do a reasonable job for IPv6.
6072 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
6073 delivery, the log line now contains "pipelined" if PIPELINING was used.
6075 4. An SMTP transport process used to panic and die if the bind() call to set
6076 an explicit outgoing interface failed. This has been changed; it is now
6077 treated in the same way as a connect() failure.
6079 5. A reference to $sender_host_name in the part of a conditional expansion
6080 that was being skipped was still causing a DNS lookup. This no longer
6083 6. The def: expansion condition was not recognizing references to header lines
6084 that used bh_ and bheader_.
6086 7. Added the _cache feature to named lists.
6088 8. The code for checking quota_filecount in the appendfile transport was
6089 allowing one more file than it should have been.
6091 9. For compatibility with Sendmail, the command line option
6099 and sets the incoming protocol and host name (for trusted callers). The
6100 host name and its colon can be omitted when only the protocol is to be set.
6101 Note the Exim already has two private options, -pd and -ps, that refer to
6102 embedded Perl. It is therefore impossible to set a protocol value of "d" or
6103 "s", but I don't think that's a major issue.
6105 10. A number of refactoring changes to the code, none of which should affect
6108 (a) The number of logging options was getting close to filling up the
6109 32-bit word that was used as a bit map. I have split them into two classes:
6110 those that are passed in the argument to log_write(), and those that are
6111 only ever tested independently outside of that function. These are now in
6112 separate 32-bit words, so there is plenty of room for expansion again.
6113 There is no change in the user interface or the logging behaviour.
6115 (b) When building, for example, log lines, the code previously used a
6116 macro that called string_cat() twice, in order to add two strings. This is
6117 not really sufficiently general. Furthermore, there was one instance where
6118 it was actually wrong because one of the argument was used twice, and in
6119 one call a function was used. (As it happened, calling the function twice
6120 did not affect the overall behaviour.) The macro has been replaced by a
6121 function that can join an arbitrary number of extra strings onto a growing
6124 (c) The code for expansion conditions now uses a table and a binary chop
6125 instead of a serial search (which was left over from when there were very
6126 few conditions). Also, it now recognizes conditions like "pam" even when
6127 the relevant support is not compiled in: a suitably worded error message is
6128 given if an attempt is made to use such a condition.
6130 11. Added ${time_interval:xxxxx}.
6132 12. A bug was causing one of the ddress fields not to be passed back correctly
6133 from remote delivery subprocesses. The field in question was not being
6134 subsequently used, so this caused to problems in practice.
6136 13. Added new log selectors queue_time and deliver_time.
6138 14. Might have fixed a bug in maildirsizefile handling that threw up
6139 "unexpected character" debug warnings, and recalculated the data
6140 unnecessarily. In any case, I expanded the warning message to give more
6143 15. Added the message "Restricted characters in address" to the statements in
6144 the default ACL that block characters like @ and % in local parts.
6146 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
6147 Three changes have been made:
6149 (a) There was a serious bug; a negative response to MAIL caused the whole
6150 recipient domain to be cached as invalid, thereby blocking all messages
6151 to all local parts at the same domain, from all senders. This bug has
6152 been fixed. The domain is no longer cached after a negative response to
6153 MAIL if the sender used is not empty.
6155 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
6158 (c) A new callout option, "use_sender" has been added for people who want
6159 the modified behaviour.
6165 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
6168 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
6169 indeed breaks things for older releases.
6171 3. Added additional logging to the case where there is a problem reading data
6172 from a filter that is running in a subprocess using a pipe, in order to
6173 try to track down a specific problem.
6175 4. Testing facility fudge: when running in the test harness and attempting
6176 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
6177 getting "No route to host". Convert this to a timeout.
6179 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
6182 6. Some OS don't have socklen_t but use size_t instead. This affects the
6183 fifth argument of getsockopt() amongst other things. This is now
6184 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
6185 can be set for individual OS. I have set it for SunOS5, OSF1, and
6186 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
6187 some earlier ones do not.
6189 7. Change 4.30/15 was not doing the test caselessly.
6191 8. The standard form for an IPv6 address literal was being rejected by address
6192 parsing in, for example, MAIL and RCPT commands. An example of this kind of
6193 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6194 this, as well as the form without the "IPv6" on the front (but only when
6195 address literals are enabled, of course).
6197 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
6199 10. Exim crashed if a message with an empty sender address specified by -f
6200 encountered a router with an errors_to setting. This could be provoked only
6201 by a command such as
6205 where an empty string was supplied; "<>" did not hit this bug.
6207 11. Installed PCRE release 4.5.
6209 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
6210 remained set. It is now erased.
6212 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
6213 times from message ids (which are base 36 rather than the normal 62).
6215 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
6216 were being counted as actual protocol errors, and logged if the log
6217 selector +smtp_protocol_error was set. One cannot be perfect in this test,
6218 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
6219 and DATA following a set of rejected RCPTs do not count as protocol errors.
6220 In other words, Exim assumes they were pipelined, though this may not
6221 actually be the case. Of course, in all cases the client gets an
6222 appropriate error code.
6224 15. If a lookup fails in an ACL condition, a message about the failure may
6225 be available; it is used if testing the ACL cannot continue, because most
6226 such messages specify what the cause of the deferral is. However, some
6227 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
6228 that caused an old message to be retained and used if a later statement
6229 caused a defer, replacing the real cause of the deferral.
6231 16. If an IP address had so many PTR records that the DNS lookup buffer
6232 was not large enough to hold them, Exim could crash while trying to process
6233 the truncated data. It now detects and logs this case.
6235 17. Further to 4.21/58, another change has been made: if (and only if) the
6236 first line of a message (the first header line) ends with CRLF, a bare LF
6237 in a subsequent header line has a space inserted after it, so as not to
6238 terminate the header.
6240 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
6241 unnecessarily, used atoi() instead of strtol(), and didn't check the
6242 termination when getting file sizes from file names by regex.
6244 19. Completely re-implemented the support for maildirsize files, in the light
6245 of a number of problems with the previous contributed implementation
6246 (4.30/29). In particular:
6248 . If the quota is zero, the maildirsize file is maintained, but no quota is
6251 . If the maildir directory does not exist, it is created before any attempt
6252 to write a maildirsize file.
6254 . The quota value in the file is just a cache; if the quota is changed in
6255 the transport, the new value overrides.
6257 . A regular expression is available for excluding directories from the
6260 20. The autoreply transport checks the characters in options that define the
6261 message's headers; it allows continued headers, but it was checking with
6262 isspace() after an embedded newline instead of explicitly looking for a
6265 21. If all the "regular" hosts to which an address was routed had passed their
6266 expiry times, and had not reached their retry times, the address was
6267 bounced, even if fallback hosts were defined. Now Exim should go on to try
6270 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
6271 equivalent code in the SMTP transport. Some hosts send humungous responses
6272 to HELO/EHLO, more than 1024 it seems.
6274 23. Refactoring: code in filter.c used (void *) for "any old type" but this
6275 gives compiler warnings in some environments. I've now done it "properly",
6278 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
6279 (because of problems with the built-in one) was declared to return uschar *
6280 instead of char *, causing compiler failure.
6282 25. Fixed a file descriptor leak when processing alias/forward files.
6284 26. Fixed a minor format string issue in dbfn.c.
6286 27. Typo in exim.c: ("dmbnz" for "dbmnz").
6288 28. If a filter file refered to $h_xxx or $message_headers, and the headers
6289 contained RFC 2047 "words", Exim's memory could, under certain conditions,
6292 29. When a sender address is verified, it is cached, to save repeating the test
6293 when there is more than one recipient in a message. However, when the
6294 verification involves a callout, it is possible for different callout
6295 options to be set for different recipients. It is too complicated to keep
6296 track of this in the cache, so now Exim always runs a verification when a
6297 callout is required, relying on the callout cache for the optimization.
6298 The overhead is duplication of the address routing, but this should not be
6301 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
6302 to be verified with callout=postmaster, and the main callout worked but the
6303 postmaster check failed, the verification correctly failed. However, if a
6304 subsequent RCPT command asked for sender verification *without* the
6305 postmaster check, incorrect caching caused this verification also to fail,
6308 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
6309 it was not caching the DNS options (qualify_single, search_parents) that
6310 were used when the lookup failed. A subsequent lookup with different
6311 options therefore always gave the same answer, though there were cases
6312 where it should not have. (Example: a "domains = !$mx_any" option on a
6313 dnslookup router: the "domains" option is always processed without any
6314 widening, but the router might have qualify_single set.) Now Exim uses the
6315 cached value only when the same options are set.
6317 32. Added John Jetmore's "exipick" utility to the distribution.
6319 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
6320 than a timeout (e.g. a certificate is required, and is not provided), an
6321 Exim server now closes the connection immediately. Previously it waited for
6322 the client to close - but if the client is SSL, it seems that they each
6323 wait for each other, leading to a delay before one of them times out.
6325 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
6326 maintained 0.8.x compatibility because I don't think many are using it, and
6327 it is clearly obsolete.
6329 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
6332 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
6333 client certificate was expired. A simple patch fixes this, though I don't
6334 understand the full logic of why the verify callback is called multiple
6337 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
6338 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
6339 which causes problems with some clients (such as the Certicom SSL Plus
6340 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
6341 disables the coutermeasure allowing Eudora to connect."
6343 38. Exim was not checking that a write() to a log file succeeded. This could
6344 lead to Bad Things if a log got too big, in particular if it hit a file
6345 size limit. Exim now panics and dies if it cannot write to a log file, just
6346 as it does if it cannot open a log file.
6348 39. Modified OS/Makefile-Linux so that it now contains
6350 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
6352 The two -D definitions ensure that Exim is compiled with large file
6353 support, which makes it possible to handle log files that are bigger than
6356 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
6357 instance) a domain was checked against a named list that involved a lookup,
6358 causing $domain_data to be set, then another domain was checked against the
6359 same list, then the first domain was re-checked, the value of $domain_data
6360 after the final check could be wrong. In particular, if the second check
6361 failed, it could be set empty. This bug probably also applied to
6364 41. The strip_trailing_dot option was not being applied to the address given
6365 with the -f command-line option.
6367 42. The code for reading a message's header from the spool was incrementing
6368 $received_count, but never initializing it. This meant that the value was
6369 incorrect (doubled) while delivering a message in the same process in which
6370 it was received. In the most common configuration of Exim, this never
6371 happens - a fresh exec is done - but it can happen when
6372 deliver_drop_privilege is set.
6374 43. When Exim logs an SMTP synchronization error - client data sent too soon -
6375 it now includes up to 150 characters of the unexpected data in the log
6378 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6379 and building data strings. The size of both of these buffers was 10 000
6380 bytes - far larger than anybody would *ever* want, thought I. Needless to
6381 say, somebody hit the limit. I have increased the maximum line length to
6382 20 000 and the maximum data length of concatenated lines to 100 000. I have
6383 also fixed two bugs, because there was no checking on these buffers. Tsk,
6384 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6385 buffer is too small.
6387 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6388 lsearch lookups. Now it does.
6390 46. When parsing a route_list item in a manualroute router, a fixed-length
6391 buffer was used for the list of hosts. I made this 1024 bytes long,
6392 thinking that nobody would ever have a list of hosts that long. Wrong.
6393 Somebody had a whole pile of complicated expansion conditions, and the
6394 string was silently truncated, leading to an expansion error. It turns out
6395 that it is easier to change to an unlimited length (owing to other changes
6396 that have happened since this code was originally written) than to build
6397 structure for giving a limitation error. The length of the item that
6398 expands into the list of hosts is now unlimited.
6400 47. The lsearch lookup could not handle data where the length of text line was
6401 more than 4095 characters. Such lines were truncated, leading to shortened
6402 data being returned. It should now handle lines of any length.
6404 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6405 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6408 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6409 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6410 no longer happens in comment lines. A list of replacements is now placed
6411 at the head of all of the source files, except those whose only change is
6412 to replace PERL_COMMAND in the very first #! line.
6414 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6415 messages on the queue, with a bottom-up merge sort, using code contributed
6416 by Michael Haardt. This should make operations like -bp somewhat faster on
6417 large queues. It won't affect queue runners, except when queue_run_in_order
6420 51. Installed eximstats 1.31 in the distribution.
6422 52. Added support for SRV lookups to the dnslookup router.
6424 53. If an ACL referred to $message_body or $message_body_end, the value was not
6425 reset for any messages that followed in the same SMTP session.
6427 54. The store-handling optimization for building very long strings was not
6428 differentiating between the different store pools. I don't think this
6429 actually made any difference in practice, but I've tidied it.
6431 55. While running the routers to verify a sender address, $sender_address
6432 was still set to the sender address. This is wrong, because when routing to
6433 send a bounce to the sender, it would be empty. Therefore, I have changed
6434 it so that, while verifying a sender address, $sender_address is set to <>.
6435 (There is no change to what happens when verifying a recipient address.)
6437 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6438 target A or AAAA records (if not already returned) without resetting the
6439 qualify_single or search_parents options of the DNS resolver. These are
6440 inappropriate in this case because the targets of MX and SRV records must
6441 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6442 target that, when qualified, matched something in the local domain. These
6443 two options are now turned off when doing these lookups.
6445 57. It seems that at least some releases of Reiserfs (which does not have the
6446 concept of a fixed number of inodes) returns zero and not -1 for the
6447 number of available inodes. This interacted badly with check_spool_inodes,
6448 which assumed that -1 was the "no such thing" setting. What I have done is
6449 to check that the total number of inodes is greater than zero before doing
6450 the test of how many are available.
6452 58. When a "warn" ACL statement has a log_message modifier, the message is
6453 remembered, and not repeated. This is to avoid a lot of repetition when a
6454 message has many recipients that cause the same warning to be written.
6455 However, Exim was preserving the list of already written lines for an
6456 entire SMTP session, which doesn't seem right. The memory is now reset if a
6457 new message is started.
6459 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6460 debugging output unless log_rewrite was also set.
6462 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6463 of (int)(handle) when we know that handle contains (void *)(-1).
6465 61. The Exim daemon panic-logs an error return when it closes the incoming
6466 connection. However "connection reset by peer" seems to be common, and
6467 isn't really an error worthy of noting specially, so that particular error
6470 62. When Exim is trying to find all the local interfaces, it used to panic and
6471 die if the ioctl to get the interface flags failed. However, it seems that
6472 on at least one OS (Solaris 9) it is possible to have an interface that is
6473 included in the list of interfaces, but for which you get a failure error
6474 for this call. This happens when the interface is not "plumbed" into a
6475 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6476 failure of the "get flags" call assumes that the interface is down.
6478 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6479 makes life easier for people who are doing arithmetic on fields extracted
6480 from dates, where you often get leading zeros that should not be
6481 interpreted as octal.
6483 64. Added qualify_domain to the redirect router, to override the global
6486 65. If a pathologically long header line contained very many addresses (the
6487 report of this problem mentioned 10 000) and each of them was rewritten,
6488 Exim could use up a very large amount of memory. (It kept on making new
6489 copies of the header line as it rewrote, and never released the old ones.)
6490 At the expense of a bit more processing, the header rewriting function has
6491 been changed so that it no longer eats memory in this way.
6493 66. The generation of the Received: header has been moved from the time that a
6494 message starts to be received, to the time that it finishes. The timestamp
6495 in the Received: header should now be very close to that of the <= log
6496 line. There are two side-effects of this change:
6498 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6499 logged header lines no longer include the local Received: line, because
6500 it has not yet been created. The same applies to a copy of the message
6501 that is returned to a non-SMTP sender when a message is rejected.
6503 (b) When a filter file is tested using -bf, no additional Received: header
6504 is added to the test message. After some thought, I decided that this
6507 This change does not affect the value of $received_for. It is still set
6508 after address rewriting, but before local_scan() is called.
6510 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6512 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6513 gave an unhelpful panic error message, and a defer error. I have managed to
6514 change this behaviour so that it now rejects any supplied certificate,
6515 which seems right, as the list of acceptable certificates is empty.
6517 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6518 gave an unhelpful defer error. I have not managed to make this reject any
6519 supplied certificates, but the error message it gives is "no certificate
6520 supplied", which is not helpful.
6522 70. exigrep's output now also includes lines that are not associated with any
6523 message, but which match the given pattern. Implemented by a patch from
6524 Martin Sluka, which also tidied up the Perl a bit.
6526 71. Recipient callout verification, like sender verification, was using <> in
6527 the MAIL FROM command. This isn't really the right thing, since the actual
6528 sender may affect whether the remote host accepts the recipient or not. I
6529 have changed it to use the actual sender in the callout; this means that
6530 the cache record is now keyed on a recipient/sender pair, not just the
6531 recipient address. There doesn't seem to be a real danger of callout loops,
6532 since a callout by the remote host to check the sender would use <>.
6533 [SEE ABOVE: changed after hitting problems.]
6535 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6536 temporary errors. However, in the case of such a code being given after
6537 the end of a data transmission (i.e. after ".") Exim was failing to write
6538 a retry record for the message. (Yes, there was some broken host that was
6539 actually sending 8xx at this point.)
6541 73. An unknown lookup type in a host list could cause Exim to panic-die when
6542 the list was checked. (An example that provoked this was putting <; in the
6543 middle of a list instead of at the start.) If this happened during a DATA
6544 ACL check, a -D file could be left lying around. This kind of configuration
6545 error no longer causes Exim to die; instead it causes a defer error. The
6546 incident is still logged to the main and panic logs.
6548 74. Buglet left over from Exim 3 conversion. The message "too many messages
6549 in one connection" was written to the rejectlog but not the mainlog, except
6550 when address rewriting (yes!) was being logged.
6552 75. Added write_rejectlog option.
6554 76. When a system filter was run not as root (that is, when system_filter_user
6555 was set), the values of the $n variables were not being returned to the
6556 main process; thus, they were not subsequently available in the $sn
6559 77. Added +return_path_on_delivery log selector.
6561 78. A connection timeout was being treated differently from recipients deferred
6562 when testing hosts_max_try with a message that was older than the host's
6563 retry timeout. (The host should not be counted, thus allowing all hosts to
6564 be tried at least once before bouncing.) This may have been the cause of an
6565 occasionally reported bug whereby a message would remain on the queue
6566 longer than the retry timeout, but would be bounced if a delivery was
6567 forced. I say "may" because I never totally pinned down the problem;
6568 setting up timeout/retry tests is difficult. See also the next item.
6570 79. The ultimate address timeout was not being applied to errors that involved
6571 a combination of host plus message (for example, a timeout on a MAIL
6572 command). When an address resolved to a number of possible hosts, and they
6573 were not all tried for each delivery (e.g. because of hosts_max_try), a
6574 message could remain on the queue longer than the retry timeout.
6576 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6579 81. Fixed an obscure SMTP outgoing bug which required at least the following
6580 conditions: (a) there was another message waiting for the same server;
6581 (b) the server returned 5xx to all RCPT commands in the first message so
6582 that the message was not completed; (c) the server dropped the connection
6583 or gave a negative response to the RSET that Exim sends to abort the
6584 transaction. The observed case was a dropped connection after DATA that had
6585 been sent in pipelining mode. That is, the server had advertised PIPELINING
6586 but was not implementing it correctly. The effect of the bug was incorrect
6587 behaviour, such as trying another host, and this could lead to a crash.
6593 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6594 and daemon.c were passed as pointers to ints; they should have been
6595 pointers to socklen_t variables (which are typically unsigned ints).
6597 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6600 3. Fixed a really odd bug that affected only the testing scheme; patching a
6601 certain fixed string in the binary changed the value of another string that
6602 happened to be identical to the end of the original first string.
6604 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6605 name", it returns that address as the IP address. On some operating
6606 systems (e.g. Solaris), it also passes back the IP address string as the
6607 "host name". However, on others (e.g. Linux), it passes back an empty
6608 string. Exim wasn't checking for this, and was changing the host name to an
6609 empty string, assuming it had been canonicalized.
6611 5. Although rare, it is permitted to have more than one PTR record for a given
6612 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6613 all the names associated with an address, because they do in Solaris.
6614 However, it seems that they do not in Linux for data that comes from the
6615 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6616 I found this out when I moved to a new Linux workstation and tried to run
6617 the Exim test suite.
6619 To get round this problem I have changed the code so that it now does its
6620 own call to the DNS to look up PTR records when searching for a host name.
6621 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6622 addresses that are only in /etc/hosts are still found.
6624 This behaviour is, however, controlled by an option called host_lookup_
6625 order, which defaults to "bydns:byaddr". If people want to use the other
6626 order, or indeed, just use one or the other means of lookup, they can
6627 specify it in this variable.
6629 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6630 some operating systems, this comes back from gethostbyaddr() as an empty
6631 string, and this is what Exim used to test for. However, it seems that in
6632 other systems, "." is yielded. Exim now tests for this case too.
6634 7. The values of check_spool_space and check_log_space are now held internally
6635 as a number of kilobytes instead of an absolute number of bytes. If a
6636 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6637 kilobyte. This means that much larger values can be stored.
6639 8. Exim monitor: an attempt to get the action menu when not actually pointing
6640 at a message produces an empty menu entitled "No message selected". This
6641 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6642 no entries in it ("Shell widget menu has zero width and/or height"). So I
6643 have added a single, blank menu entry in this case.
6645 9. Added ${quote_local_part.
6647 10. MIME decoding is now applied to the contents of Subject: header lines when
6650 11. Now that a reference to $sender_host_address automatically causes a reverse
6651 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6652 host lookup before query-style lookups in lists that might use this
6653 variable. This has therefore been abolished, and the "net-" prefix is no
6654 longer necessary for query-style lookups.
6656 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6657 have been a typo for LFLAGS, so it has been changed.
6659 13. The install script calls Exim with "-C /dev/null" in order to find the
6660 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6661 to be output. However, since Exim outputs its version number before the
6662 error, it didn't break the script. It just looked ugly. I fixed this by
6663 always allowing "-C /dev/null" if the caller is root.
6665 14. Ignore overlarge ACL variable number when reading spool file - insurance
6666 against a later release with more variables having written the file.
6668 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6669 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6670 this, as well as the form without the "IPv6" on the front.
6672 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6673 OS/Makefile-Darwin file.
6675 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6676 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6678 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6679 variable. However, some OS use ints for inodes. I've added cast to long int
6680 to get rid of the compiler warning.
6682 19. I had forgotten to lock out "/../" in configuration file names when
6683 ALT_CONFIG_PREFIX was set.
6685 20. Routers used for verification do not need to specify transports. However,
6686 if such a router generated a host list, and callout was configured, Exim
6687 crashed, because it could not find a port number from the (non-existent)
6688 transport. It now assumes port 25 in this circumstance.
6690 21. Added the -t option to exigrep.
6692 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6693 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6694 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6695 them is set, LOOKUP_LSEARCH is forced.
6697 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6699 24. Added sender and host information to the "rejected by local_scan()" log
6700 line; previously there was no indication of these.
6702 25. Added .include_if_exists.
6704 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6705 Linux. It turns out that not all file systems support this. Apparently some
6706 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6707 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6708 not supported on the file descriptor, is now ignored when Exim is trying to
6709 sync a directory. This applies only to Linux.
6711 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6713 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6714 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6717 29. Added support for maildirsize files from supplied patch (modified a bit).
6719 30. The use of :fail: followed by an empty string could lead Exim to respond to
6720 sender verification failures with (e.g.):
6722 550 Verification failed for <xxx>
6723 550 Sender verify failed
6725 where the first response line was missing the '-' that indicates it is not
6726 the final line of the response.
6728 31. The loop for finding the name of the user that called Exim had a hardwired
6729 limit of 10; it now uses the value of finduser_retries, which is used for
6730 all other user lookups.
6732 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6735 33. Exim was neglecting to zero errno before one call of strtol() when
6736 expanding a string and expecting an integer value. On some systems this
6737 resulted in spurious "integer overflow" errors. Also, it was casting the
6738 result into an int without checking.
6740 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6741 did not work. The code looks as if it has *never* worked, though it appears
6742 to have been documented since at least release 1.62. I have made it work.
6744 35. The "timeout_DNS" error in retry rules, also documented since at least
6745 1.62, also never worked. As it isn't clear exactly what this means, and
6746 clearly it isn't a major issue, I have abolished the feature by treating it
6747 as "timeout", and writing a warning to the main and panic logs.
6749 36. The display of retry rules for -brt wasn't always showing the error code
6752 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6753 timeout_connect_A, timeout_connect_MX.
6755 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6756 to the empty sender.
6758 39. The daemon was not analysing the content of -oX till after it had closed
6759 stderr and disconnected from the controlling terminal. This meant that any
6760 syntax errors were only noted on the panic log, and the return code from
6761 the command was 0. By re-arranging the code a little, I've made the
6762 decoding happen first, so such errors now appear on stderr, and the return
6763 code is 1. However, the actual setting up of the sockets still happens in
6764 the disconnected process, so errors there are still only recorded on the
6767 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6768 connections (as happens on some IP stacks) was logged at start up time as
6769 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6770 it from "IPv6 and IPv4", which means that two separate sockets are being
6773 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6774 says whether AF_INET or AF_INET6 was passed as an argument.
6776 42. Exiwhat output was messed up when time zones were included in log
6779 43. Exiwhat now gives more information about the daemon's listening ports,
6780 and whether -tls-on-connect was used.
6782 44. The "port" option of the smtp transport is now expanded.
6784 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6785 silently ignored. Now an error message is written to the main and panic
6788 46. There's a new ACL modifier called "logwrite" which writes to a log file
6789 as soon as it is encountered.
6791 47. Added $local_user_uid and $local_user_gid at routing time.
6793 48. Exim crashed when trying to verify a sender address that was being
6796 49. Exim was recognizing only a space character after ".include". It now also
6797 recognizes a tab character.
6799 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6800 extracting the relevant information from the specification. The man page no
6801 longer contains scrambled data for the -d option, and I've added a section
6802 at the front about calling Exim under different names.
6804 51. Added "extra_headers" argument to the "mail" command in filter files.
6806 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6809 53. Installed eximstats 1.29.
6811 54. Added transport_filter_timeout as a generic transport option.
6813 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6814 Cc: header lines. This was required by RFC 822, but it not required by RFC
6817 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6818 incoming messages that did not have them. Now it does so only if the
6819 message originates locally, that is, if there is no associated remote host
6820 address. When Resent- header lines are present, this applies to the Resent-
6821 lines rather than the non-Resent- lines.
6823 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6824 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6826 58. Messages for configuration errors now include the name of the main
6827 configuration file - useful now that there may be more than one file in a
6828 list (.included file names were always shown).
6830 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6831 for those rare installations that do not start the daemon as root or run it
6832 setuid root. I've cut out the call to initgroups() if the daemon is not
6835 60. The Exim user and group can now be bound into the binary as text strings
6836 that are looked up at the start of Exim's processing.
6838 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6840 62. Added $mailstore_basename variable.
6842 63. Installed patch to sieve.c from Michael Haardt.
6844 64. When Exim failed to open the panic log after failing to open the main log,
6845 the original message it was trying to log was written to stderr and debug
6846 output, but if they were not available (the usual case in production), it
6847 was lost. Now it is written to syslog before the two lines that record the
6848 failures to open the logs.
6850 65. Users' Exim filters run in subprocesses under the user's uid. It is
6851 possible for a "deliver" command or an alias in a "personal" command to
6852 provoke an address rewrite. If logging of address rewriting is configured,
6853 this fails because the process is not running as root or exim. There may be
6854 a better way of dealing with this, but for the moment (because 4.30 needs
6855 to be released), I have disabled address rewrite logging when running a
6856 filter in a non-root, non-exim process.
6862 1. The buildconfig auxiliary program wasn't quoting the value set for
6863 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6864 not defined. This bug was masked in 4.22 by the effect that was fixed in
6867 2. Some messages that were rejected after a message id was allocated were
6868 shown as "incomplete" by exigrep. It no longer does this for messages that
6869 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6871 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6872 have allow_domain_literals set, the ID did not get logged in the <= line.
6873 Domain literals are now always recognized in Message-ID: header lines.
6875 4. The first argument for a ${extract expansion item is the key name or field
6876 number. Leading and trailing spaces in this item were not being ignored,
6877 causing some misleading effects.
6879 5. When deliver_drop_privilege was set, single queue runner processes started
6880 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6881 same command in the process it spins off) were not dropping privilege.
6883 6. When the daemon running as "exim" started a queue runner, it always
6884 re-executed Exim in the spun-off process. This is a waste of effort when
6885 deliver_drop_privilege is set. The new process now just calls the
6886 queue-runner function directly.
6892 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6895 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6896 ignored. Though the use of -oP was forcing the writing of a pid file, it
6897 was always written to the default place.
6899 3. If the message "no IP address found for host xxxx" is generated during
6900 incoming verification, it is now followed by identification of the incoming
6901 connection (so you can more easily find what provoked it).
6903 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6905 5. Added some features to "harden" Exim a bit more against certain attacks:
6907 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6908 be put in Local/Makefile. This is like the never_users runtime option,
6909 but it cannot be overridden. The default setting is "root".
6911 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6912 prefix string with which any file named in a -C command line option
6915 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6916 is retained for -C and -D only if the caller of Exim is root. Without
6917 it, the exim user may also use -C and -D and retain privilege.
6919 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6920 command line option is disabled.
6922 6. Macro names set by the -D option must start with an upper case letter, just
6923 like macro names defined in the configuration file.
6925 7. Added "dereference=" facility to LDAP.
6927 8. Two instances of the typo "uknown" in the source files are fixed.
6929 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6930 the Configure-Makefile script screwed up while processing it.
6932 10. Incorporated PCRE 4.4.
6934 11. The SMTP synchronization check was not operating right at the start of an
6935 SMTP session. For example, it could not catch a HELO sent before the client
6936 waited for the greeting. There is now a check for outstanding input at the
6937 point when the greeting is written. Because of the duplex, asynchronous
6938 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6939 way, but not yet received, when the check is performed.
6941 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6942 on TCP/IP sockets, because this apparently causes some broken clients to
6945 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6946 unchanged) from the Cygwin maintainer.
6948 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6949 was supported instead of testing for mbx. Effectively a typo.
6951 15. The spa authenticator server code was not checking that the input it
6952 received was valid base64.
6954 16. The debug output line for the "set" modifier in ACLs was not showing the
6955 name of the variable that was being set.
6957 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6959 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6960 lookup on its own. Something else was needed to trigger the lookup. For
6961 example, a match in host_lookup or the need for a host name in a host list.
6962 Now, if $sender_host_name is referenced and the host name has not yet been
6963 looked up, a lookup is performed. If the lookup fails, the variable remains
6964 empty, and $host_lookup_failed is set to "1".
6966 19. Added "eqi" as a case-independent comparison operator.
6968 20. The saslauthd authentication condition could segfault if neither service
6969 nor realm was specified.
6971 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6972 error message that was logged was misleading, and incoming SMTP
6973 connections were dropped. The message is now more accurate, and temporary
6974 errors are given to SMTP connections.
6976 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6977 (except RSET) with a 421 error, until QUIT is received. However, it was
6978 failing to send a response to QUIT.
6980 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6981 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6982 ACL is now run after the helo_try_verify_hosts code.
6984 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6985 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6986 case-independent, so other case variants are also recognized). Apparently
6987 some systems use these upper case variants.
6989 25. If more than two messages were waiting for the same host, and a transport
6990 filter was specified for the transport, Exim sent two messages over the
6991 same TCP/IP connection, and then failed with "socket operation on non-
6992 socket" when it tried to send the third.
6994 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6996 27. The extern definition of crypt16() in expand.c was not being excluded when
6997 the OS had its own crypt16() function.
6999 28. Added bounce_return_body as a new option, and bounce_return_size_limit
7000 as a preferred synonym for return_size_limit, both as an option and as an
7003 29. Added LIBS=-liconv to OS/Makefile-OSF1.
7005 30. Changed the default configuration ACL to relax the local part checking rule
7006 for addresses that are not in any local domains. For these addresses,
7007 slashes and pipe symbols are allowed within local parts, but the sequence
7008 /../ is explicitly forbidden.
7010 31. SPA server authentication was not clearing the challenge buffer before
7013 32. log_message in a "warn" ACL statement was writing to the reject log as
7014 well as to the main log, which contradicts the documentation and doesn't
7015 seem right (because no rejection is happening). So I have stopped it.
7017 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
7018 However, I am unable to do any testing of this.
7020 34. Fixed an infelicity in the appendfile transport. When checking directories
7021 for a mailbox, to see if any needed to be created, it was accidentally
7022 using path names with one or more superfluous leading slashes; tracing
7023 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
7025 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
7026 contents are added to the log line that is written for every discarded
7027 recipient. (Previously a log_message setting was ignored.)
7029 36. The ${quote: operator now quotes the string if it is empty.
7031 37. The install script runs exim in order to find its version number. If for
7032 some reason other than non-existence or emptiness, which it checks, it
7033 could not run './exim', it was installing it with an empty version number,
7034 i.e. as "exim-". This error state is now caught, and the installation is
7037 38. An argument was missing from the function that creates an error message
7038 when Exim fails to connect to the socket for saslauthd authentication.
7039 This could cause Exim to crash, or give a corrupted message.
7041 39. Added isip, isip4, and isip6 to ${if conditions.
7043 40. The ACL variables $acl_xx are now saved with the message, and can be
7044 accessed later in routers, transports, and filters.
7046 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
7047 strings in the file are not string-expanded.
7049 42. If a MAIL command specified a SIZE value that was too large to fit into an
7050 int variable, the check against message_size_limit failed. Such values are
7051 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
7052 day this will have to be increased, but I don't think I want to be around
7053 when emails are that large.
7060 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
7061 iconv() is not standard in FreeBSD.
7063 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
7064 IPv6 enabled. The observed symptom was a segmentation fault on return from
7065 the function os_common_find_running_interfaces() in src/os.c.
7067 3. In the check_special_case() function in daemon.c I had used "errno" as an
7068 argument name, which causes warnings on some systems. This was basically a
7069 typo, since it was named "eno" in the comments!
7071 4. The code that waits for the clock to tick (at a resolution of some fraction
7072 of a second) so as to ensure message-id uniqueness was always waiting for
7073 at least one whole tick, when it could have waited for less. [This is
7074 almost certainly not relevant at current processor speeds, where it is
7075 unlikely to ever wait at all. But we try to future-proof.]
7077 5. The function that sleeps for a time interval that includes fractions of a
7078 second contained a race. It did not block SIGALRM between setting the
7079 timer, and suspending (a couple of lines later). If the interval was short
7080 and the sigsuspend() was delayed until after it had expired, the suspension
7081 never ended. On busy systems this could lead to processes getting stuck for
7084 6. Some uncommon configurations may cause a lookup to happen in a queue runner
7085 process, before it forks any delivery processes. The open lookup caching
7086 mechanism meant that the open file or database connection was passed into
7087 the delivery process. The problem was that delivery processes always tidy
7088 up cached lookup data. This could cause a problem for the next delivery
7089 process started by the queue runner, because the external queue runner
7090 process does not know about the closure. So the next delivery process
7091 still has data in the lookup cache. In the case of a file lookup, there was
7092 no problem because closing a file descriptor in a subprocess doesn't affect
7093 the parent. However, if the lookup was caching a connection to a database,
7094 the connection was closed, and the second delivery process was likely to
7095 see errors such as "PGSQL: query failed: server closed the connection
7096 unexpectedly". The problem has been fixed by closing all cached lookups
7097 in a queue runner before running a delivery process.
7099 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
7100 seem to have the "const" qualifier which it has on other OS. I've
7103 8. Change 4.21/2 was too strict. It is only if there are two authenticators
7104 *of the same type* (client or server) with the same public name that an
7105 error should be diagnosed.
7107 9. When Exim looked up a host name for an IP address, but failed to find the
7108 original IP address when looking up the host name (a safety check), it
7109 output the message "<ip address> does not match any IP for NULL", which was
7110 confusing, to say the least. The bug was that the host name should have
7111 appeared instead of "NULL".
7113 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
7114 user that is built into the binary, and the -C or -D options is used, root
7115 privilege is dropped before the configuration file is read. In addition,
7116 logging is switched to stderr instead of the normal log files. If the
7117 configuration then re-defines the Exim user, the unprivileged environment
7118 is probably not what is expected, so Exim logs a panic warning message (but
7121 However, if deliver_drop_privilege is set, the unprivileged state may well
7122 be exactly what is intended, so the warning has been cut out in that case,
7123 and Exim is allowed to try to write to its normal log files.
7129 1. smtp_return_error_details was not giving details for temporary sender
7130 or receiver verification errors.
7132 2. Diagnose a configuration error if two authenticators have the same public
7135 3. Exim used not to create the message log file for a message until the first
7136 delivery attempt. This could be confusing when incoming messages were held
7137 for policy or load reasons. The message log file is now created at the time
7138 the message is received, and an initial "Received" line is written to it.
7140 4. The automatically generated man page for command line options had a minor
7141 bug that caused no ill effects; however, a more serious problem was that
7142 the procedure for building the man page automatically didn't always
7143 operate. Consequently, release 4.20 contains an out-of-date version. This
7144 shouldn't happen again.
7146 5. When building Exim with embedded Perl support, the script that builds the
7147 Makefile was calling 'perl' to find its compile-time parameters, ignoring
7148 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
7150 6. The freeze_tell option was not being used for messages that were frozen on
7151 arrival, either by an ACL or by local_scan().
7153 7. Added the smtp_incomplete_transaction log selector.
7155 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
7156 was accepting AUTH without a new EHLO.
7158 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
7159 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
7162 10. Logging of TCP/IP connections (when configured) now happens in the main
7163 daemon process instead of the child process, so that the TCP/IP connection
7164 count is more accurate (but it can never be perfect).
7166 11. The use of "drop" in a nested ACL was not being handled correctly in the
7167 outer ACL. Now, if condition failure induced by the nested "drop" causes
7168 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
7169 or "require"), the connection is dropped.
7171 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
7172 that yield "discard" can now be used with an "accept" or a "discard" verb,
7173 but an error is generated for any others (because I can't see a useful way
7174 to define what should happen).
7176 13. When an ACL is read dynamically from a file (or anywhere else), the lines
7177 are now processed in the same way as lines in the Exim configuration file.
7178 In particular, continuation lines are supported.
7180 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
7182 15. Added -ti meaning -t -i.
7184 16. Check for letters, digits, hyphens, and dots in the names of dnslist
7185 domains, and warn by logging if others are found.
7187 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
7188 returned from GIFCONF when Exim is trying to find the list of interfaces on
7189 a host. The code in os.c has been modified to copy each ifreq to an aligned
7190 structure in all cases.
7192 Also, in some cases, the returned ifreq's were being copied to a 'struct
7193 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
7194 means the last couple of bytes of an IPv6 address could be chopped if the
7195 ifreq contained only a normal sockaddr (14 bytes storage).
7197 18. Named domain lists were not supported in the hosts_treat_as_local option.
7198 An entry such as +xxxx was not recognized, and was treated as a literal
7201 19. Ensure that header lines added by a DATA ACL are included in the reject log
7202 if the ACL subsequently rejects the message.
7204 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
7205 MD5 (which is deprecated).
7207 21. When testing a filter file using -bf, Exim was writing a message when it
7208 took the sender from a "From " line in the message, but it was not doing so
7209 when it took $return_path from a Return-Path: header line. It now does.
7211 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
7212 with a valid header line field name (a series of printing characters
7213 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
7215 23. Changed "disc" in the source to "disk" to conform to the documentation and
7216 the book and for uniformity.
7218 24. Ignore Sendmail's -Ooption=value command line item.
7220 25. When execve() failed while trying to run a command in a pipe transport,
7221 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
7222 could be confused with a return value of 69 from the command itself. This
7223 has been changed to 127, the value the shell returns if it is asked to run
7224 a non-existent command. The wording for the related log line suggests a
7225 non-existent command as the problem.
7227 26. If received_header_text expands to an empty string, do not add a Received:
7228 header line to the message. (Well, it adds a token one on the spool, but
7229 marks it "old" so that it doesn't get used or transmitted.)
7231 27. Installed eximstats 1.28 (addition of -nt option).
7233 28. There was no check for failure on the call to getsockname() in the daemon
7234 code. This can fail if there is a shortage of resources on the system, with
7235 ENOMEM, for example. A temporary error is now given on failure.
7237 29. Contrary to the C standard, it seems that in some environments, the
7238 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
7239 program. Exim now does this explicitly; it affects the formatting of
7240 timestamps using strftime().
7242 30. If exiqsumm was given junk data, it threw up some uninitialized variable
7243 complaints. I've now initialized all the variables, to avoid this.
7245 32. Header lines added by a system filter were not being "seen" during
7246 transport-time rewrites.
7248 33. The info_callback() function passed to OpenSSL is set up with type void
7249 (*)(SSL *, int, int), as described somewhere. However, when calling the
7250 function (actually a macro) that sets it up, the type void(*)() is
7251 expected. I've put in a cast to prevent warnings from picky compilers.
7253 34. If a DNS black list lookup found a CNAME record, but there were no A
7254 records associated with the domain it pointed at, Exim crashed.
7256 35. If a DNS black list lookup returned more than one A record, Exim ignored
7257 all but the first. It now scans all returned addresses if a particular IP
7258 value is being sought. In this situation, the contents of the
7259 $dnslist_value variable are a list of all the addresses, separated by a
7262 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
7263 to accept a host name and all its aliases if the forward lookup for any of
7264 them yielded the IP address of the incoming connection. Now it accepts only
7265 those names whose forward lookup yields the correct IP address. Any other
7266 names are discarded. This closes a loophole whereby a rogue DNS
7267 administrator could create reverse DNS records to break through a
7268 wildcarded host restriction in an ACL.
7270 37. If a user filter or a system filter that ran in a subprocess used any of
7271 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
7272 the wrong values were passed to the pipe command ($thisaddress had the
7273 value of $0, $0 had the value of $1, etc). This bug was introduced by
7274 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
7276 38. Improved the line breaking for long SMTP error messages from ACLs.
7277 Previously, if there was no break point between 40 and 75 characters, Exim
7278 left the rest of the message alone. Two changes have been made: (a) I've
7279 reduced the minimum length to 35 characters; (b) if it can't find a break
7280 point between 35 and 75 characters, it looks ahead and uses the first one
7281 that it finds. This may give the occasional overlong line, but at least the
7282 remaining text gets split now.
7284 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
7285 file descriptors might be low, and that setting 1000 would always raise it.
7286 It turns out that in some environments, the limit is already over 1000 and
7287 that lowering it causes trouble. So now Exim takes care not to decrease it.
7289 40. When delivering a message, the value of $return_path is set to $sender_
7290 address at the start of routing (routers may change the value). By an
7291 oversight, this default was not being set up when an address was tested by
7292 -bt or -bv, which affected the outcome if any router or filter referred to
7295 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
7296 log only when "message" or "log_message" are set. However, if one of the
7297 conditions was an address verification, or a call to a nested ACL, the
7298 messages generated by the underlying test were being passed through. This
7299 no longer happens. The underlying message is available in $acl_verify_
7300 message for both "message" and "log_message" expansions, so it can be
7301 passed through if needed.
7303 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
7304 new expansion $bh_ to give the encoded byte string without charset
7305 translation. Translation happens only if iconv() is available; HAVE_ICONV
7306 indicates this at build time. HEADERS_CHARSET gives the charset to
7307 translate to; headers_charset can change it in the configuration, and
7308 "headers charset" can change it in an individual filter file.
7310 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
7311 that creates RFC 2047 encoded "words" labels them as that charset instead
7312 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
7313 expansion operator; (ii) when Exim creates a From: line for a local
7314 message; (iii) when a header line is rewritten to include a "phrase" part.
7316 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
7317 buggy, causing it to skip the first lines of messages whose message ID
7318 ended in 'D'. This would not have bitten before Exim release 4.14, because
7319 message IDs were unlikely to end in 'D' before then. The effect was to have
7320 incorrect size information for certain domains.
7322 45. #include "config.h" was missing at the start of the crypt16.c module. This
7323 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
7326 46. If there was a timeout during a "random" callout check, Exim treated it as
7327 a failure of the random address, and carried on sending RSET and the real
7328 address. If the delay was just some slowness somewhere, the response to the
7329 original RCPT would be taken as a response to RSET and so on, causing
7330 mayhem of various kinds.
7332 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
7333 when I implemented it. It didn't allow for the fact that some option values
7334 may legitimately be negative (e.g. size_addition), and it didn't even do
7335 the right test for positive values.
7337 48. Domain names in DNS records are case-independent. Exim always looks them up
7338 in lower case. Some resolvers return domain names in exactly the case they
7339 appear in the zone file, that is, they may contain uppercase letters. Not
7340 all resolvers do this - some return always lower case. Exim was treating a
7341 change of case by a resolver as a change of domain, similar to a widening
7342 of a domain abbreviation. This triggered its re-routing code and so it was
7343 trying to route what was effectively the same domain again. This normally
7344 caused routing to fail (because the router wouldn't handle the domain
7345 twice). Now Exim checks for this case specially, and just changes the
7346 casing of the domain that it ultimately uses when it transmits the message
7349 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
7352 50. If a filter generated a file delivery with a non-absolute name (possible if
7353 no home directory exists for the router), the forbid_file option was not
7356 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
7357 the existing equality matching.
7359 52. Exim was using ints instead of ino_t variables in some places where it was
7360 dealing with inode numbers.
7362 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
7363 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
7364 called TMPDIR, and if it finds it is different, it changes its value.
7366 54. The smtp_printf() function is now made available to local_scan() so
7367 additional output lines can be written before returning. There is also an
7368 smtp_fflush() function to enable the detection of a dropped connection.
7369 The variables smtp_input and smtp_batched_input are exported to
7372 55. Changed the default runtime configuration: the message "Unknown user"
7373 has been removed from the ACL, and instead placed on the localuser router,
7374 using the cannot_route_message feature. This means that any verification
7375 failures that generate their own messages won't get overridden. Similarly,
7376 the "Unrouteable address" message that was in the ACL for unverifiable
7377 relay addresses has also been removed.
7379 56. Added hosts_avoid_esmtp to the smtp transport.
7381 57. The exicyclog script was not checking for the esoteric option
7382 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7383 will work only if exicyclog is run under the appropriate euid.
7385 58. Following a discussion on the list, the rules by which Exim recognises line
7386 endings on incoming messages have been changed. The -dropcr and drop_cr
7387 options are now no-ops, retained only for backwards compatibility. The
7388 following line terminators are recognized: LF CRLF CR. However, special
7389 processing applies to CR:
7391 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7392 nor a local message in the state where . is a terminator.
7394 (ii) If a bare CR is encountered in a header line, an extra space is added
7395 after the line terminator so as not to end the header. The reasoning
7396 behind this is that bare CRs in header lines are most likely either
7397 to be mistakes, or people trying to play silly games.
7399 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7400 was being incorrectly given as 18 bytes larger than it should have been.
7401 This is a VOB (very old bug).
7403 60. This may never have affected anything current, but just in case it has:
7404 When the local host is found other than at the start of a list of hosts,
7405 the local host, those with the same MX, and any that follow, are discarded.
7406 When the list in question was part of a longer list of hosts, the following
7407 hosts (not currently being processed) were also being discarded. This no
7408 longer happens. I'm not sure if this situation could ever has previously
7411 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7414 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7415 header. This is something that is recommended in a new Internet Draft, and
7416 is something that is documented as being done by Sendmail. There are two
7417 possible values. For messages generated by the autoreply transport, Exim
7420 Auto-Submitted: auto-replied
7422 whereas for all other generated messages (e.g. bounces) it adds
7424 Auto-Submitted: auto-generated
7426 63. The "personal" condition in filters now includes a test for the
7427 Auto-Submitted: header. If it contains the string "auto-" the message it
7428 not considered personal.
7430 64. Added rcpt_include_affixes as a generic transport option.
7432 65. Added queue_only_override (default true).
7434 66. Added the syslog_duplication option.
7436 67. If what should have been the first header line of a message consisted of
7437 a space followed by a colon, Exim was mis-interpreting it as a header line.
7438 It isn't of course - it is syntactically invalid and should therefore be
7439 treated as the start of the message body. The misbehaviour could have
7440 caused a number of strange effects, including loss of data in subsequent
7441 header lines, and spool format errors.
7443 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7444 client host had authenticated. This control can now be exercised by an ACL
7445 for more flexibility.
7447 69. By default, callouts do not happen when testing with -bh. There is now a
7448 variant, -bhc, which does actually run the callout code, including
7449 consulting and updating the callout cache.
7451 70. Added support for saslauthd authentication, courtesy of Alexander
7454 71. If statvfs() failed on the spool or log directories while checking their
7455 size for availability, Exim confusingly gave the error "space shortage".
7456 Furthermore, in debugging mode it crashed with a floating point exception.
7457 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7458 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7459 serious problem, Exim now writes to the main and panic logs when this
7460 happens, with details of the failure. It then refuses to accept the
7461 incoming message, giving the message "spool directory problem" or "log
7462 directory problem" with a 421 code for SMTP messages.
7464 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7465 0, 1, and 2 exist, because some OS complain for execs without them (see
7466 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7467 descriptors. However, the code omitted to check that the open succeeded,
7468 causing mysterious errors if for some reason the permissions on /dev/null
7469 got screwed. Now Exim writes a message to the main and panic logs, and
7470 bombs out if it can't open /dev/null.
7472 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7473 interact so that it is all more flexible. It is supposed to remain
7474 backwards compatible. Also added extra_local_interfaces.
7476 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7477 to bomb out with an assertion failure - to the client this appears as a
7478 connection drop. This problem occurs in the part of the code that was taken
7479 from the Samba project. Fortunately, the assertion is in a very simple
7480 function, so I have fixed this by reproducing the function inline in the
7481 one place where it is called, and arranging for authentication to fail
7482 instead of killing the process with assert().
7484 75. The SPA client code was not working when the server requested OEM rather
7485 than Unicode encoding.
7487 76. Added code to make require_files with a specific uid setting more usable in
7488 the case where statting the file as root fails - usually a non-root-mounted
7489 NFS file system. When this happens and the failure is EACCES, Exim now
7490 forks a subprocess and does the per-uid checking as the relevant uid.
7492 77. Added process_log_path.
7494 78. If log_file_path was not explicitly set, a setting of check_log_space or
7495 check_log_inodes was ignored.
7497 79. If a space check for the spool or log partitions fails, the incident is now
7498 logged. Of course, in the latter case the data may get lost...
7500 80. Added the %p formatting code to string_format() so that it can be used to
7501 print addresses in debug_print(). Adjusted all the address printing in the
7502 debugging in store.c to use %p rather than %d.
7504 81. There was a concern that a line of code in smtp_in.c could overflow a
7505 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7506 initially expressed, the concern was not well-founded, because trailing
7507 spaces are removed early. However, if the trailing spaces were followed by
7508 a NULL, they did not get removed, so the overflow was possible. Two fixes
7511 (a) I re-wrote the offending code in a cleaner fashion.
7512 (b) If an incoming SMTP command contains a NULL character, it is rejected
7515 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7516 runs initgroups(), so that if the Exim user is in any additional groups,
7517 they will be used during message reception.
7523 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob