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/02 Bug 2390: Use message_id for tempfile creation to avoid races in a
74 shared (NFS) environment.
76 HS/03 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,
164 JH/35 Bug 2409: filter out-of-spec chars from callout response before using
165 them in our smtp response.
167 JH/36 Have the general router option retry_use_local_part default to true when
168 any of the restrictive preconditions are set (to anything). Previously it
169 was only for check_local user. The change removes one item of manual
170 configuration which is required for proper retries when a remote router
171 handles a subset of addresses for a domain.
173 JH/37 Appendfile: when evaluating quota use (non-quota_size_regex) take the file
174 link count into consideration.
176 HS/04 Fix handling of very log lines in -H files. If a -<key> <value> line
177 caused the extension of big_buffer, the following lines were ignored.
179 JH/38 Bug 1395: Teach the DNS negative-cache about TTL value from the SOA in
180 accordance with RFC 2308. Previously there was no expiry, so a longlived
181 receive process (eg. due to ACL delays) versus a short SOA value could
184 HS/05 Handle trailing backslash gracefully. (CVE-2019-15846)
186 JH/39 Promote DMARC support to mainline.
188 JH/40 Bug 2452: Add a References: header to DSNs.
190 JH/41 With GnuTLS 3.6.0 (and later) do not attempt to manage Diffie-Hellman
191 parameters. The relevant library call is documented as "Deprecated: This
192 function is unnecessary and discouraged on GnuTLS 3.6.0 or later. Since
193 3.6.0, DH parameters are negotiated following RFC7919."
195 HS/06 Change the default of dnssec_request_domains to "*"
200 JH/01 Remove code calling the customisable local_scan function, unless a new
201 definition "HAVE_LOCAL_SCAN=yes" is present in the Local/Makefile.
203 JH/02 Bug 1007: Avoid doing logging from signal-handlers, as that can result in
204 non-signal-safe functions being used.
206 JH/03 Bug 2269: When presented with a received message having a stupidly large
207 number of DKIM-Signature headers, disable DKIM verification to avoid
208 a resource-consumption attack. The limit is set at twenty.
210 JH/04 Add variables $arc_domains, $arc_oldest_pass for ARC verify. Fix the
211 report of oldest_pass in ${authres } in consequence, and separate out
212 some descriptions of reasons for verification fail.
214 JH/05 Bug 2273: Cutthrough delivery left a window where the received messsage
215 files in the spool were present and unlocked. A queue-runner could spot
216 them, resulting in a duplicate delivery. Fix that by doing the unlock
217 after the unlink. Investigation by Tim Stewart. Take the opportunity to
218 add more error-checking on spoolfile handling while that code is being
221 PP/01 Refuse to open a spool data file (*-D) if it's a symlink.
222 No known attacks, no CVE, this is defensive hardening.
224 JH/06 Bug 2275: The MIME ACL unlocked the received message files early, and
225 a queue-runner could start a delivery while other operations were ongoing.
226 Cutthrough delivery was a common victim, resulting in duplicate delivery.
227 Found and investigated by Tim Stewart. Fix by using the open message data
228 file handle rather than opening another, and not locally closing it (which
229 releases a lock) for that case, while creating the temporary .eml format
230 file for the MIME ACL. Also applies to "regex" and "spam" ACL conditions.
232 JH/07 Bug 177: Make a random-recipient callout success visible in ACL, by setting
233 $sender_verify_failure/$recipient_verify_failure to "random".
235 JH/08 When generating a selfsigned cert, use serial number 1 since zero is not
238 JH/09 Bug 2274: Fix logging of cmdline args when starting in an unlinked cwd.
239 Previously this would segfault.
241 JH/10 Fix ARC signing for case when DKIM signing failed. Previously this would
244 JH/11 Bug 2264: Exim now only follows CNAME chains one step by default. We'd
245 like zero, since the resolver should be doing this for us, But we need one
246 as a CNAME but no MX presence gets the CNAME returned; we need to check
247 that doesn't point to an MX to declare it "no MX returned" rather than
248 "error, loop". A new main option is added so the older capability of
249 following some limited number of chain links is maintained.
251 JH/12 Add client-ip info to non-pass iprev ${authres } lines.
253 JH/13 For receent Openssl versions (1.1 onward) use modern generic protocol
254 methods. These should support TLS 1.3; they arrived with TLS 1.3 and the
255 now-deprecated earlier definitions used only specified the range up to TLS
256 1.2 (in the older-version library docs).
258 JH/14 Bug 2284: Fix DKIM signing for body lines starting with a pair of dots.
260 JH/15 Rework TLS client-side context management. Stop using a global, and
261 explicitly pass a context around. This enables future use of TLS for
262 connections to service-daemons (eg. malware scanning) while a client smtp
263 connection is using TLS; with cutthrough connections this is quite likely.
265 JH/16 Fix ARC verification to do AS checks in reverse order.
267 JH/17 Support a "tls" option on the ${readsocket } expansion item.
269 JH/18 Bug 2287: Fix the protocol name (eg utf8esmtp) for multiple messages
270 using the SMTPUTF8 option on their MAIL FROM commands, in one connection.
271 Previously the "utf8" would be re-prepended for every additional message.
273 JH/19 Reject MAIL FROM commands with SMTPUTF8 when the facility was not advertised.
274 Previously thery were accepted, resulting in issues when attempting to
275 forward messages to a non-supporting MTA.
277 PP/02 Let -n work with printing macros too, not just options.
279 JH/20 Bug 2296: Fix cutthrough for >1 address redirection. Previously only
280 one parent address was copied, and bogus data was used at delivery-logging
281 time. Either a crash (after delivery) or bogus log data could result.
282 Discovery and analysis by Tim Stewart.
284 PP/03 Make ${utf8clean:} expansion operator detect incomplete final character.
285 Previously if the string ended mid-character, we did not insert the
286 promised '?' replacement.
288 PP/04 Documentation: current string operators work on bytes, not codepoints.
290 JH/21 Change as many as possible of the global flags into one-bit bitfields; these
291 should pack well giving a smaller memory footprint so better caching and
292 therefore performance. Group the declarations where this can't be done so
293 that the byte-sized flag variables are not interspersed among pointer
294 variables, giving a better chance of good packing by the compiler.
296 JH/22 Bug 1896: Fix the envelope from for DMARC forensic reports to be possibly
297 non-null, to avoid issues with sites running BATV. Previously reports were
298 sent with an empty envelope sender so looked like bounces.
300 JH/23 Bug 2318: Fix the noerror command within filters. It wasn't working.
301 The ignore_error flag wasn't being returned from the filter subprocess so
302 was not set for later routers. Investigation and fix by Matthias Kurz.
304 JH/24 Bug 2310: Raise a msg:fail:internal event for each undelivered recipient,
305 and a msg:complete for the whole, when a message is manually removed using
306 -Mrm. Developement by Matthias Kurz, hacked on by JH.
308 JH/25 Avoid fixed-size buffers for pathnames in DB access. This required using
309 a "Gnu special" function, asprintf() in the DB utility binary builds; I
310 hope that is portable enough.
312 JH/26 Bug 2311: Fix DANE-TA verification under GnuTLS. Previously it was also
313 requiring a known-CA anchor certificate; make it now rely entirely on the
314 TLSA as an anchor. Checking the name on the leaf cert against the name
315 on the A-record for the host is still done for TA (but not for EE mode).
317 JH/27 Fix logging of proxy address. Previously, a pointless "PRX=[]:0" would be
318 included in delivery lines for non-proxied connections, when compiled with
319 SUPPORT_SOCKS and running with proxy logging enabled.
321 JH/28 Bug 2314: Fire msg:fail:delivery event even when error is being ignored.
322 Developement by Matthias Kurz, tweaked by JH. While in that bit of code,
323 move the existing event to fire before the normal logging of message
324 failure so that custom logging is bracketed by normal logging.
326 JH/29 Bug 2322: A "fail" command in a non-system filter (file) now fires the
327 msg:fail:internal event. Developement by Matthias Kurz.
329 JH/30 Bug 2329: Increase buffer size used for dns lookup from 2k, which was
330 far too small for todays use of crypto signatures stored there. Go all
331 the way to the max DNS message size of 64kB, even though this might be
332 overmuch for IOT constrained device use.
334 JH/31 Fix a bad use of a copy function, which could be used to pointlessly
335 copy a string over itself. The library routine is documented as not
336 supporting overlapping copies, and on MacOS it actually raised a SIGABRT.
338 JH/32 For main options check_spool_space and check_inode_space, where the
339 platform supports 64b integers, support more than the previous 2^31 kB
340 (i.e. more than 2 TB). Accept E, P and T multipliers in addition to
341 the previous G, M, k.
343 JH/33 Bug 2338: Fix the cyrus-sasl authenticator to fill in the
344 $authenticated_fail_id variable on authentication failure. Previously
347 JH/34 Increase RSA keysize of autogen selfsign cert from 1024 to 2048. RHEL 8.0
348 OpenSSL didn't want to use such a weak key. Do for GnuTLS also, and for
349 more-modern GnuTLS move from GNUTLS_SEC_PARAM_LOW to
350 GNUTLS_SEC_PARAM_MEDIUM.
352 JH/35 OpenSSL: fail the handshake when SNI processing hits a problem, server
353 side. Previously we would continue as if no SNI had been received.
355 JH/36 Harden the handling of string-lists. When a list consisted of a sole
356 "<" character, which should be a list-separator specification, we walked
357 off past the nul-terimation.
359 JH/37 Bug 2341: Send "message delayed" warning MDNs (restricted to external
360 causes) even when the retry time is not yet met. Previously they were
361 not, meaning that when (say) an account was over-quota and temp-rejecting,
362 and multiple senders' messages were queued, only one sender would get
363 notified on each configured delay_warning cycle.
365 JH/38 Bug 2351: Log failures to extract envelope addresses from message headers.
367 JH/39 OpenSSL: clear the error stack after an SSL_accept(). With anon-auth
368 cipher-suites, an error can be left on the stack even for a succeeding
369 accept; this results in impossible error messages when a later operation
372 AM/01 Bug 2359: GnuTLS: repeat lowlevel read and write operations while they
373 return error codes indicating retry. Under TLS1.3 this becomes required.
375 JH/40 Fix the feature-cache refresh for EXPERIMENTAL_PIPE_CONNECT. Previously
376 it only wrote the new authenticators, resulting in a lack of tracking of
377 peer changes of ESMTP extensions until the next cache flush.
379 JH/41 Fix the loop reading a message header line to check for integer overflow,
380 and more-often against header_maxsize. Previously a crafted message could
381 induce a crash of the recive process; now the message is cleanly rejected.
383 JH/42 Bug 2366: Fix the behaviour of the dkim_verify_signers option. It had
384 been totally disabled for all of 4.91. Discovery and fix by "Mad Alex".
390 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
391 When redis_servers is set to a list of > 1 element, and the Redis servers
392 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
393 case of MOVED into a DEFER case instead, thus moving the query onto the
394 next server in the list. For a cluster of N elements, all N servers must
395 be defined in redis_servers.
397 GF/02 Catch and remove uninitialized value warning in exiqsumm
398 Check for existence of @ARGV before looking at $ARGV[0]
400 JH/01 Replace the store_release() internal interface with store_newblock(),
401 which internalises the check required to safely use the old one, plus
402 the allocate and data copy operations duplicated in both (!) of the
403 extant use locations.
405 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
406 modifier. This matches the restriction on the commandline.
408 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
409 Previously only the last row was returned.
411 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
412 we assumed that tags in the header were well-formed, and parsed the
413 element content after inspecting only the first char of the tag.
414 Assumptions at that stage could crash the receive process on malformed
417 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
418 While running the DKIM ACL we operate on the Permanent memory pool so that
419 variables created with "set" persist to the DATA ACL. Also (at any time)
420 DNS lookups that fail create cache records using the Permanent pool. But
421 expansions release any allocations made on the current pool - so a dnsdb
422 lookup expansion done in the DKIM ACL releases the memory used for the
423 DNS negative-cache, and bad things result. Solution is to switch to the
424 Main pool for expansions.
425 While we're in that code, add checks on the DNS cache during store_reset,
426 active in the testsuite.
427 Problem spotted, and debugging aided, by Wolfgang Breyha.
429 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
430 When none of the hosts presented to a transport match an already-open
431 connection, close it and proceed with the list. Previously we would
432 queue the message. Spotted by Lena with Yahoo, probably involving
435 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
436 Previously a spurious "250 OK id=" response was appended to the proper
439 JH/08 The "support for" informational output now, which built with Content
440 Scanning support, has a line for the malware scanner interfaces compiled
441 in. Interface can be individually included or not at build time.
443 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
444 by the template makefile "src/EDITME". The "STREAM" support for an older
445 ClamAV interface method is removed.
447 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
448 rows affected is given instead).
450 JH/11 The runtime Berkeley DB library version is now additionally output by
451 "exim -d -bV". Previously only the compile-time version was shown.
453 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
454 SMTP connection. Previously, when one had more receipients than the
455 first, an abortive onward connection was made. Move to full support for
456 multiple onward connections in sequence, handling cutthrough connection
457 for all multi-message initiating connections.
459 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
460 routers. Previously, a multi-recipient message would fail to match the
461 onward-connection opened for the first recipient, and cause its closure.
463 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
464 a timeout on read on a GnuTLS initiating connection, resulting in the
465 initiating connection being dropped. This mattered most when the callout
466 was marked defer_ok. Fix to keep the two timeout-detection methods
469 JH/15 Relax results from ACL control request to enable cutthrough, in
470 unsupported situations, from error to silently (except under debug)
471 ignoring. This covers use with PRDR, frozen messages, queue-only and
474 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
476 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
477 metadata, resulting in a crash in free().
479 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
480 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
481 Broken also in d185889f4, with init system revamp.
483 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
484 Previously we abruptly closed the connection after reading a malware-
485 found indication; now we go on to read the "scan ok" response line,
488 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
489 ACL. Previously, a crash would result.
491 JH/19 Speed up macro lookups during configuration file read, by skipping non-
492 macro text after a replacement (previously it was only once per line) and
493 by skipping builtin macros when searching for an uppercase lead character.
495 JH/20 DANE support moved from Experimental to mainline. The Makefile control
496 for the build is renamed.
498 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
499 was allocated for every new TLS startup, meaning one per message. Fix
500 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
502 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
503 reported the original. Fix to report (as far as possible) the ACL
504 result replacing the original.
506 JH/23 Fix memory leak during multi-message connections using STARTTLS under
507 OpenSSL. Certificate information is loaded for every new TLS startup,
508 and the resources needed to be freed.
510 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
512 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
515 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
516 DATA response info to the (existing) per-recipient response info for
517 the "C=" log element. It can have useful tracking info from the
518 destination system. Patch from Simon Arlott.
520 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
521 length value. Previously this would segfault.
523 HS/02 Support Avast multiline protoocol, this allows passing flags to
524 newer versions of the scanner.
526 JH/28 Ensure that variables possibly set during message acceptance are marked
527 dead before release of memory in the daemon loop. This stops complaints
528 about them when the debug_store option is enabled. Discovered specifically
529 for sender_rate_period, but applies to a whole set of variables.
530 Do the same for the queue-runner and queue-list loops, for variables set
531 from spool message files. Do the same for the SMTP per-message loop, for
532 certain variables indirectly set in ACL operations.
534 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
535 as a multi-recipient message from a mailinglist manager). The coding had
536 an arbitrary cutoff number of characters while checking for more input;
537 enforced by writing a NUL into the buffer. This corrupted long / fast
538 input. The problem was exposed more widely when more pipelineing of SMTP
539 responses was introduced, and one Exim system was feeding another.
540 The symptom is log complaints of SMTP syntax error (NUL chars) on the
541 receiving system, and refused recipients seen by the sending system
542 (propating to people being dropped from mailing lists).
543 Discovered and pinpointed by David Carter.
545 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
546 replaced by the ${authresults } expansion.
548 JH/31 Bug 2257: Fix pipe transport to not use a socket-only syscall.
550 HS/03 Set a handler for SIGTERM and call exit(3) if running as PID 1. This
551 allows proper process termination in container environments.
553 JH/32 Bug 2258: Fix spool_wireformat in combination with LMTP transport.
554 Previously the "final dot" had a newline after it; ensure it is CR,LF.
556 JH/33 SPF: remove support for the "spf" ACL condition outcome values "err_temp"
557 and "err_perm", deprecated since 4.83 when the RFC-defined words
558 "temperror" and "permerror" were introduced.
560 JH/34 Re-introduce enforcement of no cutthrough delivery on transports having
561 transport-filters or DKIM-signing. The restriction was lost in the
562 consolidation of verify-callout and delivery SMTP handling.
563 Extend the restriction to also cover ARC-signing.
565 JH/35 Cutthrough: for a final-dot response timeout (and nonunderstood responses)
566 in defer=pass mode supply a 450 to the initiator. Previously the message
569 PP/02 DANE: add dane_require_tls_ciphers SMTP Transport option; if unset,
570 tls_require_ciphers is used as before.
572 HS/03 Malware Avast: Better match the Avast multiline protocol. Add
573 "pass_unscanned". Only tmpfails from the scanner are written to
574 the paniclog, as they may require admin intervention (permission
575 denied, license issues). Other scanner errors (like decompression
576 bombs) do not cause a paniclog entry.
578 JH/36 Fix reinitialisation of DKIM logging variable between messages.
579 Previously it was possible to log spurious information in receive log
582 JH/37 Bug 2255: Revert the disable of the OpenSSL session caching. This
583 triggered odd behaviour from Outlook Express clients.
585 PP/03 Add util/renew-opendmarc-tlds.sh script for safe renewal of public
588 JH/38 DKIM: accept Ed25519 pubkeys in SubjectPublicKeyInfo-wrapped form,
589 since the IETF WG has not yet settled on that versus the original
590 "bare" representation.
592 JH/39 Fix syslog logging for syslog_timestamp=no and log_selector +millisec.
593 Previously the millisecond value corrupted the output.
594 Fix also for syslog_pid=no and log_selector +pid, for which the pid
595 corrupted the output.
601 JH/01 Rework error string handling in TLS interface so that the caller in
602 more cases is responsible for logging. This permits library-sourced
603 string to be attached to addresses during delivery, and collapses
604 pairs of long lines into single ones.
606 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
607 during configuration. Wildcards are allowed and expanded.
609 JH/02 Rework error string handling in DKIM to pass more info back to callers.
610 This permits better logging.
612 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
613 do not close it down and have the child transport start it up again on
614 the passed-on TCP connection. Instead, proxy the child (and any
615 subsequent ones) for TLS via a unix-domain socket channel. Logging is
616 affected: the continued delivery log lines do not have any DNSSEC, TLS
617 Certificate or OCSP information. TLS cipher information is still logged.
619 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
620 identical IP addresses on different listening ports. Will also affect
623 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
624 add noisy ifdef guards to special-case this sillyness.
625 Patch from Bernd Kuhls.
627 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
628 than 255 are no longer allowed.
630 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
631 Disable the session-cache too, which might reduce our load. Since we
632 currrectly use a new context for every connection, both as server and
633 client, there is no benefit for these.
634 GnuTLS appears to not support tickets server-side by default (we don't
635 call gnutls_session_ticket_enable_server()) but client side is enabled
636 by default on recent versions (3.1.3 +) unless the PFS priority string
639 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
640 <https://reproducible-builds.org/specs/source-date-epoch/>.
642 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
643 the check for any unsuccessful recipients did not notice the limit, and
644 erroneously found still-pending ones.
646 JH/08 Pipeline CHUNKING command and data together, on kernels that support
647 MSG_MORE. Only in-clear (not on TLS connections).
649 JH/09 Avoid using a temporary file during transport using dkim. Unless a
650 transport-filter is involved we can buffer the headers in memory for
651 creating the signature, and read the spool data file once for the
652 signature and again for transmission.
654 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
655 4.77 as the kernel support then wasn't solid, having issues in 64bit
656 mode. Now, it's been long enough. Add support for FreeBSD also.
658 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
659 case where the routing stage had gathered several addresses to send to
660 a host before calling the transport for the first, we previously failed
661 to close down TLS in the old transport process before passing the TCP
662 connection to the new process. The new one sent a STARTTLS command
663 which naturally failed, giving a failed delivery and bloating the retry
664 database. Investigation and fix prototype from Wolfgang Breyha.
666 JH/12 Fix check on SMTP command input synchronisation. Previously there were
667 false-negatives in the check that the sender had not preempted a response
668 or prompt from Exim (running as a server), due to that code's lack of
669 awareness of the SMTP input buffering.
671 PP/04 Add commandline_checks_require_admin option.
672 Exim drops privileges sanely, various checks such as -be aren't a
673 security problem, as long as you trust local users with access to their
674 own account. When invoked by services which pass untrusted data to
675 Exim, this might be an issue. Set this option in main configuration
676 AND make fixes to the calling application, such as using `--` to stop
679 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
680 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
681 responses to those, into a single TLS record each way (this usually means
682 a single packet). As a side issue, smtp_enforce_sync now works on TLS
685 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
686 affects you only if you're dancing at the edge of the param size limits.
687 If you are, and this message makes sense to you, then: raise the
688 configured limit or use OpenSSL 1.1. Nothing we can do for older
691 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
692 cmdline element to get the documented default one. Previously it was
695 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
696 are unsuable for non-message-specific reasons - eg. connection timeout,
697 banner-time rejection.
699 JH/16 Fix logging of delivery remote port, when specified by router, under
702 PP/06 Repair manualroute's ability to take options in any order, even if one
703 is the name of a transport.
706 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
708 JH/17 Change the list-building routines interface to use the expanding-string
709 triplet model, for better allocation and copying behaviour.
711 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
712 Previously it was constructed the first time a possibly-matching string
713 was met in the configuration file input during startup; now it is done
716 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
717 compatible one, to avoid the (poorly documented) possibility of a config
718 file in the working directory redirecting the DB files, possibly correpting
719 some existing file. CVE-2017-10140 assigned for BDB.
721 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
722 cache-hot. Previously, although the result was properly cached, the
723 initial verify call returned a defer.
725 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
726 the main verify for receipient in uncached-mode.
728 JH/22 Retire historical build files to an "unsupported" subdir. These are
729 defined as "ones for which we have no current evidence of testing".
731 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
732 if present. Previously it was ignored.
734 JH/24 Start using specified-initialisers in C structure init coding. This is
735 a C99 feature (it's 2017, so now considered safe).
737 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
738 if was a fixed-sized field and bitmask ops via macros; it is now more
741 PP/07 GitHub PR 56: Apply MariaDB build fix.
742 Patch provided by Jaroslav Škarvada.
744 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
745 during Coverity cleanups [4.87 JH/47]
746 Diagnosis and fix provided by Michael Fischer v. Mollard.
748 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
749 the right size to place the terminating semicolon on its own folded
750 line, the header hash was calculated to an incorrect value thanks to
751 the (relaxed) space the fold became.
753 HS/02 Fix Bug 2130: large writes from the transport subprocess were chunked
754 and confused the parent.
756 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
757 which could crash as a result. This could lead to undeliverable messages.
759 JH/28 Logging: "next input sent too soon" now shows where input was truncated
762 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
763 matters on fast-turnover and PID-randomising systems, which were getting
764 out-of-order delivery.
766 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
767 a possibly-overlapping copy. The symptom was that "Remote host closed
768 connection in response to HELO" was logged instead of the actual 4xx
771 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
772 Previously only that bufferd was discarded, resulting in SYMTP command
775 JH/32 DKIM: when a message has multiple signatures matching an identity given
776 in dkim_verify_signers, run the dkim acl once for each. Previously only
777 one run was done. Bug 2189.
779 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
780 "panic the current process" to "deliberately defer". The panic log is
781 still written with the problem list name; the mail and reject logs now
782 get a temp-reject line for the message that was being handled, saying
783 something like "domains check lookup or other defer". The SMTP 451
784 message is still "Temporary local problem".
786 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
787 A crafted sequence of BDAT commands could result in in-use memory beeing
788 freed. CVE-2017-16943.
790 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
791 from SMTP input. Previously it was always done; now only done for DATA
792 and not BDAT commands. CVE-2017-16944.
794 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
795 to the message (such as an overlong header line). Previously this was
796 not done and we did not exit BDAT mode. Followon from the previous item
797 though a different problem.
803 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
804 than -2003 did; needs libidn2 in addition to libidn.
806 JH/02 The path option on a pipe transport is now expanded before use.
808 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
809 Patch provided by "Björn", documentation fix added too.
811 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
812 missing a wire-to-host endian conversion.
814 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
815 close after a BDAT command line could be taken as a following command,
816 giving a synch failure. Fix by only checking for synch immediately
817 before acknowledging the chunk.
819 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
820 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
821 macro. Patches provided by Josh Soref.
823 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
824 Previously we did not; the RFC seems ambiguous and VRFY is not listed
825 by IANA as a service extension. However, John Klensin suggests that we
828 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
829 the dkim code may be unix-mode line endings rather than smtp wire-format
830 CRLF, so prepend a CR to any bare LF.
832 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
833 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
835 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
836 the first were themselves being wrongly included in the feed into dkim
837 processing; with most chunk sizes in use this resulted in an incorrect
838 body hash calculated value.
840 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
841 DKIM signature block, for verification. Although advised against by
842 standards it is specifically not ruled illegal.
844 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
846 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
847 missing a body hash (the bh= tag).
849 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
850 It seems that HAProxy sends the Proxy Protocol information in clear and
851 only then does a TLS startup, so do the same.
853 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
854 TCP connections (such as for Spamd) unless the daemon successfully set
855 Fast Open mode on its listening sockets. This fixes breakage seen on
856 too-old kernels or those not configured for Fast Open, at the cost of
857 requiring both directions being enabled for TFO, and TFO never being used
858 by non-daemon-related Exim processes.
860 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
861 endings, at least on the first header line. Try to canonify any that get
862 past that check, despite the cost.
864 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
865 now limited to an arbitrary five deep, while parsing addresses with the
866 strip_excess_angle_brackets option enabled.
868 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
869 instead leave the unprompted TLS handshake in socket buffer for the
870 TLS library to consume.
872 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
874 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
876 JH/16 Drop variables when they go out of scope. Memory management drops a whole
877 region in one operation, for speed, and this leaves assigned pointers
878 dangling. Add checks run only under the testsuite which checks all
879 variables at a store-reset and panics on a dangling pointer; add code
880 explicitly nulling out all the variables discovered. Fixes one known
881 bug: a transport crash, where a dangling pointer for $sending_ip_address
882 originally assigned in a verify callout, is re-used.
884 PP/06 Drop '.' from @INC in various Perl scripts.
886 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
888 PP/08 Reduce a number of compilation warnings under clang; building with
889 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
890 should be warning-free.
892 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
894 HS/01 Fix portability problems introduced by PP/08 for platforms where
895 realloc(NULL) is not equivalent to malloc() [SunOS et al].
897 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
898 chunk. This allows us to accept broken chunked messages. We need a more
899 general solution here.
901 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
902 already-broken messages in the queue.
904 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
906 JH/19 Fix reference counting bug in routing-generated-address tracking.
912 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
913 supports it and a size is available (ie. the sending peer gave us one).
915 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
916 years of being deprecated). The replacements are the ACLs
917 acl_smtp_mime and acl_not_smtp_mime.
919 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
920 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
921 or even an in-clear connection were permitted. Now, if the host lookup
922 was dnssec and dane was requested then the host is only used if the
923 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
924 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
925 if one fails this test.
926 This means that a poorly-configured remote DNS will make it incommunicado;
927 but it protects against a DNS-interception attack on it.
929 JH/04 Bug 1810: make continued-use of an open smtp transport connection
930 non-noisy when a race steals the message being considered.
932 JH/05 If main configuration option tls_certificate is unset, generate a
933 self-signed certificate for inbound TLS connections.
935 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
936 in rewrites and routers.
938 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
939 and logged a warning sing 4.83; now they are a configuration file error.
941 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
942 (lacking @domain). Apply the same qualification processing as RCPT.
944 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
946 JH/10 Support ${sha256:} applied to a string (as well as the previous
949 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
950 a cutthrough deliver is pending, as we always want to make a connection.
951 This also avoids re-routing the message when later placing the cutthrough
952 connection after a verify cache hit.
953 Do not update it with the verify result either.
955 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
956 when routing results in more than one destination address.
958 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
959 signing (which inhibits the cutthrough capability). Previously only
960 the presence of an option was tested; now an expansion evaluating as
961 empty is permissible (obviously it should depend only on data available
962 when the cutthrough connection is made).
964 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
965 the relevant preceding SMTP command did not note the pipelining mode.
967 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
968 Previously they were not counted.
970 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
971 as one having no matching records. Previously we deferred the message
972 that needed the lookup.
974 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
975 distinguished as "(=".
977 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
978 for missing MX records. Previously it only worked for missing A records.
980 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
982 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
983 after the data-go-ahead and data-ack. Patch from Jason Betts.
985 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
986 even for a "none" policy. Patch from Tony Meyer.
988 JH/22 Fix continued use of a connection for further deliveries. If a port was
989 specified by a router, it must also match for the delivery to be
992 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
993 When one of the recipients of a message was unsuitable for the connection
994 (has no matching addresses), we lost track of needing to mark it
995 deferred. As a result mail would be lost.
997 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
999 JH/25 Decoding ACL controls is now done using a binary search; the source code
1000 takes up less space and should be simpler to maintain. Merge the ACL
1001 condition decode tables also, with similar effect.
1003 JH/26 Fix problem with one_time used on a redirect router which returned the
1004 parent address unchanged. A retry would see the parent address marked as
1005 delivered, so not attempt the (identical) child. As a result mail would
1008 JH/27 Fix a possible security hole, wherein a process operating with the Exim
1009 UID can gain a root shell. Credit to http://www.halfdog.net/ for
1010 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
1013 JH/28 Enable {spool,log} filesystem space and inode checks as default.
1014 Main config options check_{log,spool}_{inodes,space} are now
1015 100 inodes, 10MB unless set otherwise in the configuration.
1017 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
1018 Previously it only applied to the main-section connection policy
1021 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
1023 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
1024 by me. Added RFC7919 DH primes as an alternative.
1026 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
1027 are not in the system include path.
1029 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
1030 GnuTLS, when a session startup failed (eg because the client disconnected)
1031 Exim did stdio operations after fclose. This was exposed by a recent
1032 change which nulled out the file handle after the fclose.
1034 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
1035 signed directly by the cert-signing cert, rather than an intermediate
1036 OCSP-signing cert. This is the model used by LetsEncrypt.
1038 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
1040 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
1041 an incoming connection.
1043 HS/02 Bug 1802: Do not half-close the connection after sending a request
1046 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
1047 fallback to "prime256v1".
1049 JH/34 SECURITY: Use proper copy of DATA command in error message.
1050 Could leak key material. Remotely exploitable. CVE-2016-9963.
1056 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
1057 and 3.4.4 - once the server is enabled to respond to an OCSP request
1058 it does even when not requested, resulting in a stapling non-aware
1059 client dropping the TLS connection.
1061 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
1062 support variable-length bit vectors. No functional change.
1064 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
1065 The I= interface field on outgoing lines is now after the H= remote
1066 host field, same as incoming lines. There is a separate
1067 outgoing_interface log selector which allows you to disable the
1070 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
1071 If not running log_selector +smtp_connection the mainlog would be held
1072 open indefinitely after a "too many connections" event, including to a
1073 deleted file after a log rotate. Leave the per net connection logging
1074 leaving it open for efficiency as that will be quickly detected by the
1075 check on the next write.
1077 HS/01 Bug 1671: Fix post transport crash.
1078 Processing the wait-<transport> messages could crash the delivery
1079 process if the message IDs didn't exist for some reason. When
1080 using 'split_spool_directory=yes' the construction of the spool
1081 file name failed already, exposing the same netto behaviour.
1083 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
1084 mime_regex ACL conditions.
1086 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
1087 to DSN fail messages (bounces): remote IP, remote greeting, remote response
1088 to HELO, local diagnostic string.
1090 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
1091 log line to debug. Even when configured with a tls authenticator many
1092 client connections are expected to not authenticate in this way, so
1093 an authenticate fail is not an error.
1095 HS/02 Add the Exim version string to the process info. This way exiwhat
1096 gives some more detail about the running daemon.
1098 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
1099 matter for fast-change records such as DNSBLs.
1101 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
1102 retry record, even if constant. There may be multiple transports with
1103 different interface settings and the retry behaviour needs to be kept
1106 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
1108 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
1110 JH/10 Bug 840: fix log_defer_output option of pipe transport
1112 JH/11 Bug 830: use same host for all RCPTS of a message, even under
1113 hosts_randomize. This matters a lot when combined with mua_wrapper.
1115 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
1116 ${quote_pgsql:<string>} operator.
1118 JH/13 Bug 1708: avoid misaligned access in cached lookup.
1120 JH/14 Change header file name for freeradius-client. Relevant if compiling
1121 with Radius support; from the Gentoo tree and checked under Fedora.
1123 JH/15 Bug 1712: Introduce $prdr_requested flag variable
1125 JH/16 Bug 1714: Permit an empty string as expansion result for transport
1126 option transport_filter, meaning no filtering.
1128 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
1130 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
1131 defaults to "*" (all hosts). The variable is now available when not built
1132 with TLS, default unset, mainly to enable keeping the testsuite sane.
1133 If a server certificate is not supplied (via tls_certificate) an error is
1134 logged, and clients will find TLS connections fail on startup. Presumably
1135 they will retry in-clear.
1136 Packagers of Exim are strongly encouraged to create a server certificate
1137 at installation time.
1139 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
1140 with the $config_file variable.
1142 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
1143 in transport context, after the attempt, and per-recipient. The latter type
1144 is per host attempted. The event data is the error message, and the errno
1145 information encodes the lookup type (A vs. MX) used for the (first) host,
1146 and the trailing two digits of the smtp 4xx response.
1148 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
1149 to write to mainlog (or rejectlog, paniclog) in the window between file
1150 creation and permissions/ownership being changed. Particularly affects
1151 installations where exicyclog is run as root, rather than exim user;
1152 result is that the running daemon panics and dies.
1154 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
1156 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
1157 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
1158 "pri" and "weight". Note that the previous implicit priority given by the
1159 list order is no longer honoured.
1161 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
1162 for DKIM processing.
1164 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
1165 by defining SUPPORT_SOCKS.
1167 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
1168 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
1169 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
1170 variables are renamed to proxy_{local,external}_{address,port}.
1172 JH/27 Move Internationalisation support from Experimental to mainline, enabled
1173 for a build by defining SUPPORT_I18N
1175 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
1176 of the query string, and make ${quote_redis:} do that quoting.
1178 JH/29 Move Events support from Experimental to mainline, enabled by default
1179 and removable for a build by defining DISABLE_EVENT.
1181 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
1183 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
1184 cached by the daemon.
1186 JH/32 Move Redis support from Experimental to mainline, enabled for a build
1187 by defining LOOKUP_REDIS. The libhiredis library is required.
1189 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
1190 keys are given for lookup.
1192 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
1193 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
1194 only supported when built with TLS support. The PolarSSL SHA routines
1195 are still used when the TLS library is too old for convenient support.
1197 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
1198 openssl_options), for security. OpenSSL forces this from version 1.1.0
1199 server-side so match that on older versions.
1201 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
1202 allocation for $value could be released as the expansion processing
1203 concluded, but leaving the global pointer active for it.
1205 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
1206 and to use the domains and local_parts ACL conditions.
1208 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
1209 incorrectly not doubled on cutthrough transmission, hence seen as a
1210 body-termination at the receiving system - resulting in truncated mails.
1211 Commonly the sender saw a TCP-level error, and retransmitted the message
1212 via the normal store-and-forward channel. This could result in duplicates
1213 received - but deduplicating mailstores were liable to retain only the
1214 initial truncated version.
1216 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
1218 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
1220 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
1221 we're in there, support oversigning also; bug 1309.
1223 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
1225 HS/04 Add support for keep_environment and add_environment options.
1227 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
1228 either intentional arithmetic overflow during PRNG, or testing config-
1231 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
1232 delivery resulted in actual delivery. Cancel cutthrough before DATA
1235 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
1236 rejects a recipient: pass the reject to the originator.
1238 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
1239 Many were false-positives and ignorable, but it's worth fixing the
1242 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
1243 for the new environment-manipulation done at startup. Move the routines
1244 from being local to tls.c to being global via the os.c file.
1246 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
1247 an extract embedded as result-arg for a map, the first arg for extract
1248 is unavailable so we cannot tell if this is a numbered or keyed
1249 extraction. Accept either.
1255 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
1258 JH/02 The smtp transport option "multi_domain" is now expanded.
1260 JH/03 The smtp transport now requests PRDR by default, if the server offers
1263 JH/04 Certificate name checking on server certificates, when exim is a client,
1264 is now done by default. The transport option tls_verify_cert_hostnames
1265 can be used to disable this per-host. The build option
1266 EXPERIMENTAL_CERTNAMES is withdrawn.
1268 JH/05 The value of the tls_verify_certificates smtp transport and main options
1269 default to the word "system" to access the system default CA bundle.
1270 For GnuTLS, only version 3.0.20 or later.
1272 JH/06 Verification of the server certificate for a TLS connection is now tried
1273 (but not required) by default. The verification status is now logged by
1274 default, for both outbound TLS and client-certificate supplying inbound
1277 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
1280 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
1281 Status Notification (bounce) messages are now MIME format per RFC 3464.
1282 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
1283 under the control of the dsn_advertise_hosts option, and routers may
1284 have a dsn_lasthop option.
1286 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
1287 default, modifiable by a malware= option. The list separator for
1288 the options can now be changed in the usual way. Bug 68.
1290 JH/10 The smtp_receive_timeout main option is now expanded before use.
1292 JH/11 The incoming_interface log option now also enables logging of the
1293 local interface on delivery outgoing connections.
1295 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
1296 if the interface and destination host and port all match.
1298 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
1301 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
1302 Patch from Andrew Lewis.
1304 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
1305 now supports optional time-restrictions, weighting, and priority
1306 modifiers per server. Patch originally by <rommer@active.by>.
1308 JH/16 The spamd_address main option now supports a mixed list of local
1309 and remote servers. Remote servers can be IPv6 addresses, and
1310 specify a port-range.
1312 JH/17 Bug 68: The spamd_address main option now supports an optional
1313 timeout value per server.
1315 JH/18 Bug 1581: Router and transport options headers_add/remove can
1316 now have the list separator specified.
1318 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
1321 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
1324 JH/21 Support for the A6 type of dns record is withdrawn.
1326 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
1327 rather than the verbs used.
1329 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
1330 from 255 to 1024 chars.
1332 JH/24 Verification callouts now attempt to use TLS by default.
1334 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
1335 are generic router options now. The defaults didn't change.
1337 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
1338 Original patch from Alexander Shikoff, worked over by JH.
1340 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
1341 files if ZCAT_COMMAND is not executable.
1343 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
1345 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
1347 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
1348 Normally benign, it bites when the pair was led to by a CNAME;
1349 modern usage is to not canonicalize the domain to a CNAME target
1350 (and we were inconsistent anyway for A-only vs AAAA+A).
1352 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
1354 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
1355 when evaluating $sender_host_dnssec.
1357 JH/31 Check the HELO verification lookup for DNSSEC, adding new
1358 $sender_helo_dnssec variable.
1360 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
1362 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
1364 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
1366 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
1367 documented as working, but never had. Support all but $spam_report.
1369 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
1370 added for tls authenticator.
1372 HS/03 Add perl_taintmode main config option
1378 TL/01 When running the test suite, the README says that variables such as
1379 no_msglog_check are global and can be placed anywhere in a specific
1380 test's script, however it was observed that placement needed to be near
1381 the beginning for it to behave that way. Changed the runtest perl
1382 script to read through the entire script once to detect and set these
1383 variables, reset to the beginning of the script, and then run through
1384 the script parsing/test process like normal.
1386 TL/02 The BSD's have an arc4random API. One of the functions to induce
1387 adding randomness was arc4random_stir(), but it has been removed in
1388 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
1389 function when detected.
1391 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
1392 cause callback expansion.
1394 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
1395 syntax errors in an expansion can be treated as a string instead of
1396 logging or causing an error, due to the internal use of bool_lax
1397 instead of bool when processing it.
1399 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
1400 server certificates when making smtp deliveries.
1402 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
1404 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
1406 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
1408 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
1409 Merged patch from Sebastian Wiedenroth.
1411 JH/05 Fix results-pipe from transport process. Several recipients, combined
1412 with certificate use, exposed issues where response data items split
1413 over buffer boundaries were not parsed properly. This eventually
1414 resulted in duplicates being sent. This issue only became common enough
1415 to notice due to the introduction of connection certificate information,
1416 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1418 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1419 size buffer was used, resulting in syntax errors when an expansion
1422 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1423 version 3.3.6 or later.
1425 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1426 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1427 both become "event_action", the variables become $event_name, $event_data
1428 and $event_defer_errno. There is a new variable $verify_mode, usable in
1429 routers, transports and related events. The tls:cert event is now also
1430 raised for inbound connections, if the main configuration event_action
1433 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1434 early OCSP support, but no stapling (appears to be less than 1.0.0).
1436 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1437 server certificate names available under the smtp transport option
1438 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1441 JH/10 Time-related extraction expansions from certificates now use the main
1442 option "timezone" setting for output formatting, and are consistent
1443 between OpenSSL and GnuTLS compilations. Bug 1541.
1445 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1446 encoded parameter in the incoming message. Bug 1558.
1448 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1449 include certificate info, eximon was claiming there were spoolfile
1452 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1454 JH/14 Log delivery-related information more consistently, using the sequence
1455 "H=<name> [<ip>]" wherever possible.
1457 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1458 are problematic for Debian distribution, omit them from the release
1461 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1463 JH/16 Fix string representation of time values on 64bit time_t architectures.
1466 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1467 output list separator was used.
1472 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1473 checkers that were complaining about end of non-void function with no
1476 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1477 This was a regression introduced in 4.83 by another bugfix.
1479 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1481 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1482 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1488 TF/01 Correctly close the server side of TLS when forking for delivery.
1490 When a message was received over SMTP with TLS, Exim failed to clear up
1491 the incoming connection properly after forking off the child process to
1492 deliver the message. In some situations the subsequent outgoing
1493 delivery connection happened to have the same fd number as the incoming
1494 connection previously had. Exim would try to use TLS and fail, logging
1495 a "Bad file descriptor" error.
1497 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1498 utilities have not been installed.
1500 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1501 temporary space as the ACL may create new global variables.
1503 TL/01 LDAP support uses per connection or global context settings, depending
1504 upon the detected version of the libraries at build time.
1506 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1507 to extract and use the src ip:port in logging and expansions as if it
1508 were a direct connection from the outside internet. PPv2 support was
1509 updated based on HAProxy spec change in May 2014.
1511 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1513 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1514 Properly escape header and check for NULL return.
1516 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1519 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1521 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1522 characters in header names, implemented as a verify condition.
1523 Contributed by Michael Fischer v. Mollard.
1525 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1526 results permerror and temperror. Previous values are deprecated but
1527 still accepted. In a future release, err_perm and err_temp will be
1528 completely removed, which will be a backward incompatibility if the
1529 ACL tests for either of these two old results. Patch contributed by
1530 user bes-internal on the mailing list.
1532 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1534 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1535 selectors, in both main and reject logs.
1537 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1540 JH/07 Add malware type "sock" for talking to simple daemon.
1542 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1544 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1545 routers/transports under cutthrough routing.
1547 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1548 numbers. Touch up "bool" conditional to keep the same definition.
1550 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1552 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1554 JH/12 Expand items in router/transport headers_add or headers_remove lists
1555 individually rather than the list as a whole. Bug 1452.
1557 Required for reasonable handling of multiple headers_ options when
1558 they may be empty; requires that headers_remove items with embedded
1559 colons must have them doubled (or the list-separator changed).
1561 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1562 view the policy declared in the DMARC record. Currently, $dmarc_status
1563 is a combined value of both the record presence and the result of the
1566 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1568 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1569 dnslookup router and the smtp transport (applying to the forward
1572 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1573 of ldap servers used for a specific lookup. Patch provided by Heiko
1576 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1577 New variable $lookup_dnssec_authenticated for observability.
1579 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1580 Patch submitted by Lars Timman.
1582 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1584 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1585 Requires trusted mode and valid format message id, aborts otherwise.
1586 Patch contributed by Heiko Schlichting.
1588 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1589 certextract with support for various fields. Bug 1358.
1591 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1592 is requested by default, modifiable by smtp transport option
1595 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1596 operate on certificate variables to give certificate fingerprints
1597 Also new ${sha256:cert_variable}.
1599 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1601 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1602 Christian Aistleitner.
1604 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1606 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1607 file. Patch from Wolfgang Breyha.
1609 JH/25 Expand the coverage of the delivery $host and $host_address to
1610 client authenticators run in verify callout. Bug 1476.
1612 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1613 align with daemon_smtp_ports. Bug 72.
1615 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1616 support and error reporting did not work properly.
1618 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1619 and is readable. Patch from Andrew Colin Kissa.
1621 TL/14 Enhance documentation of ${run expansion and how it parses the
1622 commandline after expansion, particularly in the case when an
1623 unquoted variable expansion results in an empty value.
1625 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1627 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1628 by renaming away from it. Addresses GH issue 15, reported by
1631 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1633 TL/15 SECURITY: prevent double expansion in math comparison functions
1634 (can expand unsanitized data). Not remotely exploitable.
1641 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1643 PP/02 Make -n do something, by making it not do something.
1644 When combined with -bP, the name of an option is not output.
1646 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1649 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1650 $sender_host_name and config options to manage this, and basic check
1653 PP/05 DSCP support for outbound connections and control modifier for inbound.
1655 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1656 (Only plugin which currently uses this is kerberos4, which nobody should
1657 be using, but we should make it available and other future plugins might
1658 conceivably use it, even though it would break NAT; stuff *should* be
1659 using channel bindings instead).
1661 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1662 name; added for Sendmail compatibility; requires admin caller.
1663 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1664 just ignore it); requires trusted caller.
1665 Also parse but ignore: -Ac -Am -X<logfile>
1668 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1670 TL/02 Add +smtp_confirmation as a default logging option.
1672 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1673 Patch by Magnus Holmgren from 2007-02-20.
1675 TL/04 Bugzilla 1281 - Spec typo.
1676 Bugzilla 1283 - Spec typo.
1677 Bugzilla 1290 - Spec grammar fixes.
1679 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1681 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1683 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1684 mailing list by by Dmitry Isaikin.
1686 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1688 JH/02 Support "G" suffix to numbers in ${if comparisons.
1690 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1692 NM/01 Bugzilla 1197 - Spec typo
1693 Bugzilla 1196 - Spec examples corrections
1695 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1697 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1698 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1701 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1702 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1704 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1705 "acl {{name}{arg}...}", and optional args on acl condition
1708 JH/05 Permit multiple router/transport headers_add/remove lines.
1710 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1712 JH/07 Avoid using a waiting database for a single-message-only transport.
1713 Performance patch from Paul Fisher. Bugzilla 1262.
1715 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1718 JH/09 Add $headers_added variable, with content from use of ACL modifier
1719 add_header (but not yet added to the message). Bugzilla 199.
1721 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1722 Pulled from Bugzilla 817 by Wolfgang Breyha.
1724 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1726 (nb: this is the same fix as in Exim 4.80.1)
1728 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1731 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1734 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1736 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1737 advertise SMTP AUTH mechanism to us, instead of a generic
1738 protocol violation error. Also, make Exim more robust to bad
1739 data from the Dovecot auth socket.
1741 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1743 When a queue runner is handling a message, Exim first routes the
1744 recipient addresses, during which it prunes them based on the retry
1745 hints database. After that it attempts to deliver the message to
1746 any remaining recipients. It then updates the hints database using
1749 So if a recipient address works intermittently, it can get repeatedly
1750 deferred at routing time. The retry hints record remains fresh so the
1751 address never reaches the final cutoff time.
1753 This is a fairly common occurrence when a user is bumping up against
1754 their storage quota. Exim had some logic in its local delivery code
1755 to deal with this. However it did not apply to per-recipient defers
1756 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1758 This change adds a proper retry rule check during routing so that the
1759 final cutoff time is checked against the message's age. We only do
1760 this check if there is an address retry record and there is not a
1761 domain retry record; this implies that previous attempts to handle
1762 the address had the retry_use_local_parts option turned on. We use
1763 this as an approximation for the destination being like a local
1764 delivery, as in LMTP.
1766 I suspect this new check makes the old local delivery cutoff check
1767 redundant, but I have not verified this so I left the code in place.
1769 TF/02 Correct gecos expansion when From: is a prefix of the username.
1771 Test 0254 submits a message to Exim with the header
1775 When I ran the test suite under the user fanf2, Exim expanded
1776 the header to contain my full name, whereas it should have added
1777 a Resent-Sender: header. It erroneously treats any prefix of the
1778 username as equal to the username.
1780 This change corrects that bug.
1782 GF/01 DCC debug and logging tidyup
1783 Error conditions log to paniclog rather than rejectlog.
1784 Debug lines prefixed by "DCC: " to remove any ambiguity.
1786 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1788 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1789 Bug spotted by Jeremy Harris; was flawed since initial commit.
1790 Would have resulted in OCSP responses post-SNI triggering an Exim
1791 NULL dereference and crash.
1793 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1795 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1796 Bug detection, analysis and fix by Samuel Thibault.
1797 Bugzilla 1331, Debian bug #698092.
1799 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1801 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1802 Server implementation by Todd Lyons, client by JH.
1803 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1804 config variable "prdr_enable" controls whether the server
1805 advertises the facility. If the client requests PRDR a new
1806 acl_data_smtp_prdr ACL is called once for each recipient, after
1807 the body content is received and before the acl_smtp_data ACL.
1808 The client is controlled by both of: a hosts_try_prdr option
1809 on the smtp transport, and the server advertisement.
1810 Default client logging of deliveries and rejections involving
1811 PRDR are flagged with the string "PRDR".
1813 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1814 fclose(). Diagnosis by Todd Lyons.
1816 PP/17 Update configure.default to handle IPv6 localhost better.
1817 Patch by Alain Williams (plus minor tweaks).
1820 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1821 This is now consistent with GnuTLS, and is now documented: the
1822 previous undocumented portable approach to treating the option as
1823 unset was to force an expansion failure. That still works, and
1824 an empty string is now equivalent.
1826 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1827 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1828 not performing validation itself.
1830 PP/20 Added force_command boolean option to pipe transport.
1831 Patch from Nick Koston, of cPanel Inc.
1833 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1836 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1838 PP/21 Fix eximon continuous updating with timestamped log-files.
1839 Broken in a format-string cleanup in 4.80, missed when I repaired the
1840 other false fix of the same issue.
1841 Report and fix from Heiko Schlichting.
1844 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1845 Report from Prashanth Katuri.
1847 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1848 It's SecureTransport, so affects any MacOS clients which use the
1849 system-integrated TLS libraries, including email clients.
1851 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1852 using a MIME ACL for non-SMTP local injection.
1853 Report and assistance in diagnosis by Warren Baker.
1855 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1857 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1859 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1860 last id that failed so it may be referenced in subsequent ACL's.
1862 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1865 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1866 ldap library initialization, allowing self-signed CA's to be
1867 used. Also properly sets require_cert option later in code by
1868 using NULL (global ldap config) instead of ldap handle (per
1869 session). Bug diagnosis and testing by alxgomz.
1871 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1872 the src/util/ subdirectory.
1874 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1875 renamed to Transport Post Delivery Action by Jeremy Harris, as
1878 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1879 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1880 redis_servers = needs to be configured which will be used by the redis
1881 lookup. Patch from Warren Baker, of The Packet Hub.
1883 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1885 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1886 hostname or reverse DNS when processing a host list. Used suggestions
1887 from multiple comments on this bug.
1889 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1891 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1892 Missed a few lines, added it to make the runtest require no keyboard
1895 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1896 contains upper case chars. Make router use caseful_local_part.
1898 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1899 support when GnuTLS has been built with p11-kit.
1905 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1907 This, or similar/improved, will also be change PP/11 of 4.82.
1913 PP/01 Handle short writes when writing local log-files.
1914 In practice, only affects FreeBSD (8 onwards).
1915 Bugzilla 1053, with thanks to Dmitry Isaikin.
1917 NM/01 Bugzilla 949 - Documentation tweak
1919 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1922 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1924 PP/02 Implemented gsasl authenticator.
1926 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1928 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1929 `pkg-config foo` for cflags/libs.
1931 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1932 with rest of GSASL and with heimdal_gssapi.
1934 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1935 `pkg-config foo` for cflags/libs for the TLS implementation.
1937 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1938 properties get this fed in as external SSF. A number of robustness
1939 and debugging improvements to the cyrus_sasl authenticator.
1941 PP/08 cyrus_sasl server now expands the server_realm option.
1943 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1944 Patch by Jeremy Harris.
1946 PP/10 Added dbmjz lookup type.
1948 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1950 PP/12 MAIL args handles TAB as well as SP, for better interop with
1951 non-compliant senders.
1952 Analysis and variant patch by Todd Lyons.
1954 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1955 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1956 Patch from Dirk Mueller <dmueller@suse.com>
1958 PP/13 tls_peerdn now print-escaped for spool files.
1959 Observed some $tls_peerdn in wild which contained \n, which resulted
1960 in spool file corruption.
1962 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1963 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1964 or write after TLS renegotiation, which otherwise led to messages
1967 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1968 as a tracking header (ie: a signed header comes before the signature).
1969 Patch from Wolfgang Breyha.
1971 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1972 comma-sep list; embedded commas doubled.
1974 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1976 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1978 Report and patch from Dmitry Banschikov.
1980 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1981 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1982 failures. We appear to now support TLS1.1+ with Exim.
1984 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1985 lets Exim select keys and certificates based upon TLS SNI from client.
1986 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1987 before an outbound SMTP session. New log_selector, +tls_sni.
1989 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1990 NULL dereference. Report and patch from Alun Jones.
1992 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1993 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1994 Not seeing resolver debug output on NetBSD, but suspect this is a
1995 resolver implementation change.
1997 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1998 Left warnings. Added "eximon gdb" invocation mode.
2000 PP/21 Defaulting "accept_8bitmime" to true, not false.
2002 PP/22 Added -bw for inetd wait mode support.
2004 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
2005 locate the relevant includes and libraries. Made this the default.
2007 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
2008 Bugzilla 1246, report and most of solution from Tomasz Kusy.
2010 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
2011 This may cause build issues on older platforms.
2013 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
2014 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
2015 gnutls_require_mac & gnutls_require_protocols (no longer supported).
2016 Added SNI support via GnuTLS too.
2017 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
2019 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
2021 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
2022 Applied second patch from Janne, implementing suggestion to default
2023 multiple-strings-in-record handling to match SPF spec.
2025 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
2027 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
2028 read-only, out of scope).
2029 Patch from Wolfgang Breyha, report from Stuart Northfield.
2031 PP/29 Fix three issues highlighted by clang analyser static analysis.
2032 Only crash-plausible issue would require the Cambridge-specific
2033 iplookup router and a misconfiguration.
2034 Report from Marcin Mirosław.
2036 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
2038 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
2039 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
2040 As part of this, removing so much warning spew let me fix some minor
2041 real issues in debug logging.
2043 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
2044 assignment on my part. Fixed.
2046 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
2047 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
2048 Janne Snabb (who went above and beyond: thank you).
2050 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
2051 string otherwise requires a connection and a bunch more work and it's
2052 relatively easy to get wrong. Should also expose TLS library linkage
2055 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
2056 64-bit ${eval} (JH/03).
2058 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
2059 GNU libc to support some of the 64-bit stuff, should not lead to
2060 conflicts. Defined before os.h is pulled in, so if a given platform
2061 needs to override this, it can.
2063 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
2064 protection layer was required, which is not implemented.
2065 Bugzilla 1254, patch from Wolfgang Breyha.
2067 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
2068 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
2069 tls_dhparam take prime identifiers. Also unbreak combination of
2070 OpenSSL+DH_params+TLSSNI.
2072 PP/39 Disable SSLv2 by default in OpenSSL support.
2078 PP/01 Solaris build fix for Oracle's LDAP libraries.
2079 Bugzilla 1109, patch from Stephen Usher.
2081 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
2083 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
2086 TF/02 Fix a couple more cases where we did not log the error message
2087 when unlink() failed. See also change 4.74-TF/03.
2089 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
2090 lock up or crash if it happened to be inside a call to libc when it
2091 got a SIGUSR1 from exiwhat.
2093 The SIGUSR1 handler appends the current process status to the process
2094 log which is later printed by exiwhat. It used to use the general
2095 purpose logging code to do this, but several functions it calls are
2096 not safe for signals.
2098 The new output code in the SIGUSR1 handler is specific to the process
2099 log, and simple enough that it's easy to inspect for signal safety.
2100 Removing some special cases also simplifies the general logging code.
2101 Removing the spurious timestamps from the process log simplifies
2104 TF/04 Improved ratelimit ACL condition.
2106 The /noupdate option has been deprecated in favour of /readonly which
2107 has clearer semantics. The /leaky, /strict, and /readonly update modes
2108 are mutually exclusive. The update mode is no longer included in the
2109 database key; it just determines when the database is updated. (This
2110 means that when you upgrade Exim will forget old rate measurements.)
2112 Exim now checks that the per_* options are used with an update mode that
2113 makes sense for the current ACL. For example, when Exim is processing a
2114 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
2115 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
2116 must specify per_mail/readonly. If you omit the update mode it defaults to
2117 /leaky where that makes sense (as before) or /readonly where required.
2119 The /noupdate option is now undocumented but still supported for
2120 backwards compatibility. It is equivalent to /readonly except that in
2121 ACLs where /readonly is required you may specify /leaky/noupdate or
2122 /strict/noupdate which are treated the same as /readonly.
2124 A useful new feature is the /count= option. This is a generalization
2125 of the per_byte option, so that you can measure the throughput of other
2126 aggregate values. For example, the per_byte option is now equivalent
2127 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
2129 The per_rcpt option has been generalized using the /count= mechanism
2130 (though it's more complicated than the per_byte equivalence). When it is
2131 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
2132 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
2133 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
2134 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
2135 Note that using per_rcpt with a non-readonly update mode in more than
2136 one ACL will cause the recipients to be double-counted. (The per_mail
2137 and per_byte options don't have this problem.)
2139 The handling of very low rates has changed slightly. If the computed rate
2140 is less than the event's count (usually one) then this event is the first
2141 after a long gap. In this case the rate is set to the same as this event's
2142 count, so that the first message of a spam run is counted properly.
2144 The major new feature is a mechanism for counting the rate of unique
2145 events. The new per_addr option counts the number of different
2146 recipients that someone has sent messages to in the last time period. It
2147 behaves like per_rcpt if all the recipient addresses are different, but
2148 duplicate recipient addresses do not increase the measured rate. Like
2149 the /count= option this is a general mechanism, so the per_addr option
2150 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
2151 example, measure the rate that a client uses different sender addresses
2152 with the options per_mail/unique=$sender_address. There are further
2153 details in the main documentation.
2155 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
2157 TF/06 Removed a few PCRE remnants.
2159 TF/07 Automatically extract Exim's version number from tags in the git
2160 repository when doing development or release builds.
2162 PP/02 Raise smtp_cmd_buffer_size to 16kB.
2163 Bugzilla 879. Patch from Paul Fisher.
2165 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
2166 Heavily based on revision 40f9a89a from Simon Arlott's tree.
2169 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
2171 PP/05 Variable $av_failed, true if the AV scanner deferred.
2172 Bugzilla 1078. Patch from John Horne.
2174 PP/06 Stop make process more reliably on build failure.
2175 Bugzilla 1087. Patch from Heiko Schlittermann.
2177 PP/07 Make maildir_use_size_file an _expandable_ boolean.
2178 Bugzilla 1089. Patch from Heiko Schlittermann.
2180 PP/08 Handle ${run} returning more data than OS pipe buffer size.
2181 Bugzilla 1131. Patch from Holger Weiß.
2183 PP/09 Handle IPv6 addresses with SPF.
2184 Bugzilla 860. Patch from Wolfgang Breyha.
2186 PP/10 GnuTLS: support TLS 1.2 & 1.1.
2188 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
2191 PP/11 match_* no longer expand right-hand-side by default.
2192 New compile-time build option, EXPAND_LISTMATCH_RHS.
2193 New expansion conditions, "inlist", "inlisti".
2195 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
2197 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
2199 PP/14 fix log_write() format string regression from TF/03.
2200 Bugzilla 1152. Patch from Dmitry Isaikin.
2206 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
2208 PP/02 Harmonised TLS library version reporting; only show if debugging.
2209 Layout now matches that introduced for other libraries in 4.74 PP/03.
2211 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
2213 PP/04 New "dns_use_edns0" global option.
2215 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
2218 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
2219 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
2221 TK/01 Updated PolarSSL code to 0.14.2.
2222 Bugzilla 1097. Patch from Andreas Metzler.
2224 PP/07 Catch divide-by-zero in ${eval:...}.
2225 Fixes bugzilla 1102.
2227 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
2230 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
2231 format-string attack -- SECURITY: remote arbitrary code execution.
2233 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
2234 time unintentionally subject to list matching rules, letting the header
2235 cause arbitrary Exim lookups (of items which can occur in lists, *not*
2236 arbitrary string expansion). This allowed for information disclosure.
2238 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
2239 INT_MIN/-1 -- value coerced to INT_MAX.
2245 NM/01 Workaround for PCRE version dependency in version reporting
2248 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
2249 This fixes portability to compilers other than gcc, notably
2250 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
2252 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
2253 makefiles for portability to HP-UX and POSIX correctness.
2255 PP/01 Permit LOOKUP_foo enabling on the make command-line.
2256 Also via indented variable definition in the Makefile.
2257 (Debugging by Oliver Heesakkers).
2259 PP/02 Restore caching of spamd results with expanded spamd_address.
2260 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
2262 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
2263 Improves build reliability. Fix from: Frank Elsner
2265 NM/02 Fix wide character breakage in the rfc2047 coding
2266 Fixes bug 1064. Patch from Andrey N. Oktyabrski
2268 NM/03 Allow underscore in dnslist lookups
2269 Fixes bug 1026. Patch from Graeme Fowler
2271 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
2272 Code patches from Adam Ciarcinski of NetBSD.
2274 NM/04 Fixed exiqgrep to cope with mailq missing size issue
2277 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
2278 is logged, to avoid truncation. Patch from John Horne.
2280 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
2281 Patch from Jakob Hirsch.
2283 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
2284 SQL string expansion failure details.
2285 Patch from Andrey Oktyabrski.
2287 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
2288 Patch from Simon Arlott.
2290 PP/09 New lookups functionality failed to compile on old gcc which rejects
2291 extern declarations in function scope.
2292 Patch from Oliver Fleischmann
2294 PP/10 Use sig_atomic_t for flags set from signal handlers.
2295 Check getgroups() return and improve debugging.
2296 Fixed developed for diagnosis in bug 927 (which turned out to be
2299 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
2300 Patch from Mark Zealey.
2302 PP/12 Bugzilla 1056: Improved spamd server selection.
2303 Patch from Mark Zealey.
2305 PP/13 Bugzilla 1086: Deal with maildir quota file races.
2306 Based on patch from Heiko Schlittermann.
2308 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
2309 Patch from Uwe Doering, sign-off by Michael Haardt.
2311 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
2312 variable declaration deep within a block. Bug and patch from
2315 PP/15 lookups-Makefile IRIX compatibility coercion.
2317 PP/16 Make DISABLE_DKIM build knob functional.
2319 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
2320 Patch by Simon Arlott
2322 TF/03 Fix valgrind.h portability to C89 compilers that do not support
2323 variable argument macros. Our copy now differs from upstream.
2329 TF/01 Failure to get a lock on a hints database can have serious
2330 consequences so log it to the panic log.
2332 TF/02 Log LMTP confirmation messages in the same way as SMTP,
2333 controlled using the smtp_confirmation log selector.
2335 TF/03 Include the error message when we fail to unlink a spool file.
2337 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
2338 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
2339 for maintaining out-of-tree patches for some time.
2341 PP/01 Bugzilla 139: Documentation and portability issues.
2342 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
2343 Handle per-OS dynamic-module compilation flags.
2345 PP/02 Let /dev/null have normal permissions.
2346 The 4.73 fixes were a little too stringent and complained about the
2347 permissions on /dev/null. Exempt it from some checks.
2348 Reported by Andreas M. Kirchwitz.
2350 PP/03 Report version information for many libraries, including
2351 Exim version information for dynamically loaded libraries. Created
2352 version.h, now support a version extension string for distributors
2353 who patch heavily. Dynamic module ABI change.
2355 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
2356 privilege escalation vulnerability whereby the Exim run-time user
2357 can cause root to append content of the attacker's choosing to
2360 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
2363 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
2364 If dropping privileges for untrusted macros, we disabled normal logging
2365 on the basis that it would fail; for the Exim run-time user, this is not
2366 the case, and it resulted in successful deliveries going unlogged.
2367 Fixed. Reported by Andreas Metzler.
2373 PP/01 Date: & Message-Id: revert to normally being appended to a message,
2374 only prepend for the Resent-* case. Fixes regression introduced in
2375 Exim 4.70 by NM/22 for Bugzilla 607.
2377 PP/02 Include check_rfc2047_length in configure.default because we're seeing
2378 increasing numbers of administrators be bitten by this.
2380 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
2382 PP/03 Bugzilla 994: added openssl_options main configuration option.
2384 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
2386 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
2388 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
2390 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
2391 without a peer certificate, leading to a segfault because of an
2392 assumption that peers always have certificates. Be a little more
2393 paranoid. Problem reported by Martin Tscholak.
2395 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
2396 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
2397 NB: ClamAV planning to remove STREAM in "middle of 2010".
2398 CL also introduces -bmalware, various -d+acl logging additions and
2399 more caution in buffer sizes.
2401 PP/09 Implemented reverse_ip expansion operator.
2403 PP/10 Bugzilla 937: provide a "debug" ACL control.
2405 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
2407 PP/12 Bugzilla 973: Implement --version.
2409 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2411 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2413 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2415 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2416 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2417 ignore trailing whitespace.
2419 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2421 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2424 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2425 Notification from Dr Andrew Aitchison.
2427 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2428 ExtendedDetectionInfo response format.
2429 Notification from John Horne.
2431 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2434 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2435 XSL and documented dependency on system catalogs, with examples of how
2438 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2441 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2442 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2443 configuration file which is writeable by the Exim user or group.
2445 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2446 of configuration files to cover files specified with the -C option if
2447 they are going to be used with root privileges, not just the default
2450 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2451 option (effectively making it always true).
2453 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2454 files to be used while preserving root privileges.
2456 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2457 that rogue child processes cannot use them.
2459 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2460 run-time user, instead of root.
2462 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2463 Exim run-time user without dropping privileges.
2465 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2466 result string, instead of calling string_vformat() twice with the same
2469 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2470 for other users. Others should always drop root privileges if they use
2471 -C on the command line, even for a whitelisted configure file.
2473 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2475 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2481 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2482 $data_path, and $header_path variables; fixed documentation bugs and
2485 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2486 exipick to access non-standard spools, including the "frozen" queue
2489 NM/01 Bugzilla 965: Support mysql stored procedures.
2490 Patch from Alain Williams
2492 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2494 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2495 Patch from Andreas Metzler
2497 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2498 Patch from Kirill Miazine
2500 NM/05 Bugzilla 671: Added umask to procmail example.
2502 JJ/03 installed exipick 20100323.0, fixing doc bug
2504 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2505 directory. Notification and patch from Dan Rosenberg.
2507 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2509 TK/02 Improve log output when DKIM signing operation fails.
2511 MH/01 Treat the transport option dkim_domain as a colon separated
2512 list, not as a single string, and sign the message with each element,
2513 omitting multiple occurences of the same signer.
2515 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2516 Bugzilla 985, 986. Patch by Simon Arlott
2518 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2519 Patch by Simon Arlott
2521 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2522 MBX locking. Notification from Dan Rosenberg.
2528 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2530 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2532 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2534 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2536 NM/04 Bugzilla 915: Flag broken perl installation during build.
2542 TK/01 Added patch by Johannes Berg that expands the main option
2543 "spamd_address" if it starts with a dollar sign.
2545 TK/02 Write list of recipients to X-Envelope-Sender header when building
2546 the mbox-format spool file for content scanning (suggested by Jakob
2549 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2550 (http://www.dcc-servers.net/) support via dccifd. Activated by
2551 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2553 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2554 by Mark Daniel Reidel <mr@df.eu>.
2556 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2557 When building exim an external PCRE library is now needed -
2558 PCRE is a system library on the majority of modern systems.
2559 See entry on PCRE_LIBS in EDITME file.
2561 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2562 conversation. Added nologin parameter to request.
2563 Patch contributed by Kirill Miazine.
2565 TF/01 Do not log submission mode rewrites if they do not change the address.
2567 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2569 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2570 log files in place. Contributed by Roberto Lima.
2572 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2574 TF/03 Bugzilla 615: When checking the local_parts router precondition
2575 after a local_part_suffix or local_part_prefix option, Exim now
2576 does not use the address's named list lookup cache, since this
2577 contains cached lookups for the whole local part.
2579 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2580 Robert Millan. Documentation is in experimental-spec.txt.
2582 TF/04 Bugzilla 668: Fix parallel build (make -j).
2584 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2586 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2587 Patch provided by Jan Srzednicki.
2589 TF/05 Leading white space used to be stripped from $spam_report which
2590 wrecked the formatting. Now it is preserved.
2592 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2593 that they are available at delivery time.
2595 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2597 TF/08 TLS error reporting now respects the incoming_interface and
2598 incoming_port log selectors.
2600 TF/09 Produce a more useful error message if an SMTP transport's hosts
2601 setting expands to an empty string.
2603 NM/06 Bugzilla 744: EXPN did not work under TLS.
2604 Patch provided by Phil Pennock.
2606 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2607 Patch provided by Richard Godbee.
2609 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2610 acl_smtp_notquit, added index entry.
2612 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2613 Patch provided by Eugene Bujak.
2615 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2616 accept(). Patch provided by Maxim Dounin.
2618 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2619 Patch provided by Phil Pennock.
2621 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2623 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2624 Patch provided by Brad "anomie" Jorsch.
2626 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2627 Patch provided by Dean Brooks.
2629 TK/05 Add native DKIM support (does not depend on external libraries).
2631 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2632 Patch provided by Graeme Fowler.
2634 NM/16 Bugzilla 851: Documentation example syntax fix.
2636 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2638 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2641 NM/19 Bugzilla 745: TLS version reporting.
2642 Patch provided by Phil Pennock.
2644 NM/20 Bugzilla 167: bool: condition support.
2645 Patch provided by Phil Pennock.
2647 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2648 clients. Patch provided by Phil Pennock.
2650 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2651 Patch provided by Brad "anomie" Jorsch.
2653 NM/23 Bugzilla 687: Fix misparses in eximstats.
2654 Patch provided by Heiko Schlittermann.
2656 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2657 Patch provided by Heiko Schlittermann.
2659 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2660 plus update to original patch.
2662 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2664 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2665 Patch provided by David Brownlee.
2667 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2669 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2671 NM/30 Bugzilla 888: TLS documentation bugfixes.
2673 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2675 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2676 Unlike the original bugzilla I have changed all shell scripts in src tree.
2678 NM/33 Bugzilla 898: Transport filter timeout fix.
2679 Patch by Todd Rinaldo.
2681 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2682 Patch by Serge Demonchaux.
2684 NM/35 Bugzilla 39: Base64 decode bug fixes.
2685 Patch by Jakob Hirsch.
2687 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2689 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2691 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2693 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2699 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2700 ALT-N's libdkim that I have put here:
2701 http://duncanthrax.net/exim-experimental/
2703 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2704 were called 'true' and it seems that C99 defines that as a reserved
2705 keyword to be used with 'bool' variable types. That means you could
2706 not include C99-style headers which use bools without triggering
2707 build errors in sieve.c.
2709 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2710 as mailq or other aliases. Changed the --help handling significantly
2711 to do whats expected. exim_usage() emits usage/help information.
2713 SC/01 Added the -bylocaldomain option to eximstats.
2715 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2717 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2719 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2725 PH/01 Another patch from the Sieve maintainer.
2727 PH/02 When an IPv6 address is converted to a string for single-key lookup
2728 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2729 dots are used instead of colons so that keys in lsearch files need not
2730 contain colons. This was done some time before quoting was made available
2731 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2732 (notated using the quote facility) so as to distinguish them from IPv4
2733 keys. This meant that lookups for IP addresses in host lists did not work
2734 for iplsearch lookups.
2736 This has been fixed by arranging for IPv6 addresses to be expressed with
2737 colons if the lookup type is iplsearch. This is not incompatible, because
2738 previously such lookups could never work.
2740 The situation is now rather anomalous, since one *can* have colons in
2741 ordinary lsearch keys. However, making the change in all cases is
2742 incompatible and would probably break a number of configurations.
2744 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2747 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2748 conversion specification without a maximum field width, thereby enabling
2749 a rogue spamd server to cause a buffer overflow. While nobody in their
2750 right mind would setup Exim to query an untrusted spamd server, an
2751 attacker that gains access to a server running spamd could potentially
2752 exploit this vulnerability to run arbitrary code as the Exim user.
2754 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2755 $primary_hostname instead of what libspf2 thinks the hosts name is.
2757 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2758 a directory entry by the name of the lookup key. Previously, if a
2759 symlink pointed to a non-existing file or a file in a directory that
2760 Exim lacked permissions to read, a lookup for a key matching that
2761 symlink would fail. Now it is enough that a matching directory entry
2762 exists, symlink or not. (Bugzilla 503.)
2764 PH/03 The body_linecount and body_zerocount variables are now exported in the
2767 PH/04 Added the $dnslist_matched variable.
2769 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2770 This means they are set thereafter only if the connection becomes
2773 PH/06 Added the client_condition to authenticators so that some can be skipped
2774 by clients under certain conditions.
2776 PH/07 The error message for a badly-placed control=no_multiline_responses left
2777 "_responses" off the end of the name.
2779 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2781 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2782 (without spaces) instead of just copying the configuration text.
2784 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2786 PH/11 Added $max_received_linelength.
2788 PH/12 Added +ignore_defer and +include_defer to host lists.
2790 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2791 way in which PCRE > 7.0 is built.
2793 PH/14 Implemented queue_only_load_latch.
2795 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2796 MAIL command. The effect was to mangle the value on 64-bit systems.
2798 PH/16 Another patch from the Sieve maintainer.
2800 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2802 PH/18 If a system quota error occurred while trying to create the file for
2803 a maildir delivery, the message "Mailbox is full" was not appended to the
2804 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2805 only to a quota excession during the actual writing of the file.
2807 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2808 characters?) which causes problems in log lines. The DN values are now
2809 passed through string_printing() before being added to log lines.
2811 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2812 and InterBase are left for another time.)
2814 PH/21 Added message_body_newlines option.
2816 PH/22 Guard against possible overflow in moan_check_errorcopy().
2818 PH/23 POSIX allows open() to be a macro; guard against that.
2820 PH/24 If the recipient of an error message contained an @ in the local part
2821 (suitably quoted, of course), incorrect values were put in $domain and
2822 $local_part during the evaluation of errors_copy.
2828 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2829 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2832 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2833 issue a MAIL command.
2835 PH/02 In an ACL statement such as
2837 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2839 if a client was not listed at all, or was listed with a value other than
2840 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2841 the condition was not true (as it should be), so access was not denied.
2842 The bug was that the ! inversion was incorrectly passed on to the second
2843 item. This has been fixed.
2845 PH/03 Added additional dnslists conditions == and =& which are different from
2846 = and & when the dns lookup returns more than one IP address.
2848 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2849 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2851 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2852 FSYNC, which compiles an option called disable_fsync that allows for
2853 bypassing fsync(). The documentation is heavily laced with warnings.
2855 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2857 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2858 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2859 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2860 including adding "make clean"; (3) Added -fPIC when compiling the test
2861 dynamically loaded module, to get rid of a warning.
2863 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2864 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2865 = 0s. The bug is otherwise harmless.
2867 PH/07 There was a bug in the dovecot authenticator such that the value of
2868 $auth1 could be overwritten, and so not correctly preserved, after a
2869 successful authentication. This usually meant that the value preserved by
2870 the server_setid option was incorrect.
2872 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2874 PH/09 Installed PCRE release 7.0.
2876 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2877 run for batched SMTP input. It is now run at the start of every message
2878 in the batch. While fixing this I discovered that the process information
2879 (output by running exiwhat) was not always getting set for -bs and -bS
2880 input. This is fixed, and it now also says "batched" for BSMTP.
2882 PH/11 Added control=no_pipelining.
2884 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2885 patch, slightly modified), and move the expansion of helo_data till after
2886 the connection is made in the smtp transport (so it can use these
2889 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2891 PH/14 Added log_selector = +pid.
2893 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2895 PH/16 Add ${if forany and ${if forall.
2897 PH/17 Added dsn_from option to vary the From: line in DSNs.
2899 PH/18 Flush SMTP output before performing a callout, unless control =
2900 no_callout_flush is set.
2902 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2903 was true (the default) a successful delivery failed to delete the retry
2904 item, thus causing premature timeout of the address. The bug is now
2907 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2909 PH/21 Long custom messages for fakedefer and fakereject are now split up
2910 into multiline responses in the same way that messages for "deny" and
2911 other ACL rejections are.
2913 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2914 with slight modification.
2916 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2917 draft, changing the syntax and factoring some duplicate code".
2919 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2920 for deliveries of the second and subsequent messages over the same SMTP
2923 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2924 ${reduce, with only minor "tidies".
2926 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2928 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2929 expansion side effects.
2931 PH/27 When a message times out after an over-quota error from an Exim-imposed
2932 quota, the bounce message says "mailbox is full". This message was not
2933 being given when it was a system quota that was exceeded. It now should
2936 MH/03 Made $recipients available in local_scan(). local_scan() already has
2937 better access to the recipient list through recipients_list[], but
2938 $recipients can be useful in postmaster-provided expansion strings.
2940 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2941 in the case of a MAIL command with additional options following the
2942 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2943 were accidentally chopped off.
2945 PH/29 SMTP synchronization checks are implemented when a command is read -
2946 there is a check that no more input is waiting when there shouldn't be
2947 any. However, for some commands, a delay in an ACL can mean that it is
2948 some time before the response is written. In this time, more input might
2949 arrive, invalidly. So now there are extra checks after an ACL has run for
2950 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2951 pipelining has not been advertised.
2953 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2955 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2956 returned in the SMTP response when smtp_return_error_details was false.
2957 This has been fixed.
2959 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2960 instead of the C I/O that was originally supplied, because problems were
2961 reported on Solaris.
2963 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2964 Exim which did not show up earlier: it was assuming that a call to
2965 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2966 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2967 was a macro that became an assignment, so it seemed to work. This has
2968 changed to a proper function call with a void return, hence the compile
2969 error. Exim's code has been fixed.
2971 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2974 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2976 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2978 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2979 because it turns out to be too big or there is a timeout) there is no
2980 "Completed" line in the log. When some messages of this type were
2981 selected by exigrep, they were listed as "not completed". Others were
2982 picked up by some special patterns. I have improved the selection
2983 criteria to be more general.
2985 PH/38 The host_find_failed option in the manualroute router can now be set
2986 to "ignore", to completely ignore a host whose IP address cannot be
2987 found. If all hosts are ignored, the behaviour is controlled by the new
2988 host_all_ignored option.
2990 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2991 homing or because of multiple MX records with /mx) generated more than
2992 one IP address, and the following item turned out to be the local host,
2993 all the secondary addresses of the first item were incorrectly removed
2994 from the list, along with the local host and any following hosts (which
2995 is what is supposed to happen).
2997 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2998 whoever called Exim into the -H file. In the case of the daemon it was
2999 behaving confusingly. When first started, it used values for whoever
3000 started the daemon, but after a SIGHUP it used the Exim user (because it
3001 calls itself on a restart). I have changed the code so that it now always
3004 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
3005 message are rejected with the same error (e.g. no authentication or bad
3006 sender address), and a DATA command is nevertheless sent (as can happen
3007 with PIPELINING or a stupid MUA), the error message that was given to the
3008 RCPT commands is included in the rejection of the DATA command. This is
3009 intended to be helpful for MUAs that show only the final error to their
3012 PH/42 Another patch from the Sieve maintainer.
3014 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
3015 Eximstats - Fixed some broken HTML links and added missing column headers
3017 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
3018 columns for Rejects, Temp Rejects, Ham, and Spam rows.
3020 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
3022 PH/43 Yet another patch from the Sieve maintainer.
3024 PH/44 I found a way to check for a TCP/IP connection going away before sending
3025 the response to the final '.' that terminates a message, but only in the
3026 case where the client has not sent further data following the '.'
3027 (unfortunately, this is allowed). However, in many cases there won't be
3028 any further data because there won't be any more messages to send. A call
3029 to select() can be used: if it shows that the input is "ready", there is
3030 either input waiting, or the socket has been closed. An attempt to read
3031 the next input character can distinguish the two cases. Previously, Exim
3032 would have sent an OK response which the client would never have see.
3033 This could lead to message repetition. This fix should cure that, at
3034 least in a lot of common cases.
3036 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
3037 advertised in response to EHLO.
3043 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
3044 fixed by 4.65/MH/01 (is this a record?) are fixed:
3046 (i) An empty string was always treated as zero by the numeric comparison
3047 operators. This behaviour has been restored.
3049 (ii) It is documented that the numeric comparison operators always treat
3050 their arguments as decimal numbers. This was broken in that numbers
3051 starting with 0 were being interpreted as octal.
3053 While fixing these problems I realized that there was another issue that
3054 hadn't been noticed. Values of message_size_limit (both the global option
3055 and the transport option) were treated as octal if they started with 0.
3056 The documentation was vague. These values are now always treated as
3057 decimal, and I will make that clear in the documentation.
3063 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
3064 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
3067 MH/01 Don't check that the operands of numeric comparison operators are
3068 integers when their expansion is in "skipping" mode (fixes bug
3069 introduced by 4.64-PH/07).
3071 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
3072 child addresses, Exim now panics and dies. Previously, because the count
3073 is held in a short int, deliveries were likely to be lost. As such a
3074 large number of recipients for a single message is ridiculous
3075 (performance will be very, very poor), I have chosen to impose a limit
3076 rather than extend the field.
3082 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
3083 leftover -K file (the existence of which was triggered by #402).
3084 While we were at it, introduced process PID as part of the -K
3085 filename. This should rule out race conditions when creating
3088 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
3089 processing considerably. Previous code took too long for large mails,
3090 triggering a timeout which in turn triggers #401.
3092 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
3093 in the DK code in transports.c. sendfile() is not really portable,
3094 hence the _LINUX specificness.
3096 TF/01 In the add_headers option to the mail command in an Exim filter,
3097 there was a bug that Exim would claim a syntax error in any
3098 header after the first one which had an odd number of characters
3101 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
3102 callout verification, Exim cached a "reject" for the entire domain. This
3103 is correct for most verifications, but it is not correct for a recipient
3104 verification with use_sender or use_postmaster set, because in that case
3105 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
3106 case of MAIL FROM:<> rejection from other early rejections (e.g.
3107 rejection of HELO). When verifying a recipient using a non-null MAIL
3108 address, the cache is ignored if it shows MAIL FROM:<> rejection.
3109 Whatever the result of the callout, the value of the domain cache is
3110 left unchanged (for any other kind of callout, getting as far as trying
3111 RCPT means that the domain itself is ok).
3113 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
3116 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
3117 manifest itself as EPIPE rather than ECONNECT. When tidying away a
3118 session, the daemon ignores ECONNECT errors and logs others; it now
3119 ignores EPIPE as well.
3121 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
3122 (quoted-printable decoding).
3124 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
3125 later the small subsequent patch to fix an introduced bug.
3127 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
3129 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
3131 PH/08 An error is now given if message_size_limit is specified negative.
3133 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
3134 to be given (somewhat) arbitrary names.
3136 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
3139 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
3140 miscellaneous code fixes
3142 PH/10 Added the log_reject_target ACL modifier to specify where to log
3145 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
3146 hostname. This is wrong, because it relates to the incoming message (and
3147 probably the interface on which it is arriving) and not to the outgoing
3148 callout (which could be using a different interface). This has been
3149 changed to use the value of the helo_data option from the smtp transport
3150 instead - this is what is used when a message is actually being sent. If
3151 there is no remote transport (possible with a router that sets up host
3152 addresses), $smtp_active_hostname is used.
3154 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
3155 tweaks were necessary in order to get it to work (see also 21 below):
3156 (a) The code assumed that strncpy() returns a negative number on buffer
3157 overflow, which isn't the case. Replaced with Exim's string_format()
3159 (b) There were several signed/unsigned issues. I just did the minimum
3160 hacking in of casts. There is scope for a larger refactoring.
3161 (c) The code used strcasecmp() which is not a standard C function.
3162 Replaced with Exim's strcmpic() function.
3163 (d) The code set only $1; it now sets $auth1 as well.
3164 (e) A simple test gave the error "authentication client didn't specify
3165 service in request". It would seem that Dovecot has changed its
3166 interface. Fortunately there's a specification; I followed it and
3167 changed what the client sends and it appears to be working now.
3169 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
3172 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
3173 address A is aliased to B and C, where B exists and C does not. Without
3174 -v the output is "A verified" because verification stops after a
3175 successful redirection if more than one address is generated. However,
3176 with -v the child addresses are also verified. Exim was outputting "A
3177 failed to verify" and then showing the successful verification for C,
3178 with its parentage. It now outputs "B failed to verify", showing B's
3179 parentage before showing the successful verification of C.
3181 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
3182 look up a TXT record in a specific list after matching in a combined
3185 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
3186 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
3187 they consult the DNS. I had assumed they would set it the way they
3188 wanted; and indeed my experiments on Linux seem to show that in some
3189 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
3190 To be on the safe side, however, I have now made the interface to
3191 host_find_byname() similar to host_find_bydns(), with an argument
3192 containing the DNS resolver options. The host_find_byname() function now
3193 sets these options at its start, just as host_find_bydns() does. The smtp
3194 transport options dns_qualify_single and dns_search_parents are passed to
3195 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
3196 of host_find_byname() use the default settings of RES_DEFNAMES
3197 (qualify_single) but not RES_DNSRCH (search_parents).
3199 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
3200 spool_read_header() do less string testing, by means of a preliminary
3201 switch on the second character of optional "-foo" lines. (This is
3202 overdue, caused by the large number of possibilities that now exist.
3203 Originally there were few.) While I was there, I also converted the
3204 str(n)cmp tests so they don't re-test the leading "-" and the first
3205 character, in the hope this might squeeze out yet more improvement.
3207 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
3208 flag allowing group syntax was set by the header_syntax check but not
3209 turned off, possible causing trouble later; (2) The flag was not being
3210 set at all for the header_verify test, causing "group"-style headers to
3211 be rejected. I have now set it in this case, and also caused header_
3212 verify to ignore an empty address taken from a group. While doing this, I
3213 came across some other cases where the code for allowing group syntax
3214 while scanning a header line wasn't quite right (mostly, not resetting
3215 the flag correctly in the right place). These bugs could have caused
3216 trouble for malformed header lines. I hope it is now all correct.
3218 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
3219 with the "reply" argument non-NULL. The code, however (which originally
3220 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
3221 but it didn't always do it. This confused somebody who was copying the
3222 code for some other use. I have removed all the tests.
3224 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
3225 feature that was used to support insecure browsers during the U.S. crypto
3226 embargo. It requires special client support, and Exim is probably the
3227 only MTA that supported it -- and would never use it because real RSA is
3228 always available. This code has been removed, because it had the bad
3229 effect of slowing Exim down by computing (never used) parameters for the
3230 RSA_EXPORT functionality.
3232 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
3233 authenticator to fail if there's a tab character in the incoming data
3234 (there should never be unless someone is messing about, as it's supposed
3235 to be base64-encoded). Also added, on Timo's advice, the "secured" option
3236 if the connection is using TLS or if the remote IP is the same as the
3237 local IP, and the "valid-client-cert option" if a client certificate has
3240 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
3241 authenticators. This can be used for authorization after authentication
3242 succeeds. (In the case of plaintext, it servers for both authentication
3245 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
3246 if any retry times were supplied.
3248 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
3249 connection, where there is no client IP address to check. In this
3250 situation, the verify now always succeeds.
3252 PH/25 Applied John Jetmore's -Mset patch.
3254 PH/26 Added -bem to be like -Mset, but loading a message from a file.
3256 PH/27 In a string expansion for a processed (not raw) header when multiple
3257 headers of the same name were present, leading whitespace was being
3258 removed from all of them, but trailing whitespace was being removed only
3259 from the last one. Now trailing whitespace is removed from each header
3260 before concatenation. Completely empty headers in a concatenation (as
3261 before) are ignored.
3263 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
3264 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
3266 PH/29 [Removed. This was a change that I later backed out, and forgot to
3267 correct the ChangeLog entry (that I had efficiently created) before
3268 committing the later change.]
3270 PH/30 Exim was sometimes attempting to deliver messages that had suffered
3271 address errors (4xx response to RCPT) over the same connection as other
3272 messages routed to the same hosts. Such deliveries are always "forced",
3273 so retry times are not inspected. This resulted in far too many retries
3274 for the affected addresses. The effect occurred only when there were more
3275 hosts than the hosts_max_try setting in the smtp transport when it had
3276 the 4xx errors. Those hosts that it had tried were not added to the list
3277 of hosts for which the message was waiting, so if all were tried, there
3278 was no problem. Two fixes have been applied:
3280 (i) If there are any address or message errors in an SMTP delivery, none
3281 of the hosts (tried or untried) are now added to the list of hosts
3282 for which the message is waiting, so the message should not be a
3283 candidate for sending over the same connection that was used for a
3284 successful delivery of some other message. This seems entirely
3285 reasonable: after all the message is NOT "waiting for some host".
3286 This is so "obvious" that I'm not sure why it wasn't done
3287 previously. Hope I haven't missed anything, but it can't do any
3288 harm, as the worst effect is to miss an optimization.
3290 (ii) If, despite (i), such a delivery is accidentally attempted, the
3291 routing retry time is respected, so at least it doesn't keep
3292 hammering the server.
3294 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
3295 in ${readsocket because some servers need this prod.
3297 PH/32 Added some extra debug output when updating a wait-xxx database.
3299 PH/33 The hint "could be header name not terminated by colon", which has been
3300 given for certain expansion errors for a long time, was not being given
3301 for the ${if def:h_colon_omitted{... case.
3303 PH/34 The spec says: "With one important exception, whenever a domain list is
3304 being scanned, $domain contains the subject domain." There was at least
3305 one case where this was not true.
3307 PH/35 The error "getsockname() failed: connection reset by peer" was being
3308 written to the panic log as well as the main log, but it isn't really
3309 panic-worthy as it just means the connection died rather early on. I have
3310 removed the panic log writing for the ECONNRESET error when getsockname()
3313 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
3314 runs only) independently of the message's sender address. This meant
3315 that, if the 4xx error was in fact related to the sender, a different
3316 message to the same recipient with a different sender could confuse
3317 things. In particular, this can happen when sending to a greylisting
3318 server, but other circumstances could also provoke similar problems.
3319 I have changed the default so that the retry time for these errors is now
3320 based a combination of the sender and recipient addresses. This change
3321 can be overridden by setting address_retry_include_sender=false in the
3324 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
3325 remote server are returned as part of bounce messages. This was not
3326 happening for LMTP over a pipe (the lmtp transport), but now it is the
3327 same for both kinds of LMTP.
3329 PH/38 Despite being documented as not happening, Exim was rewriting addresses
3330 in header lines that were in fact CNAMEs. This is no longer the case.
3332 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
3333 and queue runs started by the daemon processed all messages. This has
3334 been fixed so that -R and -S can now usefully be given with -q<time>.
3336 PH/40 Import PCRE release 6.7 (fixes some bugs).
3338 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
3340 PH/42 Give an error if -q is specified more than once.
3342 PH/43 Renamed the variables $interface_address and $interface_port as
3343 $received_ip_address and $received_port, to make it clear that these
3344 values apply to message reception, and not to the outgoing interface when
3345 a message is delivered. (The old names remain recognized, of course.)
3347 PH/44 There was no timeout on the connect() call when using a Unix domain
3348 socket in the ${readsocket expansion. There now is.
3350 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
3351 be meaningful with "accept".
3353 SC/01 Eximstats V1.43
3354 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
3356 SC/02 Eximstats V1.44
3357 Use a glob alias rather than an array ref in the generated
3358 parser. This improves both readability and performance.
3360 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
3361 Collect SpamAssassin and rejection statistics.
3362 Don't display local sender or destination tables unless
3363 there is data to show.
3364 Added average volumes into the top table text output.
3366 SC/04 Eximstats V1.46
3367 Collect data on the number of addresses (recipients)
3368 as well as the number of messages.
3370 SC/05 Eximstats V1.47
3371 Added 'Message too big' to the list of mail rejection
3372 reasons (thanks to Marco Gaiarin).
3374 SC/06 Eximstats V1.48
3375 Mainlog lines which have GMT offsets and are too short to
3376 have a flag are now skipped.
3378 SC/07 Eximstats V1.49 (Alain Williams)
3379 Added the -emptyok flag.
3381 SC/08 Eximstats V1.50
3382 Fixes for obtaining the IP address from reject messages.
3384 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
3385 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
3386 whitespace changes from 4.64-PH/27
3388 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
3391 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
3392 are found, allow negative numbers in numeric criteria)
3394 JJ/06 exipick.20061117.2, added new $message_body_missing variable
3396 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
3397 to match changes made in 4.64-PH/43
3399 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
3401 PH/47 Put in an explicit test for a DNS lookup of an address record where the
3402 "domain" is actually an IP address, and force a failure. This locks out
3403 those revolvers/nameservers that support "A-for-A" lookups, in
3404 contravention of the specifications.
3406 PH/48 When a host name was looked up from an IP address, and the subsequent
3407 forward lookup of the name timed out, the host name was left in
3408 $sender_host_name, contrary to the specification.
3410 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3411 restricted to single-key lookups, Exim was not diagnosing an error if
3412 * or *@ was used with a query-style lookup.
3414 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3416 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3417 long ago, but noone interested enough thought of it. Let's just say that
3418 the "1.1" means that there are some new functions that weren't there at
3419 some point in the past.
3421 PH/51 Error processing for expansion failure of helo_data from an smtp
3422 transport during callout processing was broken.
3424 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3425 tested/used via the -bh/-bhc/-bs options.
3427 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3428 bug, fixed in subsequent PCRE releases).
3430 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3431 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3433 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3439 SC/01 Use a glob alias rather than an array ref in eximstats generated
3440 parser. This improves both readability and performance.
3442 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3443 Don't display local sender or destination tables in eximstats unless
3444 there is data to show.
3445 Added average volumes into the eximstats top table text output.
3447 SC/03 Collect data on the number of addresses (recipients) as well
3448 as the number of messages in eximstats.
3450 TF/01 Correct an error in the documentation for the redirect router. Exim
3451 does (usually) call initgroups() when daemonizing.
3453 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3454 with consistent privilege compared to when running as a daemon.
3456 TF/03 Note in the spec that $authenticated_id is not set for local
3457 submissions from trusted users.
3459 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3460 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3462 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3463 by adding some example configuration directives to the default
3464 configuration file. A little bit of work is required to uncomment the
3465 directives and define how usernames and passwords are checked, but
3466 there is now a framework to start from.
3468 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3469 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3470 without this. I don't know how relevant this is to other LDAP libraries.
3472 PH/02 Add the verb name to the "unknown ACL verb" error.
3474 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3476 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3478 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3479 directory not expanded when it should be if an expanded home directory
3480 was set for the address (which is overridden by the transport).
3482 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3485 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3486 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3487 because it is too late at that time, and has no effect.
3489 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3490 security issue with \' (bugzilla #107). I could not use the
3491 PQescapeStringConn() function, because it needs a PGconn value as one of
3494 PH/08 When testing addresses using -bt, indicate those final addresses that
3495 are duplicates that would not cause an additional delivery. At least one
3496 person was confused, thinking that -bt output corresponded to deliveries.
3497 (Suppressing duplicates isn't a good idea as you lose the information
3498 about possibly different redirections that led to the duplicates.)
3500 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3501 systems where poll() doesn't work, in particular OS X.
3503 PH/10 Added more information to debugging output for retry time not reached.
3505 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3506 operations in malware.c.
3508 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3511 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3512 syslog_duplication set false, log lines that would normally be written
3513 both the the main log and to the reject log were not written to syslog at
3516 PH/14 In the default configuration, change the use of "message" in ACL warn
3517 statements to "add_header".
3519 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3520 not followed by a command (e.g. "seen endif").
3522 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3523 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3526 PH/17 Added extra conditions to the default value of delay_warning_condition
3530 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3531 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3532 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3535 The Auto-Submitted: and various List- headers are standardised, whereas I
3536 don't think Precedence: ever was.
3538 PH/18 Refactored debugging code in route_finduser() to show more information,
3539 in particular, the error code if getpwnam() issues one.
3541 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3542 This is apparently needed in addition to the PH/07 change above to avoid
3543 any possible encoding problems.
3545 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3546 but not after initializing Perl.
3548 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3549 output them only if debugging. By default they are written stderr,
3550 apparently, which is not desirable.
3552 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3555 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3558 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3560 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3561 authenticated or an ident call has been made. Suppress the default
3562 values for $authenticated_id and $authenticated_sender (but permit -oMai
3563 and -oMas) when testing with -bh.
3565 PH/24 Re-jigged the order of the tests in the default configuration so that the
3566 tests for valid domains and recipients precede the DNS black list and CSA
3567 tests, on the grounds that those ones are more expensive.
3569 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3570 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3571 This bug exists in every version of Exim that I still have, right back to
3574 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3575 However, an attempt to turn on case-sensitivity in a regex key by
3576 including (?-i) didn't work because the subject string was already
3577 lowercased, and the effects were non-intuitive. It turns out that a
3578 one-line patch can be used to allow (?-i) to work as expected.
3584 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3585 other effects) broke the use of negated acl sub-conditions.
3587 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3590 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3591 "Deny" causes Exim to reject the incoming connection with a 554 error.
3592 Unfortunately, if there is a major crisis, such as a disk failure,
3593 tcp-wrappers gives "deny", whereas what one would like would be some
3594 kind of temporary error. A kludge has been added to help with this.
3595 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3596 554 error is used if errno is still zero or contains ENOENT (which occurs
3597 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3600 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3602 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3603 errors. Otherwise a message that provokes a temporary error (when other
3604 messages do not) can cause a whole host to time out.
3606 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3607 addresses were routed directly to files or pipes from a redirect router.
3608 File deliveries just didn't batch; pipe deliveries might have suffered
3611 PH/06 A failure to get a lock for a hints database would erroneously always say
3612 "Failed to get write lock", even when it was really a read lock.
3614 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3615 of 0600. This has been changed to use the value of the lockfile_mode
3616 option (which defaults to 0600).
3618 PH/08 Applied small patch from the Sieve maintainer.
3620 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3621 folder from quota calculations, a direct delivery into this folder messed
3622 up the contents of the maildirsize file. This was because the regex was
3623 used only to exclude .Trash (or whatever) when the size of the mailbox
3624 was calculated. There was no check that a delivery was happening into an
3625 excluded directory. This bug has been fixed by ignoring all quota
3626 processing for deliveries into excluded directories.
3628 PH/10 Added the maildirfolder_create_regex option to appendfile.
3634 PH/01 The code for finding all the local interface addresses on a FreeBSD
3635 system running IPv6 was broken. This may well have applied to all BSD
3636 systems, as well as to others that have similar system calls. The broken
3637 code found IPv4 interfaces correctly, but gave incorrect values for the
3638 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3639 that it would not match correctly against @[] and not recognize the IPv6
3642 PH/02 The ipliteral router was not recognizing addresses of the form user@
3643 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3645 PH/03 Added disable_ipv6.
3647 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3648 decoded form, because it is most often used to construct To: headers
3649 lines in autoreplies, and the decoded form may well be syntactically
3650 invalid. However, $reply_address has leading white space removed, and all
3651 newlines turned into spaces so that the autoreply transport does not
3654 PH/05 If group was specified without a user on a router, and no group or user
3655 was specified on a transport, the group from the router was ignored.
3657 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3658 for visible compile-time settings that can be used to change these
3659 numbers, for those that want even more. Backwards compatibility with old
3660 spool files has been maintained. However, going back to a previous Exim
3661 release will lost any variables that are in spool files.
3663 PH/07 Two small changes when running in the test harness: increase delay when
3664 passing a TCP/IP connection to a new process, in case the original
3665 process has to generate a bounce, and remove special handling of
3666 127.0.0.2 (sic), which is no longer necessary.
3668 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3669 be the same on different OS.
3671 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3674 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3675 whether --show-vars was specified or not
3677 JJ/02 exipick: Added support for new ACL variable spool format introduced
3680 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3681 syntactically invalid From: or Reply-to: line, and a filter used this to
3682 generate an autoreply, and therefore failed to obtain an address for the
3683 autoreply, Exim could try to deliver to a non-existent relative file
3684 name, causing unrelated and misleading errors. What now happens is that
3685 it logs this as a hard delivery error, but does not attempt to create a
3688 PH/11 The exinext utility has a -C option for testing purposes, but although
3689 the given file was scanned by exinext itself; it wasn't being passed on
3690 when Exim was called.
3692 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3693 an end-of-file indication when reading a command response.
3695 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3696 compiled. In many other places in Exim, IPv6 addresses are always
3697 recognized, so I have changed this. It also means that IPv4 domain
3698 literals of the form [IPV4:n.n.n.n] are now always recognized.
3700 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3701 used if the router is not running as root, for example, when verifying at
3702 ACL time, or when using -bh. The debugging output from this situation was
3703 non-existent - all you got was a failure to exec. I have made two
3706 (a) Failures to set uid/gid, the current directory, or a process leader
3707 in a subprocess such as that created by queryprogram now generate
3708 suitable debugging output when -d is set.
3710 (b) The queryprogram router detects when it is not running as root,
3711 outputs suitable debugging information if -d is set, and then runs
3712 the subprocess without attempting to change uid/gid.
3714 PH/15 Minor change to Makefile for building test_host (undocumented testing
3717 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3718 additional section of a DNS packet that returns MX or SRV records.
3719 Instead, it always explicitly searches for A/AAAA records. This avoids
3720 major problems that occur when a DNS server includes only records of one
3721 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3722 fixed another bug: if SRV records were looked up and the corresponding
3723 address records were *not* found in the additional section, the port
3724 values from the SRV records were lost.
3726 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3727 using the correct key (the original address) when searching the retry
3728 rules in order to find which one to use for generating the retry hint.
3730 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3731 adding the default one. Similarly, if it contains a Reply-To: header, the
3732 errors_reply_to option, if set, is not used.
3734 PH/19 When calculating a retry time, Exim used to measure the "time since
3735 failure" by looking at the "first failed" field in the retry record. Now
3736 it does not use this if it is later than than the arrival time of the
3737 message. Instead it uses the arrival time. This makes for better
3738 behaviour in cases where some deliveries succeed, thus re-setting the
3739 "first failed" field. An example is a quota failure for a huge message
3740 when small messages continue to be delivered. Without this change, the
3741 "time since failure" will always be short, possible causing more frequent
3742 delivery attempts for the huge message than are intended.
3743 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3745 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3746 $1, $2, $3) because the numerical variables can be reset during some
3747 expansion items (e.g. "match"), thereby losing the authentication data.
3749 PH/21 Make -bV show the size of off_t variables so that the test suite can
3750 decide whether to run tests for quotas > 2G.
3752 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3753 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3754 filecount value is greater than 2G or if a quota value is greater than 2G
3755 on a system where the size of off_t is not greater than 4, a panic error
3758 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3759 never match. The debug and -bh output now contains an explicit error
3760 message indicating a malformed IPv4 address or mask.
3762 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3763 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3764 PH/23 above applies.
3766 PH/25 Do not write to syslog when running in the test harness. The only
3767 occasion when this arises is a failure to open the main or panic logs
3768 (for which there is an explicit test).
3770 PH/26 Added the /no_tell option to "control=freeze".
3772 PH/27 If a host name lookup failed very early in a connection, for example, if
3773 the IP address matched host_lookup and the reverse lookup yielded a name
3774 that did not have a forward lookup, an error message of the form "no IP
3775 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3776 could be logged. Now it outputs the IP address instead of "NULL".
3778 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3779 allows the sender and the authenticated sender to be set when
3780 submitting a message from within Exim. Since child_open_exim() is
3781 documented for local_scan(), the new function should be too.
3783 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3784 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3785 results in an empty string is now treated as unset.
3787 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3789 PH/31 Added sender_verify_fail logging option.
3791 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3792 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3793 tidied the source and removed it altogether.
3795 PH/33 When a queue run was abandoned because the load average was too high, a
3796 log line was always written; now it is written only if the queue_run log
3797 selector is set. In addition, the log line for abandonment now contains
3798 information about the queue run such as the pid. This is always present
3799 in "start" and "stop" lines but was omitted from the "abandon" line.
3801 PH/34 Omit spaces between a header name and the colon in the error message that
3802 is given when verify = headers_syntax fails (if there are lots of them,
3803 the message gets confusing).
3805 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3806 names, as there are now some PTR records that contain slashes. This check
3807 is only to protect against broken name servers that fall over on strange
3808 characters, so the fact that it applies to all lookups doesn't matter.
3810 PH/36 Now that the new test suite is complete, we can remove some of the
3811 special code in Exim that was needed for the old test suite. For example,
3812 sorting DNS records because real resolvers return them in an arbitrary
3813 order. The new test suite's fake resolver always returns records in the
3816 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3817 bounces) except when queue_only is set, to avoid logging races between
3818 the different processes.
3820 PH/38 Panic-die if .include specifies a non-absolute path.
3822 PH/39 A tweak to the "H" retry rule from its user.
3824 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3825 a label. They prevented compilation on older perls.
3827 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3828 a warning to be raised on newish perls.
3830 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3831 on queue. Changes to match documented behaviour of showing count of
3832 messages matching specified criteria.
3834 PH/40 Changed the default ident timeout from 30s to 5s.
3836 PH/41 Added support for the use of login_cap features, on those BSD systems
3837 that have them, for controlling the resources used by pipe deliveries.
3839 PH/42 The content-scanning code uses fopen() to create files in which to put
3840 message data. Previously it was not paying any attention to the mode of
3841 the files. Exim runs with umask(0) because the rest of the code creates
3842 files with open(), and sets the required mode explicitly. Thus, these
3843 files were ending up world-writeable. This was not a big issue, because,
3844 being within the spool directory, they were not world-accessible. I have
3845 created a function called modefopen, which takes an additional mode
3846 argument. It sets umask(777), creates the file, chmods it to the required
3847 mode, then resets the umask. All the relevant calls to fopen() in the
3848 content scanning code have been changed to use this function.
3850 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3851 to 24 hours. This avoids potential overflow problems when processing G
3852 and H retry rules. I suspect nobody ever tinkers with this value.
3854 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3856 PH/45 When the plaintext authenticator is running as a client, the server's
3857 challenges are checked to ensure they are valid base64 strings. By
3858 default, the authentication attempt is cancelled if an invalid string is
3859 received. Setting client_ignore_invalid_base64 true ignores these errors.
3860 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3861 they are received. Thus, the responses can be made to depend on the
3862 challenges. If an invalid string is ignored, an empty string is placed in
3865 PH/46 Messages that are created by the autoreply transport now contains a
3866 References: header, in accordance with RFCs 2822 and 3834.
3868 PH/47 Added authenticated_sender_force to the smtp transport.
3870 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3872 PH/49 Installed latest patch from the Sieve maintainer.
3874 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3875 was also set, the appendfile transport was unnecessarily scanning a
3876 directory of message files (e.g. for maildir delivery) to find the count
3877 of files (along with the size), even though it did not need this
3878 information. It now does the scan only if it needs to find either the
3879 size of the count of files.
3881 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3883 PH/52 Two bugs concerned with error handling when the smtp transport is
3886 (i) Exim was not creating retry information for temporary errors given
3887 for individual recipients after the DATA command when the smtp transport
3888 was used in LMTP mode. This meant that they could be retried too
3889 frequently, and not timed out correctly.
3891 (ii) Exim was setting the flag that allows error details to be returned
3892 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3893 recipients that were returned after the DATA command.
3895 PH/53 This is related to PH/52, but is more general: for any failing address,
3896 when detailed error information was permitted to be returned to the
3897 sender, but the error was temporary, then after the final timeout, only
3898 "retry timeout exceeded" was returned. Now it returns the full error as
3899 well as "retry timeout exceeded".
3901 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3902 do this, and (what is worse) MTAs that accept it.
3904 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3905 will now be deprecated.
3907 PH/56 New os.c-cygwin from the Cygwin maintainer.
3909 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3910 formats (previously only available in exim formats via -bpr, -bpru,
3911 and -bpra. Now also available in native and exiqgrep formats)
3913 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3914 with very large, slow to parse queues
3916 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3918 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3920 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3921 responses to authentication challenges, though it was showing the
3922 challenges; (ii) I've removed the CR characters from the debug output for
3925 PH/58 Allow for the insertion of a newline as well as a space when a string
3926 is turned into more than one encoded-word during RFC 2047 encoding. The
3927 Sieve code now uses this.
3929 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3930 data_4xx, lost_connection, tls_required.
3932 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3933 message was being sent as an SMTP response.
3935 PH/61 Add -l and -k options to exicyclog.
3937 PH/62 When verifying, if an address was redirected to one new address, so that
3938 verification continued, and the new address failed or deferred after
3939 having set something in $address_data, the value of $address_data was not
3940 passed back to the ACL. This was different to the case when no
3941 redirection occurred. The value is now passed back in both cases.
3943 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3944 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3945 use login_cap.h, so on its own it isn't the distinguishing feature. The
3946 new name refers directly to the setclassresources() function.
3948 PH/65 Added configuration files for NetBSD3.
3950 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3952 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3953 is preferred over IPv4.
3955 PH/68 The bounce_return_message and bounce_return_body options were not being
3956 honoured for bounces generated during the reception of non-SMTP messages.
3957 In particular, this applied to messages rejected by the ACL. This bug has
3958 been fixed. However, if bounce_return_message is true and bounce_return_
3959 body is false, the headers that are returned for a non-SMTP message
3960 include only those that have been read before the error was detected.
3961 (In the case of an ACL rejection, they have all been read.)
3963 PH/69 The HTML version of the specification is now built in a directory called
3964 spec_html instead of spec.html, because the latter looks like a path with
3965 a MIME-type, and this confuses some software.
3967 PH/70 Catch two compiler warnings in sieve.c.
3969 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3970 function verify_get_ident() calls ip_connect() to connect a socket, but
3971 if the "connect()" function timed out, ip_connect() used to close the
3972 socket. However, verify_get_ident() also closes the socket later, and in
3973 between Exim writes to the log, which may get opened at this point. When
3974 the socket was closed in ip_connect(), the log could get the same file
3975 descriptor number as the socket. This naturally causes chaos. The fix is
3976 not to close the socket in ip_connect(); the socket should be closed by
3977 the function that creates it. There was only one place in the code where
3978 this was missing, in the iplookup router, which I don't think anybody now
3979 uses, but I've fixed it anyway.
3981 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3982 well as to direct DNS lookups. Otherwise the handling of names in host
3983 lists is inconsistent and therefore confusing.
3989 PH/01 Two changes to the default runtime configuration:
3991 (1) Move the checks for relay_from_hosts and authenticated clients from
3992 after to before the (commented out) DNS black list checks.
3994 (2) Add control=submission to the relay_from_hosts and authenticated
3995 clients checks, on the grounds that messages accepted by these
3996 statements are most likely to be submissions.
3998 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
4000 (1) Generate an error if the third argument for the ${prvs expansion is
4003 (2) Treat a missing third argument of ${prvscheck as if it were an empty
4006 (3) Reset the variables that are obtained from the first argument of
4007 ${prvscheck and used in the second argument before leaving the code,
4008 because their memory is reclaimed, so using them afterwards may do
4011 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
4012 one (it's much easier than Tom thought :-).
4014 (5) Because of (4), we can now allow for the use of $prvscheck_result
4015 inside the third argument.
4017 PH/03 For some reason, the default setting of PATH when running a command from
4018 a pipe transport was just "/usr/bin". I have changed it to
4021 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
4022 anything to be listed in the output from -bV.
4024 PH/05 When a filter generated an autoreply, the entire To: header line was
4025 quoted in the delivery log line, like this:
4027 => >A.N.Other <ano@some.domain> <original@ddress> ...
4029 This has been changed so that it extracts the operative address. There
4030 may be more than one such address. If so, they are comma-separated, like
4033 => >ano@some.domain,ona@other.domain <original@ddress> ...
4035 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
4036 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
4037 not being looked up in the rDNS to get a host name, Exim was showing the
4038 IP address twice in Received: lines, even though the IP addresses were
4039 identical. For example:
4041 Received: from [1.2.3.4] (helo=[1.2.3.4])
4043 However, if the real host name was known, it was omitting the HELO data
4044 if it matched the actual IP address. This has been tidied up so that it
4045 doesn't show the same IP address twice.
4047 PH/07 When both +timestamp and +memory debugging was on, the value given by
4048 $tod_xxx expansions could be wrong, because the tod_stamp() function was
4049 called by the debug printing, thereby overwriting the timestamp buffer.
4050 Debugging no longer uses the tod_stamp() function when +timestamp is set.
4052 PH/08 When the original message was included in an autoreply transport, it
4053 always said "this is a copy of the message, including all the headers",
4054 even if body_only or headers_only was set. It now gives an appropriate
4057 PH/09 Applied a patch from the Sieve maintainer which:
4059 o fixes some comments
4060 o adds the (disabled) notify extension core
4061 o adds some debug output for the result of if/elsif tests
4062 o points to the current vacation draft in the documentation
4063 and documents the missing references header update
4067 o fixes a bug in processing the envelope test (when testing
4068 multiple envelope elements, the last element determined the
4071 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
4072 Electronic Mail") by including:
4074 Auto-submitted: auto-generated
4076 in the messages that it generates (bounce messages and others, such as
4077 warnings). In the case of bounce messages for non-SMTP messages, there was
4078 also a typo: it was using "Auto_submitted" (underscore instead of
4079 hyphen). Since every message generated by Exim is necessarily in response
4080 to another message, thes have all been changed to:
4082 Auto-Submitted: auto-replied
4084 in accordance with these statements in the RFC:
4086 The auto-replied keyword:
4088 - SHOULD be used on messages sent in direct response to another
4089 message by an automatic process,
4091 - MUST NOT be used on manually-generated messages,
4093 - MAY be used on Delivery Status Notifications (DSNs) and Message
4094 Disposition Notifications (MDNs),
4096 - MUST NOT be used on messages generated by automatic or periodic
4097 processes, except for messages which are automatic responses to
4100 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
4101 to the default Received: header definition.
4103 PH/12 Added log selector acl_warn_skipped (default on).
4105 PH/13 After a successful wildlsearch lookup, discard the values of numeric
4106 variables because (a) they are in the wrong storage pool and (b) even if
4107 they were copied, it wouldn't work properly because of the caching.
4109 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
4110 checking when decoding. Apparently there are clients that generate
4111 overlong encoded strings. Why am I not surprised?
4113 PH/15 If the first argument of "${if match_address" was not empty, but did not
4114 contain an "@" character, Exim crashed. Now it writes a panic log message
4115 and treats the condition as false.
4117 PH/16 In autoreply, treat an empty string for "once" the same as unset.
4119 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
4120 extension "envelope-auth". The code is finished and in agreement with
4121 other implementations, but there is no documentation so far and in fact,
4122 nobody wrote the draft yet. This extension is currently #undef'ed, thus
4123 not changing the active code.
4125 Print executed "if" and "elsif" statements when debugging is used. This
4126 helps a great deal to understand what a filter does.
4128 Document more things not specified clearly in RFC3028. I had all this
4129 sorted out, when out of a sudden new issues came to my mind. Oops."
4131 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
4134 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
4135 canonical form (as documented). However, after a host name lookup from
4136 the IP address, check_host() was doing a simple string comparison with
4137 addresses acquired from the DNS when checking that the found name did
4138 have the original IP as one of its addresses. Since any found IPv6
4139 addresses are likely to be in abbreviated form, the comparison could
4140 fail. Luckily, there already exists a function for doing the comparison
4141 by converting both addresses to binary, so now that is used instead of
4142 the text comparison.
4144 PH/20 There was another similar case to PH/19, when a complete host name was
4145 given in a host list; looking up its IP address could give an abbreviated
4146 form, whereas the current host's name might or might not be abbreviated.
4147 The same fix has been applied.
4153 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
4154 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
4157 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
4158 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
4160 PH/03 Typo: missing ".o" in src/pcre/Makefile.
4162 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
4163 header line, restrict the check to what is listed in RFCs 2369 and 2929.
4164 Also, for "Auto-Submitted", treat anything other than "no" as
4165 non-personal, in accordance with RFC 3834. (Previously it treated
4166 anything starting "auto-" as non-personal.)
4168 TF/01 The control=submission/name=... option had a problem with syntax
4169 errors if the name included a slash character. The /name= option
4170 now slurps the rest of the string, so it can include any characters
4171 but it must come last in the list of options (after /sender_retain
4174 PH/05 Some modifications to the interface to the fake nameserver for the new
4182 TK/01 Added the "success_on_redirect" address verification option. See
4183 NewStuff for rationale and an example.
4185 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
4187 PH/02 Patch to exigrep to allow it to work on syslog lines.
4189 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
4190 fread() to skip over the body file's header line, because in Cygwin the
4191 header line is locked and is inaccessible.
4193 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
4194 co-exist for some time) to make it clear that it is the Exim ID that is
4195 referenced, not the Message-ID: header line.
4197 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
4198 string_format() function, because snprintf() does not exist on all
4201 PH/06 The use of forbid_filter_existstest now also locks out the use of the
4202 ${stat: expansion item.
4204 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
4205 protocol synchronization error", to keep the pedants happy.
4207 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
4208 well as for IRIX systems, when gcc is being used. See the host.c source
4211 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
4213 PH/10 Named domain lists were not working if used in a queue_smtp_domains
4216 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
4217 transport and to the smtp transport in LMTP mode.
4219 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
4221 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
4222 run a filter in a subprocess. This could lead to confusion in subsequent
4223 lookups in the parent process. There should also be a search_tidyup() at
4224 the end of the subprocess.
4226 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
4227 only if the host matched helo_try_verify_hosts, which caused the
4228 verification to occur when the EHLO/HELO command was issued. The ACL just
4229 tested the remembered result. Now, if a previous verification attempt has
4230 not happened, "verify = helo" does it there and then.
4232 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
4234 TK/03 Fix log output including CR from clamd.
4236 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
4237 exist provoked a memory error which could cause a segfault.
4239 PH/15 Installed PCRE 6.2
4241 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
4243 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
4244 of the problem. Specifically, suggested +O2 rather than +O1 for the
4247 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
4249 PH/20 If a delivery was routed to a non-standard port by means of an SRV
4250 record, the port was not correctly logged when the outgoing_port log
4251 selector was set (it logged the transort's default port).
4253 PH/21 Added support for host-specific ports to manualroute, queryprogram,
4254 fallback_hosts, and "hosts" in the smtp transport.
4256 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
4257 host errors such as "Connection refused".
4259 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
4260 authentication with radiusclient 0.4.9:
4262 - Error returned from rc_read_config was caught wrongly
4263 - Username/password not passed on to radius server due to wrong length.
4265 The presumption is that some radiusclient API changes for 4.51/PH/17
4266 were not taken care of correctly. The code is still untested by me (my
4267 Linux distribution still has 0.3.2 of radiusclient), but it was
4268 contributed by a Radius user.
4270 PH/24 When doing a callout, the value of $domain wasn't set correctly when
4271 expanding the "port" option of the smtp transport.
4273 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
4274 while reading a MIME header. Thanks to Tom Hughes for a patch.
4276 PH/24 Include config.h inside local_scan.h so that configuration settings are
4279 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
4280 that in an ACL for RCPT (for example), you can examine exactly what was
4283 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
4284 commands, but it was not correctly comparing the address with the actual
4285 client host address. Thus, it would show the EHLO address in Received:
4286 header lines when this was not necessary.
4288 PH/27 Added the % operator to ${eval:}.
4290 PH/28 Exim tries to create and chdir to its spool directory when it starts;
4291 it should be ignoring failures (because with -C, for example, it has lost
4292 privilege). It wasn't ignoring creation failures other than "already
4295 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
4298 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
4299 because an input line was too long, either on its own, or by virtue of
4300 too many continuations, the temporary file was not being removed, and the
4301 return code was incorrect.
4303 PH/31 Missing "BOOL" in function definition in filtertest.c.
4305 PH/32 Applied Sieve patches from the maintainer.
4307 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
4309 PH/33 Added "verify = not_blind".
4311 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
4312 Local/Makefile (with some defaults set). These are used in built scripts
4313 such as exicyclog, but they have never been used in the exim_install
4314 script (though there are many overriding facilities there). I have
4315 arranged that the exim_install script now takes note of these two
4318 PH/35 Installed configuration files for Dragonfly.
4320 PH/36 When a locally submitted message by a trusted user did not contain a
4321 From: header, and the sender address was obtained from -f or from an SMTP
4322 MAIL command, and the trusted user did not use -F to supply a sender
4323 name, $originator_name was incorrectly used when constructing a From:
4324 header. Furthermore, $originator_name was used for submission mode
4325 messages from external hosts without From: headers in a similar way,
4326 which is clearly wrong.
4328 PH/37 Added control=suppress_local_fixups.
4330 PH/38 When log_selector = +received_sender was set, and the addition of the
4331 sender made the log line's construction buffer exactly full, or one byte
4332 less than full, an overflow happened when the terminating "\n" was
4335 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
4336 when the result of a list match is failure because a DNS lookup failed.
4338 PH/40 RM_COMMAND is now used in the building process.
4340 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
4341 the "build-* directories that it finds.
4343 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
4344 address in a domain literal was a prefix of an interface address.
4346 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
4347 when verifying a sender address, unless rewrite_headers is false.
4349 PH/44 Wrote a long comment about why errors_to addresses are verified as
4350 recipients, not senders.
4352 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
4353 the ratelimit ACL was added.
4355 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
4357 PH/46 Added extra information about PostgreSQL errors to the error string.
4359 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
4360 avoiding the need to install special zones in a real server. This is
4361 backwards compatible; if it can't find the fake resolver, it drops back.
4362 Thus, both old and new test suites can be run.
4364 TF/02 Added util/ratelimit.pl
4366 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
4367 clock is set back in time.
4369 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
4370 Candler <B.Candler@pobox.com>.
4372 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
4373 OK for addresses that are the result of redirections.
4375 PH/48 A number of further additions for the benefit of the new test suite,
4376 including a fake gethostbyname() that interfaces to the fake DNS resolver
4379 TF/06 The fix for widen_domains has also been applied to qualify_single and
4380 search_parents which are the other dnslookup options that can cause
4383 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
4386 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
4388 TF/07 Exim produced the error message "an SRV record indicated no SMTP
4389 service" if it encountered an MX record with an empty target hostname.
4390 The message is now "an MX or SRV record indicated no SMTP service".
4392 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
4393 if the DNS of the sending site is misconfigured. This is quite a
4394 common situation. This change restores the behaviour of treating a
4395 helo verification defer as a failure.
4397 PH/51 If self=fail was set on a router, the bounce message did not include the
4398 actual error message.
4404 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
4406 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
4407 command itself ended in error, the underlying message about the transport
4408 filter timeout was being overwritten with the pipe command error. Now the
4409 underlying error message should be appended to the second error message.
4411 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4413 PH/02 Reduce the amount of output that "make" produces by default. Full output
4414 can still be requested.
4416 PH/03 The warning log line about a condition test deferring for a "warn" verb
4417 was being output only once per connection, rather than after each
4418 occurrence (because it was using the same function as for successful
4419 "warn" verbs). This seems wrong, so I have changed it.
4421 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4422 it should not have, which might have caused a crash in the right
4423 circumstances, but probably never did.
4425 PH/04 Installed a modified version of Tony Finch's patch to make submission
4426 mode fix the return path as well as the Sender: header line, and to
4427 add a /name= option so that you can make the user's friendly name appear
4430 TF/03 Added the control = fakedefer ACL modifier.
4432 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4433 Mark Lowes for thorough testing.
4435 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4437 TK/03 Merged latest SRS patch from Miles Wilton.
4439 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4440 with the definition in sysexits.h (which is #included earlier).
4441 Fortunately, Exim does not actually use EX_OK. The code used to try to
4442 preserve the sysexits.h value, by assuming that macro definitions were
4443 scanned for macro replacements. I have been disabused of this notion,
4444 so now the code just undefines EX_OK before #including unistd.h.
4446 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4447 in the smtp transport. When a block could not be written in a single
4448 write() function, the timeout was being re-applied to each part-write.
4449 This seems wrong - if the receiver was accepting one byte at a time it
4450 would take for ever. The timeout is now adjusted when this happens. It
4451 doesn't have to be particularly precise.
4453 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4454 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4456 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4457 without a domain if the check to <postmaster@domain> fails.
4459 SC/01 Eximstats: added -xls and the ability to specify output files
4460 (patch written by Frank Heydlauf).
4462 SC/02 Eximstats: use FileHandles for outputting results.
4464 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4466 SC/04 Eximstats: fixed display of large numbers with -nvr option
4468 SC/05 Eximstats: fixed merging of reports with empty tables.
4470 SC/06 Eximstats: added the -include_original_destination flag
4472 SC/07 Eximstats: removed tabs and trailing whitespace.
4474 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4476 TK/06 MBOX spool code: Add real "From " MBOX separator line
4477 so the .eml file is really in mbox format (even though
4478 most programs do not really care). Patch from Alex Miller.
4480 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4481 The latter is generated from $received_to and is only set if the
4482 message has one envelope recipient. SA can use these headers,
4483 obviously out-of-the-box. Patch from Alex Miller.
4485 PH/08 The ${def test on a variable was returning false if the variable's
4486 value was "0", contrary to what the specification has always said!
4487 The result should be true unless the variable is empty.
4489 PH/09 The syntax error of a character other than { following "${if
4490 def:variable_name" (after optional whitespace) was not being diagnosed.
4491 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4492 accidental colon was present, for example, could give incorrect results.
4494 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4495 result is used (not including appendfile, where other changes are about
4498 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4499 This involved changing a lot of size variables from int to off_t. It
4500 should work with maildirs and everything.
4502 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4503 spamd dying while we are connected to it.
4505 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4508 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4509 function for BDB 4.3.
4511 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4513 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4514 the sources for PCRE have been re-arranged and more files are now
4517 PH/15 The code I had for printing potentially long long variables in PH/11
4518 above was not the best (it lost precision). The length of off_t variables
4519 is now inspected at build time, and an appropriate printing format (%ld
4520 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4521 to be "long long int" or "long int". This is needed for the internal
4522 formatting function string_vformat().
4524 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4525 the configuration file to be ":syslog", then the script "guesses" where
4526 the logs files are, rather than using the compiled in default. In our
4527 case the guess is not the same as the compiled default, so the script
4528 suddenly stopped working when I started to use syslog. The patch checks
4529 to see if log_file_path is "". If so, it attempts to read it from exim
4530 with no configuration file to get the compiled in version, before it
4531 falls back to the previous guessing code."
4533 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4534 implementing BATV in an Exim configuration. See NewStuff for the gory
4537 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4538 Makefile that are specific to HP-UX.
4540 PH/18 If the "use_postmaster" option was set for a recipient callout together
4541 with the "random" option, the postmaster address was used as the MAIL
4542 FROM address for the random test, but not for the subsequent recipient
4543 test. It is now used for both.
4545 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4546 patch removes a few documentation additions to RFC 3028, because the
4547 latest draft now contains them. It adds the new en;ascii-case comparator
4548 and a new error check for 8bit text in MIME parts. Comparator and
4549 require names are now matched exactly. I enabled the subaddress
4550 extension, but it is not well tested yet (read: it works for me)."
4552 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4553 rework some of the code of TK/09 above to avoid the hardwired use of
4554 "%lld" and "long long". Replaced the call to snprintf() with a call to
4557 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4558 records point to non-existent hosts", "retry timeout exceeded", and
4559 "retry time not reached for any host after a long failure period".
4561 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4562 experimental DomainKeys support:
4564 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4565 (2) On an error such as an illegally used "control", the wrong name for
4566 the control was given.
4568 These problems did NOT occur unless DomainKeys support was compiled.
4570 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4572 PH/24 Added ${if match_ip condition.
4574 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4575 databases so that it will be absolutely obvious if a crash occurs in the
4576 DB library. This is a regular occurrence (often caused by mis-matched
4579 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4580 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4581 picked up on a user's system that detects such things. There doesn't seem
4582 to be a gcc warning option for this - only an attribute that has to be
4583 put on the function's prototype. It seems that in Fedora Core 4 they have
4584 set this on a number of new functions. No doubt there will be more in due
4587 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4588 specify a transport. However, if an address that was verified by such a
4589 router was the subject of a callout, Exim crashed because it tried to
4590 read the rcpt_include_affixes from the non-existent transport. Now it
4591 just assumes that the setting of that option is false. This bug was
4592 introduced by 4.51/PH/31.
4594 PH/28 Changed -d+all to exclude +memory, because that information is very
4595 rarely of interest, but it makes the output a lot bigger. People tend to
4596 do -d+all out of habit.
4598 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4599 code in os-type was giving problems when libc.so lives in lib64, like on
4602 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4603 aren't the modern standard, and it seems that some systems' include files
4604 don't always have them. Exim was already checking for some of the newer
4605 ones like T_AAAA, and defining it itself. I've added checks for all the
4606 record types that Exim uses.
4608 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4609 not automatically generating a new one, as it is supposed to. This
4610 prevented TLS from working. If the file did exist, but contained invalid
4611 data, a new version was generated, as expected. It was only the case of a
4612 non-existent file that was broken.
4614 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4615 with a change in libdomainkeys > 0.64.
4617 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4618 from DNS. If the selector record carries the flag, it now has
4619 precedence over the domain-wide flag.
4621 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4623 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4624 the use of a port name that isn't defined in /etc/services) occurred, the
4625 message was deferred as in a normal delivery, and thus remained on the
4626 spool, instead of being failed because of the mua_wrapper setting. This
4627 is now fixed, and I tidied up some of the mua_wrapper messages at the
4630 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4631 the messages in a hash of arrays rather than using individual hashes.
4632 This is a bit cleaner and results in dramatic memory savings, albeit
4633 at a slight CPU cost.
4635 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4636 as requested by Marc Sherman.
4638 SC/10 Eximstats: added histograms for user specified patterns as requested
4641 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4643 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4644 fopen() in the content-scanning modules that did not already have it.
4650 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4651 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4653 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4655 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4657 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4658 to test Sieve filters that use "vacation".
4660 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4661 that changes the way the GnuTLS parameters are stored in the cache file.
4662 The new format can be generated externally. For backward compatibility,
4663 if the data in the cache doesn't make sense, Exim assumes it has read an
4664 old-format file, and it generates new data and writes a new file. This
4665 means that you can't go back to an older release without removing the
4668 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4669 work if there are any delivery delays because "one_time" forces the
4670 parent to be marked "delivered", so its unseen clone is never tried
4671 again. For this reason, Exim now forbids the simultaneous setting of
4674 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4675 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4676 entry if you want to know the details. The fix, however, neglected to
4677 consider the case when local delivery batching is involved. The test for
4678 "previously delivered" was not happening when checking to see if an
4679 address could be batched with a previous (undelivered) one; under
4680 certain circumstances this could lead to multiple deliveries to the same
4683 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4684 in its include files, and this causes problems building Exim.
4686 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4687 header_syntax) but Exim was just ignoring anything given after a slash.
4688 In particular, this caused confusion with an attempt to use "verify =
4689 reverse_host_lookup/defer_ok". An error is now given when options are
4690 supplied for verify items that do not have them. (Maybe reverse_host_
4691 lookup should have a defer_ok option, but that's a different point.)
4693 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4694 defined by RFC 821) to 2048, because there were problems with some AUTH
4695 commands, and RFC 1869 says the size should be increased for extended
4696 SMTP commands that take arguments.
4698 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4701 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4702 "unknown" error; now it says that the functionality isn't in the binary.
4704 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4705 an address' error message when a string expansion fails (syntax or
4706 whatever). Otherwise the password may appear in the log. Following change
4707 PH/42 below, there is no longer a chance of it appearing in a bounce
4710 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4712 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4713 of its addresses was ever tried. (Bugzilla bug #2.)
4715 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4716 the result incorrectly in the debug output. (It correctly added a newline
4717 to what was transported.)
4719 TF/01 Added $received_time.
4721 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4722 commented out examples of how to interface to a virus scanner and to
4723 SpamAssassin. Also added commented examples of av_scanner and
4724 spamd_address settings.
4726 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4727 and controls are allowed in which ACLs. There were a couple of minor
4728 errors. Some of the entries in the conditions table (which is a table of
4729 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4730 negation of where the condition IS allowed.
4732 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4734 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4735 header file does not have a version number, so I've had to invent a new
4736 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4737 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4738 radiusclient), but it was contributed by a Radius user.
4740 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4741 files or directories, for OpenSSL.
4743 PH/19 When an Exim process that is running as root has to create an Exim log
4744 file, it does so in a subprocess that runs as exim:exim so as to get the
4745 ownership right at creation (otherwise, other Exim processes might see
4746 the file with the wrong ownership). There was no test for failure of this
4747 fork() call, which would lead to the process getting stuck as it waited
4748 for a non-existent subprocess. Forks do occasionally fail when resources
4749 run out. I reviewed all the other calls to fork(); they all seem to check
4752 PH/20 When checking for unexpected SMTP input at connect time (before writing
4753 the banner), Exim was not dealing correctly with a non-positive return
4754 from the read() function. If the client had disconnected by this time,
4755 the result was a log entry for a synchronization error with an empty
4756 string after "input=" when read() returned zero. If read() returned -1
4757 (an event I could not check), uninitialized data bytes were printed.
4758 There were reports of junk text (parts of files, etc) appearing after
4761 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4763 PH/22 Added support for macro redefinition, and (re)definition in between
4764 driver and ACL definitions.
4766 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4767 forgetting to use the resulting value; it was using the unexpanded value.
4769 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4770 hadn't been configured. The fix is from Juergen Kreileder, who
4771 understands it better than I do:
4773 "Here's what I see happening with three configured cyrus_sasl
4774 authenticators configured (plain, login, cram-md5):
4776 On startup auth_cyrus_sasl_init() gets called for each of these.
4777 This means three calls to sasl_listmech() without a specified mech_list.
4778 => SASL tests which mechs of all available mechs actually work
4779 => three warnings about OTP not working
4780 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4782 With the patch, sasl_listmech() also gets called three times. But now
4783 SASL's mech_list option is set to the server_mech specified in the the
4784 authenticator. Or in other words, the answer from sasl_listmech()
4785 gets limited to just the mech you're testing for (which is different
4787 => the return list contains just 'plain' or 'login', 'cram-md5' or
4788 nothing depending on the value of ob->server_mech.
4790 I've just tested the patch: Authentication still works fine,
4791 unavailable mechs specified in the exim configuration are still
4792 caught, and the auth.log warnings about OTP are gone."
4794 PH/25 When debugging is enabled, the contents of the command line are added
4795 to the debugging output, even when log_selector=+arguments is not
4798 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4799 answer is "GNU", and only if the return is "GNU/something" is the answer
4802 PH/27 $acl_verify_message is now set immediately after the failure of a
4803 verification in an ACL, and so is available in subsequent modifiers. In
4804 particular, the message can be preserved by coding like this:
4806 warn !verify = sender
4807 set acl_m0 = $acl_verify_message
4809 Previously, $acl_verify_message was set only while expanding "message"
4810 and "log_message" when a very denied access.
4812 PH/28 Modified OS/os.c-Linux with
4814 -#ifndef OS_LOAD_AVERAGE
4815 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4817 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4818 nomenclature these days.)
4820 PH/29 Installed patch from the Sieve maintainer that adds the options
4821 sieve_useraddress and sieve_subaddress to the redirect router.
4823 PH/30 In these circumstances:
4824 . Two addresses routed to the same list of hosts;
4825 . First host does not offer TLS;
4826 . First host accepts first address;
4827 . First host gives temporary error to second address;
4828 . Second host offers TLS and a TLS session is established;
4829 . Second host accepts second address.
4830 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4831 and peerdn, if requested) that were in fact used only for the second
4834 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4835 attention to any local part prefix or suffix that was matched by the
4836 router that accepted the address. It now behaves in the same way as it
4837 does for delivery: the affixes are removed from the local part unless
4838 rcpt_include_affixes is set on the transport.
4840 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4841 timeout during the DATA phase of an incoming message.
4843 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4844 applied a patch sanctioned by the Sieve maintainer.
4846 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4847 the uid or gid is negative. A case of a negative gid caused this to be
4848 noticed. The fix allows for either to be negative.
4850 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4851 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4854 PH/36 The change PH/12 above was broken. Fixed it.
4856 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4857 the grounds that routing the same address twice would always produce the
4858 same answer. This might have been true once, but it is certainly no
4859 longer true now. Routing a child address may depend on the previous
4860 routing that produced that child. Some complicated redirection strategies
4861 went wrong when messages had multiple recipients, and made Exim's
4862 behaviour dependent on the order in which the addresses were given.
4864 I have moved the duplicate checking until after the routing is complete.
4865 Exim scans the addresses that are assigned to local and remote
4866 transports, and removes any duplicates. This means that more work will be
4867 done, as duplicates will always all be routed, but duplicates are
4868 presumably rare, so I don't expect this is of any significance.
4870 For deliveries to pipes, files, and autoreplies, the duplicate checking
4871 still happens during the routing process, since they are not going to be
4874 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4875 It corrects a timeout issue with spamd. This is Ian's comment: "The
4876 background is that sometimes spamd either never reads data from a
4877 connection it has accepted, or it never writes response data. The exiscan
4878 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4879 blindly assumes that writes won't block so it may never time out."
4881 PH/39 Allow G after quota size as well as K and M.
4883 PH/40 The value set for $authenticated_id in an authenticator may not contain
4884 binary zeroes or newlines because the value is written to log lines and
4885 to spool files. There was no check on this. Now the value is run through
4886 the string_printing() function so that such characters are converted to
4887 printable escape sequences.
4889 PH/41 $message_linecount is a new variable that contains the total number of
4890 lines in the message. Compare $body_linecount, which is the count for the
4893 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4894 bounce and delay warning messages, except in certain special cases, which
4897 (a) An SMTP error message from a remote host;
4898 (b) A message specified in a :fail: redirection;
4899 (c) A message specified in a "fail" command in a system filter;
4900 (d) A message specified in a FAIL return from the queryprogram router;
4901 (e) A message specified by the cannot_route_message router option.
4903 In these cases only, Exim does include the error details in bounce and
4904 warning messages. There are also a few cases where bland messages such
4905 as "unrouteable address" or "local delivery error" are given.
4907 PH/43 $value is now also set for the "else" part of a ${run expansion.
4909 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4910 being worked on, but at least Exim now implements the latest version to
4913 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4914 process to complete was treated as a delivery failure, a timeout while
4915 writing the message to the pipe was logged, but erroneously treated as a
4916 successful delivery. Such timeouts include transport filter timeouts. For
4917 consistency with the overall process timeout, these timeouts are now
4918 treated as errors, giving rise to delivery failures by default. However,
4919 there is now a new Boolean option for the pipe transport called
4920 timeout_defer, which, if set TRUE, converts the failures into defers for
4921 both kinds of timeout. A transport filter timeout is now identified in
4924 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4925 systems where "make" and "gmake" are different, calling "gmake" at top
4926 level broke things. I've arranged for the value of $(MAKE) to be passed
4927 from the Makefile to this script so that it can call the same version of
4931 A note about Exim versions 4.44 and 4.50
4932 ----------------------------------------
4934 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4935 changes of various kinds. As a consequence, a big documentation update was
4936 needed. This delayed the release for rather longer than seemed good, especially
4937 in the light of a couple of (minor) security issues. Therefore, the changes
4938 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4939 release. So 4.44 and 4.50 are in effect two different branches that both start
4942 I have left the 4.50 change log unchanged; it contains all the changes since
4943 4.43. The change log for 4.44 is below; many of its items are identical to
4944 those for 4.50. This seems to be the most sensible way to preserve the
4945 historical information.
4951 1. Minor wording change to the doc/README.SIEVE file.
4953 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4954 computation of the current number of files was incorrect.
4956 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4957 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4960 4. Give more explanation in the error message when the command for a transport
4961 filter fails to execute.
4963 5. There are several places where Exim runs a non-Exim command in a
4964 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4965 was being done only for the command run by the queryprogram router. It is
4966 now done for all such subprocesses. The other cases are: ${run, transport
4967 filters, and the commands run by the lmtp and pipe transports.
4969 6. Added CONFIGURE_GROUP build-time option.
4971 7. Some older OS have a limit of 256 on the maximum number of file
4972 descriptors. Exim was using setrlimit() to set 1000 as a large value
4973 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4974 systems. I've change it so that if it can't get 1000, it tries for 256.
4976 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4977 was an oversight, and furthermore, ever since the addition of extra
4978 controls (e.g. 4.43/32), the checks on when to allow different forms of
4979 "control" were broken. There should now be diagnostics for all cases when a
4980 control that does not make sense is encountered.
4982 9. Added the /retain_sender option to "control=submission".
4984 10. $recipients is now available in the predata ACL (oversight).
4986 11. Tidy the search cache before the fork to do a delivery from a message
4987 received from the command line. Otherwise the child will trigger a lookup
4988 failure and thereby defer the delivery if it tries to use (for example) a
4989 cached ldap connection that the parent has called unbind on.
4991 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4992 of $address_data from the recipient verification was clobbered by the
4993 sender verification.
4995 13. The value of address_data from a sender verification is now available in
4996 $sender_address_data in subsequent conditions in the ACL statement.
4998 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
5000 15. Added a new option "connect=<time>" to callout options, to set a different
5003 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5004 was its contents. (It was OK if the option was not defined at all.)
5006 17. A "Completed" log line is now written for messages that are removed from
5007 the spool by the -Mrm option.
5009 18. New variables $sender_verify_failure and $recipient_verify_failure contain
5010 information about exactly what failed.
5012 19. Added -dd to debug only the daemon process.
5014 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
5015 handles timeouts, both on the server side and network timeouts. Renamed the
5016 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
5018 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
5019 It is now set to "smtps".
5021 22. $host_address is now set to the target address during the checking of
5022 ignore_target_hosts.
5024 23. When checking ignore_target_hosts for an ipliteral router, no host name was
5025 being passed; this would have caused $sender_host_name to have been used if
5026 matching the list had actually called for a host name (not very likely,
5027 since this list is usually IP addresses). A host name is now passed as
5030 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5031 code that specifies a non-restarting handler (typically sigaction() in
5032 modern systems) in an attempt to fix a rare and obscure crash bug.
5034 25. Narrowed the window for a race in the daemon that could cause it to ignore
5035 SIGCHLD signals. This is not a major problem, because they are used only to
5036 wake it up if nothing else does.
5038 26. A malformed maildirsize file could cause Exim to calculate negative values
5039 for the mailbox size or file count. Odd effects could occur as a result.
5040 The maildirsize information is now recalculated if the size or filecount
5043 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5044 support for a long time. Removed HAVE_SYS_VFS_H.
5046 28. Installed the latest version of exipick from John Jetmore.
5048 29. In an address list, if the pattern was not a regular expression, an empty
5049 subject address (from a bounce message) matched only if the pattern was an
5050 empty string. Non-empty patterns were not even tested. This was the wrong
5051 because it is perfectly reasonable to use an empty address as part of a
5052 database query. An empty address is now tested by patterns that are
5053 lookups. However, all the other forms of pattern expect the subject to
5054 contain a local part and a domain, and therefore, for them, an empty
5055 address still always fails if the pattern is not itself empty.
5057 30. Exim went into a mad DNS loop when attempting to do a callout where the
5058 host was specified on an smtp transport, and looking it up yielded more
5059 than one IP address.
5061 31. Re-factored the code for checking spool and log partition space into a
5062 function that finds that data and another that does the check. The former
5063 is then used to implement four new variables: $spool_space, $log_space,
5064 $spool_inodes, and $log_inodes.
5066 32. The RFC2047 encoding function was originally intended for short strings
5067 such as real names; it was not keeping to the 75-character limit for
5068 encoded words that the RFC imposes. It now respects the limit, and
5069 generates multiple encoded words if necessary. To be on the safe side, I
5070 have increased the buffer size for the ${rfc2047: expansion operator from
5073 33. It is now permitted to omit both strings after an "if" condition; if the
5074 condition is true, the result is "true". As before, when the second string
5075 is omitted, a false condition yields an empty string. This makes it less
5076 cumbersome to write custom ACL and router conditions.
5078 34. Failure to deliver a bounce message always caused it to be frozen, even if
5079 there was an errors_to setting on the router. The errors_to setting is now
5082 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5083 canonical form (fully expanded) before being placed in
5084 $sender_host_address.
5086 36. The table in the code that translates DNS record types into text (T_A to
5087 "A" for instance) was missing entries for NS and CNAME. It is just possible
5088 that this could have caused confusion if both these types were looked up
5089 for the same domain, because the text type is used as part of Exim's
5090 per-process caching. But the chance of anyone hitting this buglet seems
5093 37. The dnsdb lookup has been extended in a number of ways.
5095 (1) There is a new type, "zns", which walks up the domain tree until it
5096 finds some nameserver records. It should be used with care.
5098 (2) There is a new type, "mxh", which is like "mx" except that it returns
5099 just the host names, not the priorities.
5101 (3) It is now possible to give a list of domains (or IP addresses) to be
5102 looked up. The behaviour when one of the lookups defers can be
5103 controlled by a keyword.
5105 (4) It is now possible to specify the separator character for use when
5106 multiple records are returned.
5108 38. The dnslists ACL condition has been extended: it is now possible to supply
5109 a list of IP addresses and/or domains to be looked up in a particular DNS
5112 39. Added log_selector=+queue_time_overall.
5114 40. When running the queue in the test harness, wait just a tad after forking a
5115 delivery process, to get repeatability of debugging output.
5117 41. Include certificate and key file names in error message when GnuTLS fails
5118 to set them up, because the GnuTLS error message doesn't include the name
5119 of the failing file when there is a problem reading it.
5121 42. Allow both -bf and -bF in the same test run.
5123 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
5125 44. The "Exiscan patch" is now merged into the mainline Exim source.
5127 45. Sometimes the final signoff response after QUIT could fail to get
5128 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5129 < 0 before doing a fflush(). This bug looks as though it goes back to the
5130 introduction of TLS in release 3.20, but "sometimes" must have been rare
5131 because the tests only now provoked it.
5133 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
5134 (this can affect the format of dates).
5136 47. exim_tidydb, when checking for the continued existence of a message for
5137 which it has found a message-specific retry record, was not finding
5138 messages that were in split spool directories. Consequently, it was
5139 deleting retry records that should have stayed in existence.
5141 48. Steve fixed some bugs in eximstats.
5143 49. The SPA authentication driver was not abandoning authentication and moving
5144 on to the next authenticator when an expansion was forced to fail,
5145 contradicting the general specification for all authenticators. Instead it
5146 was generating a temporary error. It now behaves as specified.
5148 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
5149 (the order specifies the preference for clients). The order is now AES256,
5150 AES128, 3DES, ARCFOUR128.
5152 51. Small patch to Sieve code - explicitly set From: when generating an
5155 52. Exim crashed if a remote delivery caused a very long error message to be
5156 recorded - for instance if somebody sent an entire SpamAssassin report back
5157 as a large number of 550 error lines. This bug was coincidentally fixed by
5158 increasing the size of one of Exim's internal buffers (big_buffer) that
5159 happened as part of the Exiscan merge. However, to be on the safe side, I
5160 have made the code more robust (and fixed the comments that describe what
5163 53. Now that there can be additional text after "Completed" in log lines (if
5164 the queue_time_overall log selector is set), a one-byte patch to exigrep
5165 was needed to allow it to recognize "Completed" as not the last thing in
5168 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
5169 patch that reportedly fixes this has been added. I am not expert enough to
5170 create a test for it. This is what the patch creator wrote:
5172 "I found a little strange behaviour of ldap code when working with
5173 Windows 2003 AD Domain, where users was placed in more than one
5174 Organization Units. When I tried to give exim partial DN, the exit code
5175 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
5176 But simultaneously result of request was absolutely normal ldap result,
5177 so I produce this patch..."
5179 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
5180 REFERENCE, so I have modified the code to exclude the patch when that macro
5183 55. Some experimental protocols are using DNS PTR records for new purposes. The
5184 keys for these records are domain names, not reversed IP addresses. The
5185 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5186 leaves it alone. Component reversal etc. now happens only for IP addresses.
5189 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
5191 57. Double the size of the debug message buffer (to 2048) so that more of very
5192 long debug lines gets shown.
5194 58. The exicyclog utility now does better if the number of log files to keep
5195 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
5197 59. Two changes related to the smtp_active_hostname option:
5199 (1) $smtp_active_hostname is now available as a variable.
5200 (2) The default for smtp_banner uses $smtp_active_hostname instead
5201 of $primary_hostname.
5203 60. The host_aton() function is supposed to be passed a string that is known
5204 to be a valid IP address. However, in the case of IPv6 addresses, it was
5205 not checking this. This is a hostage to fortune. Exim now panics and dies
5206 if the condition is not met. A case was found where this could be provoked
5207 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5208 components; fortuitously, this particular loophole had already been fixed
5209 by change 4.50/55 above.
5211 If there are any other similar loopholes, the new check in host_aton()
5212 itself should stop them being exploited. The report I received stated that
5213 data on the command line could provoke the exploit when Exim was running as
5214 exim, but did not say which command line option was involved. All I could
5215 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5216 running as the user.
5219 61. There was a buffer overflow vulnerability in the SPA authentication code
5220 (which came originally from the Samba project). I have added a test to the
5221 spa_base64_to_bits() function which I hope fixes it.
5224 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
5225 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
5227 63. The daemon start-up calls getloadavg() while still root for those OS that
5228 need the first call to be done as root, but it missed one case: when
5229 deliver_queue_load_max is set with deliver_drop_privilege. This is
5230 necessary for the benefit of the queue runner, because there is no re-exec
5231 when deliver_drop_privilege is set.
5233 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
5234 This has been fixed.
5236 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5237 was in use, was not putting the data itself into the right store pool;
5238 consequently, it could be overwritten for a subsequent message in the same
5239 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5242 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
5244 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
5245 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
5246 the function were treating the return as a boolean value, which happened to
5247 work because 0=false and not-0=true, but is not correct code.
5249 68. The host_aton() function was not handling scoped IPv6 addresses (those
5250 with, for example, "%eth0" on the end) correctly.
5252 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
5253 negated items (that is, ~something) in unsigned ints. Some compilers
5254 apparently mutter when there is no cast.
5256 70. If an address verification called from an ACL failed, and did not produce a
5257 user-specific message (i.e. there was only a "system" message), nothing was
5258 put in $acl_verify_message. In this situation, it now puts the system
5261 71. Change 4.23/11 added synchronization checking at the start of an SMTP
5262 session; change 4.31/43 added the unwanted input to the log line - except
5263 that it did not do this in the start of session case. It now does.
5265 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
5266 This is wrong and can cause the other end to generate a synchronization
5267 error if it is another Exim or anything else that does the synchronization
5268 check. A QUIT command is no longer sent after a timeout.
5270 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
5271 during host lookups.
5273 74. The defer_ok option of callout verification was not working if it was used
5274 when verifying addresses in header lines, that is, for this case:
5276 verify = header_sender/callout=defer_ok
5278 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
5279 those file descriptors could be used for SMTP connections. If anything
5280 wrote to stderr (the example that came up was "warn" in embedded Perl), it
5281 could be sent to the SMTP client, causing chaos. The daemon now opens
5282 stdin, stdout, and stderr to /dev/null when it puts itself into the
5285 76. Arrange for output from Perl's "warn" command to be written to Exim's main
5286 log by default. The user can override this with suitable Perl magic.
5288 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
5289 the log message when discard triggers, was not working for the DATA ACL or
5290 for the non-SMTP ACL.
5292 78. Error message wording change in sieve.c.
5294 79. If smtp_accept_max_per_host was set, the number of connections could be
5295 restricted to fewer than expected, because the daemon was trying to set up
5296 a new connection before checking whether the processes handling previous
5297 connections had finished. The check for completed processes is now done
5298 earlier. On busy systems, this bug wouldn't be noticed because something
5299 else would have woken the daemon, and it would have reaped the completed
5302 80. If a message was submitted locally by a user whose login name contained one
5303 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
5304 It caused a spool format error. I have fixed the spool reading code. A
5305 related problem was that the "from" clause in the Received: line became
5306 illegal because of the space(s). It is now covered by ${quote_local_part.
5308 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
5311 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
5313 83. Patch from Sieve maintainer to fix unterminated string problem in
5314 "vacation" handling.
5316 84. Some minor changes to the Linux configuration files to help with other
5317 OS variants using glibc.
5319 85. One more patch for Sieve to update vacation handling to latest spec.
5322 ----------------------------------------------------
5323 See the note above about the 4.44 and 4.50 releases.
5324 ----------------------------------------------------
5330 1. Change 4.43/35 introduced a bug that caused file counts to be
5331 incorrectly computed when quota_filecount was set in an appendfile
5334 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5335 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5338 3. Give more explanation in the error message when the command for a transport
5339 filter fails to execute.
5341 4. There are several places where Exim runs a non-Exim command in a
5342 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5343 was being done only for the command run by the queryprogram router. It is
5344 now done for all such subprocesses. The other cases are: ${run, transport
5345 filters, and the commands run by the lmtp and pipe transports.
5347 5. Some older OS have a limit of 256 on the maximum number of file
5348 descriptors. Exim was using setrlimit() to set 1000 as a large value
5349 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5350 systems. I've change it so that if it can't get 1000, it tries for 256.
5352 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5353 was an oversight, and furthermore, ever since the addition of extra
5354 controls (e.g. 4.43/32), the checks on when to allow different forms of
5355 "control" were broken. There should now be diagnostics for all cases when a
5356 control that does not make sense is encountered.
5358 7. $recipients is now available in the predata ACL (oversight).
5360 8. Tidy the search cache before the fork to do a delivery from a message
5361 received from the command line. Otherwise the child will trigger a lookup
5362 failure and thereby defer the delivery if it tries to use (for example) a
5363 cached ldap connection that the parent has called unbind on.
5365 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5366 of $address_data from the recipient verification was clobbered by the
5367 sender verification.
5369 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5370 was its contents. (It was OK if the option was not defined at all.)
5372 11. A "Completed" log line is now written for messages that are removed from
5373 the spool by the -Mrm option.
5375 12. $host_address is now set to the target address during the checking of
5376 ignore_target_hosts.
5378 13. When checking ignore_target_hosts for an ipliteral router, no host name was
5379 being passed; this would have caused $sender_host_name to have been used if
5380 matching the list had actually called for a host name (not very likely,
5381 since this list is usually IP addresses). A host name is now passed as
5384 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5385 code that specifies a non-restarting handler (typically sigaction() in
5386 modern systems) in an attempt to fix a rare and obscure crash bug.
5388 15. Narrowed the window for a race in the daemon that could cause it to ignore
5389 SIGCHLD signals. This is not a major problem, because they are used only to
5390 wake it up if nothing else does.
5392 16. A malformed maildirsize file could cause Exim to calculate negative values
5393 for the mailbox size or file count. Odd effects could occur as a result.
5394 The maildirsize information is now recalculated if the size or filecount
5397 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5398 support for a long time. Removed HAVE_SYS_VFS_H.
5400 18. Updated exipick to current release from John Jetmore.
5402 19. Allow an empty sender to be matched against a lookup in an address list.
5403 Previously the only cases considered were a regular expression, or an
5406 20. Exim went into a mad DNS lookup loop when doing a callout where the
5407 host was specified on the transport, if the DNS lookup yielded more than
5410 21. The RFC2047 encoding function was originally intended for short strings
5411 such as real names; it was not keeping to the 75-character limit for
5412 encoded words that the RFC imposes. It now respects the limit, and
5413 generates multiple encoded words if necessary. To be on the safe side, I
5414 have increased the buffer size for the ${rfc2047: expansion operator from
5417 22. Failure to deliver a bounce message always caused it to be frozen, even if
5418 there was an errors_to setting on the router. The errors_to setting is now
5421 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5422 canonical form (fully expanded) before being placed in
5423 $sender_host_address.
5425 24. Updated eximstats to version 1.33
5427 25. Include certificate and key file names in error message when GnuTLS fails
5428 to set them up, because the GnuTLS error message doesn't include the name
5429 of the failing file when there is a problem reading it.
5431 26. Expand error message when OpenSSL has problems setting up cert/key files.
5434 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5435 (this can affect the format of dates).
5437 28. exim_tidydb, when checking for the continued existence of a message for
5438 which it has found a message-specific retry record, was not finding
5439 messages that were in split spool directories. Consequently, it was
5440 deleting retry records that should have stayed in existence.
5442 29. eximstats updated to version 1.35
5443 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5444 1.35 - bugfix such that pie charts by volume are generated correctly
5446 30. The SPA authentication driver was not abandoning authentication and moving
5447 on to the next authenticator when an expansion was forced to fail,
5448 contradicting the general specification for all authenticators. Instead it
5449 was generating a temporary error. It now behaves as specified.
5451 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5452 (the order specifies the preference for clients). The order is now AES256,
5453 AES128, 3DES, ARCFOUR128.
5455 31. Small patch to Sieve code - explicitly set From: when generating an
5458 32. Exim crashed if a remote delivery caused a very long error message to be
5459 recorded - for instance if somebody sent an entire SpamAssassin report back
5460 as a large number of 550 error lines. This bug was coincidentally fixed by
5461 increasing the size of one of Exim's internal buffers (big_buffer) that
5462 happened as part of the Exiscan merge. However, to be on the safe side, I
5463 have made the code more robust (and fixed the comments that describe what
5466 33. Some experimental protocols are using DNS PTR records for new purposes. The
5467 keys for these records are domain names, not reversed IP addresses. The
5468 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5469 leaves it alone. Component reversal etc. now happens only for IP addresses.
5472 34. The host_aton() function is supposed to be passed a string that is known
5473 to be a valid IP address. However, in the case of IPv6 addresses, it was
5474 not checking this. This is a hostage to fortune. Exim now panics and dies
5475 if the condition is not met. A case was found where this could be provoked
5476 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5477 components; fortuitously, this particular loophole had already been fixed
5478 by change 4.50/55 or 4.44/33 above.
5480 If there are any other similar loopholes, the new check in host_aton()
5481 itself should stop them being exploited. The report I received stated that
5482 data on the command line could provoke the exploit when Exim was running as
5483 exim, but did not say which command line option was involved. All I could
5484 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5485 running as the user.
5488 35. There was a buffer overflow vulnerability in the SPA authentication code
5489 (which came originally from the Samba project). I have added a test to the
5490 spa_base64_to_bits() function which I hope fixes it.
5493 36. The daemon start-up calls getloadavg() while still root for those OS that
5494 need the first call to be done as root, but it missed one case: when
5495 deliver_queue_load_max is set with deliver_drop_privilege. This is
5496 necessary for the benefit of the queue runner, because there is no re-exec
5497 when deliver_drop_privilege is set.
5499 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5500 was in use, was not putting the data itself into the right store pool;
5501 consequently, it could be overwritten for a subsequent message in the same
5502 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5505 38. Sometimes the final signoff response after QUIT could fail to get
5506 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5507 < 0 before doing a fflush(). This bug looks as though it goes back to the
5508 introduction of TLS in release 3.20, but "sometimes" must have been rare
5509 because the tests only now provoked it.
5515 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5516 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5517 Now it returns an integer. A number of calls were still expecting a T/F
5518 return. Fortuitously, in all cases, the tests worked in OK situations,
5519 which is the norm. However, things would have gone wrong on any write
5520 failures on the smtp file descriptor. This function is used when sending
5521 messages over SMTP and also when doing verify callouts.
5523 2. When Exim is called to do synchronous delivery of a locally submitted
5524 message (the -odf or -odi options), it no longer closes stderr before doing
5527 3. Implemented the mua_wrapper option.
5529 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5531 5. Implemented the functions header_remove(), header_testname(),
5532 header_add_at_position(), and receive_remove_recipient(), and exported them
5535 6. If an ACL "warn" statement specified the addition of headers, Exim already
5536 inserted X-ACL-Warn: at the start if there was no header name. However, it
5537 was not making this test for the second and subsequent header lines if
5538 there were newlines in the string. This meant that an invalid header could
5539 be inserted if Exim was badly configured.
5541 7. Allow an ACL "warn" statement to add header lines at the start or after all
5542 the Received: headers, as well as at the end.
5544 8. Added the rcpt_4xx retry error code.
5546 9. Added postmaster_mailfrom=xxx to callout verification option.
5548 10. Added mailfrom=xxxx to the callout verification option, for verify=
5551 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5552 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5554 12. Inserted some casts to stop certain compilers warning when using pointer
5555 differences as field lengths or precisions in printf-type calls (mostly
5556 affecting debugging statements).
5558 13. Added optional readline() support for -be (dynamically loaded).
5560 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5561 same clock tick as a message's arrival, so that its received time was the
5562 same as the "first fail" time on the retry record, and that message
5563 remained on the queue past the ultimate address timeout, every queue runner
5564 would try a delivery (because it was past the ultimate address timeout) but
5565 after another failure, the ultimate address timeout, which should have then
5566 bounced the address, did not kick in. This was a "< instead of <=" error;
5567 in most cases the first failure would have been in the next clock tick
5568 after the received time, and all would be well.
5570 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5571 being recognized when the domain list was tested by the match_domain
5572 condition in an expansion string.
5574 16. Added the ${str2b64: operator.
5576 17. Exim was always calling setrlimit() to set a large limit for the number of
5577 processes, without checking whether the existing limit was already
5578 adequate. (It did check for the limit on file descriptors.) Furthermore,
5579 errors from getrlimit() and setrlimit() were being ignored. Now they are
5580 logged to the main and panic logs, but Exim does carry on, to try to do its
5581 job under whatever limits there are.
5583 18. Imported PCRE 5.0.
5585 19. Trivial typo in log message " temporarily refused connection" (the leading
5588 20. If the log selector return_path_on_delivery was set and an address was
5589 redirected to /dev/null, the delivery process crashed because it assumed
5590 that a return path would always be set for a "successful" delivery. In this
5591 case, the whole delivery is bypassed as an optimization, and therefore no
5594 21. Internal re-arrangement: the function for sending a challenge and reading
5595 a response while authentication was assuming a zero-terminated challenge
5596 string. It's now changed to take a pointer and a length, to allow for
5597 binary data in such strings.
5599 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5601 23. Exim was not respecting finduser_retries when seeking the login of the
5602 uid under which it was called; it was always trying 10 times. (The default
5603 setting of finduser_retries is zero.) Also, it was sleeping after the final
5604 failure, which is pointless.
5606 24. Implemented tls_on_connect_ports.
5608 25. Implemented acl_smtp_predata.
5610 26. If the domain in control=submission is set empty, Exim assumes that the
5611 authenticated id is a complete email address when it generates From: or
5612 Sender: header lines.
5614 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5615 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5616 chown and chgrp in /bin and hostname in /usr/bin.
5618 28. Exim was keeping the "process log" file open after each use, just as it
5619 does for the main log. This opens the possibility of it remaining open for
5620 long periods when the USR1 signal hits a daemon. Occasional processlog
5621 errors were reported, that could have been caused by this. Anyway, it seems
5622 much more sensible not to leave this file open at all, so that is what now
5625 29. The long-running daemon process does not normally write to the log once it
5626 has entered its main loop, and it closes the log before doing so. This is
5627 so that log files can straightforwardly be renamed and moved. However,
5628 there are a couple of unusual error situations where the daemon does write
5629 log entries, and I had neglected to close the log afterwards.
5631 30. The text of an SMTP error response that was received during a remote
5632 delivery was being truncated at 512 bytes. This is too short for some of
5633 the long messages that one sometimes sees. I've increased the limit to
5636 31. It is now possible to make retry rules that apply only when a message has a
5637 specific sender, in particular, an empty sender.
5639 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5640 it possible to be selective about when SMTP synchronization is enforced.
5642 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5644 32. Implemented hosts_connection_nolog.
5646 33. Added an ACL for QUIT.
5648 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5651 35. Added mailbox_size and mailbox_filecount to appendfile.
5653 36. Added control = no_multiline_responses to ACLs.
5655 37. There was a bug in the logic of the code that waits for the clock to tick
5656 in the case where the clock went backwards by a substantial amount such
5657 that the microsecond fraction of "now" was more than the microsecond
5658 fraction of "then" (but the whole seconds number was less).
5660 38. Added support for the libradius Radius client library this is found on
5661 FreeBSD (previously only the radiusclient library was supported).
5667 1. When certain lookups returned multiple values in the form name=value, the
5668 quoting of the values was not always being done properly. Specifically:
5669 (a) If the value started with a double quote, but contained no whitespace,
5671 (b) If the value contained whitespace other than a space character (i.e.
5672 tabs or newlines or carriage returns) it was not quoted.
5673 This fix has been applied to the mysql and pgsql lookups by writing a
5674 separate quoting function and calling it from the lookup code. The fix
5675 should probably also be applied to nisplus, ibase and oracle lookups, but
5676 since I cannot test any of those, I have not disturbed their existing code.
5678 2. A hit in the callout cache for a specific address caused a log line with no
5679 reason for rejecting RCPT. Now it says "Previous (cached) callout
5680 verification failure".
5682 3. There was an off-by-one bug in the queryprogram router. An over-long
5683 return line was truncated at 256 instead of 255 characters, thereby
5684 overflowing its buffer with the terminating zero. As well as fixing this, I
5685 have increased the buffer size to 1024 (and made a note to document this).
5687 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5688 when Exim is waiting for an SMTP response from a remote server, Exim
5689 restarts its select() call on the socket, thereby resetting its timeout.
5690 This is not a problem when such interrupts are rare. Somebody set up a cron
5691 job to run exiwhat every 2 minutes, which is less than the normal select()
5692 timeout (5 or 10 minutes). This meant that the select() timeout never
5693 kicked in because it was always reset. I have fixed this by comparing the
5694 time when an interrupt arrives with the time at the start of the first call
5695 to select(). If more time than the timeout has elapsed, the interrupt is
5696 treated as a timeout.
5698 5. Some internal re-factoring in preparation for the addition of Sieve
5699 extensions (by MH). In particular, the "personal" test is moved to a
5700 separate function, and given an option for scanning Cc: and Bcc: (which is
5701 not set for Exim filters).
5703 6. When Exim created an email address using the login of the caller as the
5704 local part (e.g. when creating a From: or Sender: header line), it was not
5705 quoting the local part when it contained special characters such as @.
5707 7. Installed new OpenBSD configuration files.
5709 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5710 try to make them clearer.
5712 9. Callout options, other than the timeout value, were being ignored when
5713 verifying sender addresses in header lines. For example, when using
5715 verify = header_sender/callout=no_cache
5717 the cache was (incorrectly) being used.
5719 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5720 only the Cygwin environment.
5722 11. When return_path_on_delivery was set as a log selector, if different remote
5723 addresses in the same message used different return paths and parallel
5724 remote delivery occurred, the wrong values would sometimes be logged.
5725 (Whenever a remote delivery process finished, the return path value from
5726 the most recently started remote delivery process was logged.)
5728 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5729 lines when AUTH and/or TLS are in use. This is the "received protocol"
5730 field. Exim used to use "asmtp" for authenticated SMTP, without any
5731 indication (in the protocol name) for TLS use. Now it follows the RFC and
5732 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5733 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5734 names appear in log lines as well as in Received: header lines.
5736 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5737 extensions, and comparison tests, and to fix some bugs.
5739 14. Changes to the "personal" filter test:
5741 (1) The test was buggy in that it was just doing the equivalent of
5742 "contains" tests on header lines. For example, if a user's address was
5743 anne@some.where, the "personal" test would incorrectly be true for
5745 To: susanne@some.where
5747 This test is now done by extracting each address from the header in turn,
5748 and checking the entire address. Other tests that are part of "personal"
5749 are now done using regular expressions (for example, to check local parts
5750 of addresses in From: header lines).
5752 (2) The list of non-personal local parts in From: addresses has been
5753 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5754 taken from the Sieve specification recommendations.
5756 (3) If the message contains any header line starting with "List-" it is
5757 treated as non-personal.
5759 (4) The test for "circular" in the Subject: header line has been removed
5760 because it now seems ill-conceived.
5762 15. Minor typos in src/EDITME comments corrected.
5764 16. Installed latest exipick from John Jetmore.
5766 17. If headers_add on a router specified a text string that was too long for
5767 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5768 of string_sprintf() is now avoided.
5770 18. $message_body_size was not set (it was always zero) when running the DATA
5771 ACL and the local_scan() function.
5773 19. For the "mail" command in an Exim filter, no default was being set for
5774 the once_repeat time, causing a random time value to be used if "once" was
5775 specified. (If the value happened to be <= 0, no repeat happened.) The
5776 default is now 0s, meaning "never repeat". The "vacation" command was OK
5777 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5778 (I found it when inspecting the code).
5780 20. There is now an overall timeout for performing a callout verification. It
5781 defaults to 4 times the callout timeout, which applies to individual SMTP
5782 commands during the callout. The overall timeout applies when there is more
5783 than one host that can be tried. The timeout is checked before trying the
5784 next host. This prevents very long delays if there are a large number of
5785 hosts and all are timing out (e.g. when the network connections are timing
5786 out). The value of the overall timeout can be changed by specifying an
5787 additional sub-option for "callout", called "maxwait". For example:
5789 verify = sender/callout=5s,maxwait=20s
5791 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5792 to the end before writing, but this should make it even safer).
5794 22. Exim was forgetting that it had advertised PIPELINING for the second and
5795 subsequent messages on an SMTP connection. It was also not resetting its
5796 memory on STARTTLS and an internal HELO.
5798 23. When Exim logs an SMTP synchronization error within a session, it now
5799 records whether PIPELINING has been advertised or not.
5801 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5802 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5803 rather than long int.
5805 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5807 26. Added the never_mail option to autoreply.
5813 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5814 crash if the getsockname() call failed; this can happen if a connection is
5815 closed very soon after it is established. The problem was simply in the
5816 order in which certain operations were done, causing Exim to try to write
5817 to the SMTP stream before it had set up the file descriptor. The bug has
5818 been fixed by making things happen in the correct order.
5824 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5825 before the connection was closed, thus losing the rejection response.
5827 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5828 some early Solaris releases, but causes trouble in current releases where
5829 socklen_t is defined.
5831 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5834 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5837 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5838 that releasing the top of it at the end releases what was used for sub-
5839 expansions (unless the block got too big). However, discard this block if
5840 the first thing is a variable or header, so that we can use its block when
5841 it is dynamic (useful for very large $message_headers, for example).
5843 6. Lookups now cache *every* query, not just the most recent. A new, separate
5844 store pool is used for this. It can be recovered when all lookup caches are
5845 flushed. Lookups now release memory at the end of their result strings.
5846 This has involved some general refactoring of the lookup sources.
5848 7. Some code has been added to the store_xxx() functions to reduce the amount
5849 of flapping under certain conditions.
5851 8. log_incoming_interface used to affect only the <= reception log lines. Now
5852 it causes the local interface and port to be added to several more SMTP log
5853 lines, for example "SMTP connection from", and rejection lines.
5855 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5857 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5859 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5860 could be overwritten at the end of the current message (or the start of a
5861 new message if it was set in a HELO ACL). The value is now preserved for
5862 the duration of the SMTP connection.
5864 12. If a transport had a headers_rewrite setting, and a matching header line
5865 contained an unqualified address, that address was qualified, even if it
5866 did not match any rewriting rules. The underlying bug was that the values
5867 of the flags that permit the existence of unqualified sender and recipient
5868 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5869 non-local messages, and by -bnq for local messages) were not being
5870 preserved with the message after it was received.
5872 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5873 "next input=" as part of the text comprising the host identity instead of
5874 the correct text. The code was using the same buffer for two different
5875 strings. However, depending on which order the printing function evaluated
5876 its arguments, the bug did not always show up. Under Linux, for example, my
5877 test suite worked just fine.
5879 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5880 doesn't work with some older versions of Perl. It has been changed to "my",
5881 which in any case is probably the better facility to use.
5883 15. A really picky compiler found some instances of statements for creating
5884 error messages that either had too many or two few arguments for the format
5887 16. The size of the buffer for calls to the DNS resolver has been increased
5888 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5889 for addresses that have a lot of PTR records. This alleviates a problem; it
5890 does not fully solve it.
5892 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5893 buffer now truncates the list and logs the incident, which is the same
5894 action as happens when Exim is looking up a host name and its aliases.
5895 Previously in this situation something unpredictable would happen;
5896 sometimes it was "internal error: store_reset failed".
5898 18. If a server dropped the connection unexpectedly when an Exim client was
5899 using GnuTLS and trying to read a response, the client delivery process
5900 crashed while trying to generate an error log message.
5902 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5903 string, for example:
5905 warn message = H1: something\nH2: something
5907 the text was added as a single header line from Exim's point of view
5908 though it ended up OK in the delivered message. However, searching for the
5909 second and subsequent header lines using $h_h2: did not work. This has been
5910 fixed. Similarly, if a system filter added multiple headers in this way,
5911 the routers could not see them.
5913 20. Expanded the error message when iplsearch is called with an invalid key to
5914 suggest using net-iplsearch in a host list.
5916 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5917 ACLs are no longer actually imposed (and a message to that effect is
5920 22. If a "gecos" field in a passwd entry contained escaped characters, in
5921 particular, if it contained a \" sequence, Exim got it wrong when building
5922 a From: or a Sender: header from that name. A second bug also caused
5923 incorrect handling when an unquoted " was present following a character
5924 that needed quoting.
5926 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5927 was not being matched caselessly.
5929 24. Arranged for all hyphens in the exim.8 source to be escaped with
5932 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5933 properly. Recipient callout cache records were still being keyed to include
5934 the sender, even when use_sender was set false. This led to far more
5935 callouts that were necessary. The sender is no longer included in the key
5936 when use_sender is false.
5938 26. Added "control = submission" modifier to ACLs.
5940 27. Added the ${base62d: operator to decode base 62 numbers.
5942 28. dnsdb lookups can now access SRV records.
5944 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5945 the configuration file.
5947 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5948 (-v) mode. This makes the output for a verbose queue run more intelligible.
5950 31. Added a use_postmaster feature to recipient callouts.
5952 32. Added the $body_zerocount variable, containing the number of binary zero
5953 bytes in the message body.
5955 33. The time of last modification of the "new" subdirectory is now used as the
5956 "mailbox time last read" when there is a quota error for a maildir
5959 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5961 35. Added +ignore_unknown as a special item in host lists.
5963 36. Code for decoding IPv6 addresses in host lists is now included, even if
5964 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5965 address was recognized as an IP address, but was then not correctly decoded
5966 into binary, causing unexpected and incorrect effects when compared with
5973 1. Very minor rewording of debugging text in manualroute to say "list of
5974 hosts" instead of "hostlist".
5976 2. If verify=header_syntax was set, and a header line with an unqualified
5977 address (no domain) and a large number of spaces between the end of the
5978 name and the colon was received, the reception process suffered a buffer
5979 overflow, and (when I tested it) crashed. This was caused by some obsolete
5980 code that should have been removed. The fix is to remove it!
5982 3. When running in the test harness, delay a bit after writing a bounce
5983 message to get a bit more predictability in the log output.
5985 4. Added a call to search_tidyup() just before forking a reception process. In
5986 theory, someone could use a lookup in the expansion of smtp_accept_max_
5987 per_host which, without the tidyup, could leave open a database connection.
5989 5. Added the variables $recipient_data and $sender_data which get set from a
5990 lookup success in an ACL "recipients" or "senders" condition, or a router
5991 "senders" option, similar to $domain_data and $local_part_data.
5993 6. Moved the writing of debug_print from before to after the "senders" test
5996 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5997 problems for message scanning, either using a data ACL, or using
5998 local_scan() because the Received: header was not generated till after they
5999 were called (in order to set the time as the time of reception completion).
6000 I have revised the way this works. The header is now generated after the
6001 body is received, but before the ACL or local_scan() are called. After they
6002 are run, the timestamp in the header is updated.
6008 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
6009 before starting a queue runner without re-exec. This happened only when
6010 deliver_drop_privilege was set or when the Exim user was set to root. The
6011 effect of the bug was that timeouts during subsequent deliveries caused
6012 crashes instead of being properly handled. The handler is now left at its
6013 default (and expected) setting.
6015 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
6016 message, again when deliver_drop_privilege is set or Exim is run as root.
6017 The bug described in (1) was not present in this case, but the tidying up
6018 of the other signals was missing. I have made the two cases consistent.
6020 3. The ignore_target_hosts setting on a manualroute router was being ignored
6021 for hosts that were looked up using the /MX notation.
6023 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
6026 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
6027 operated on the sender address. After changing the $sender_address to <>
6028 for the sender address verify, Exim was re-instated it as the original
6029 (before rewriting) address, but remembering that it had rewritten it, so it
6030 wasn't rewriting it again. This bug also had the effect of breaking the
6031 sender address verification caching when the sender address was rewritten.
6033 6. The ignore_target_hosts option was being ignored by the ipliteral router.
6034 This has been changed so that if the ip literal address matches
6035 ignore_target_hosts, the router declines.
6037 7. Added expansion conditions match_domain, match_address, and match_local_
6038 part (NOT match_host).
6040 8. The placeholder for the Received: header didn't have a length field set.
6042 9. Added code to Exim itself and to exim_lock to test for a specific race
6043 condition that could lead to file corruption when using MBX delivery. The
6044 issue is with the lockfile that is created in /tmp. If this file is removed
6045 after a process has opened it but before that process has acquired a lock,
6046 there is the potential for a second process to recreate the file and also
6047 acquire a lock. This could lead to two Exim processes writing to the file
6048 at the same time. The added code performs the same test as UW imapd; it
6049 checks after acquiring the lock that its file descriptor still refers to
6050 the same named file.
6052 10. The buffer for building added header lines was of fixed size, 8192 bytes.
6053 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
6056 11. Added the smtp_active_hostname option. If used, this will typically be made
6057 to depend on the incoming interface address. Because $interface_address is
6058 not set up until the daemon has forked a reception process, error responses
6059 that can happen earlier (such as "too many connections") no longer contain
6062 12. If an expansion in a condition on a "warn" statement fails because a lookup
6063 defers, the "warn" statement is abandoned, and the next ACL statement is
6064 processed. Previously this caused the whole ACL to be aborted.
6066 13. Added the iplsearch lookup type.
6068 14. Added ident_timeout as a log selector.
6070 15. Added tls_certificate_verified as a log selector.
6072 16. Added a global option tls_require_ciphers (compare the smtp transport
6073 option of the same name). This controls incoming TLS connections.
6075 17. I finally figured out how to make tls_require_ciphers do a similar thing
6076 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
6077 before starting the TLS session.
6079 18. Tabs are now shown as \t in -bP output.
6081 19. If the log selector return_path_on_delivery was set, Exim crashed when
6082 bouncing a message because it had too many Received: header lines.
6084 20. If two routers both had headers_remove settings, and the first one included
6085 a superfluous trailing colon, the final name in the first list and the
6086 first name in the second list were incorrectly joined into one item (with a
6087 colon in the middle).
6093 1. Added -C and -D options to the exinext utility, mainly to make it easier
6094 to include in the automated testing, but these could be helpful when
6095 multiple configurations are in use.
6097 2. The exinext utility was not formatting the output nicely when there was
6098 an alternate port involved in the retry record key, nor when there was a
6099 message id as well (for retries that were specific to a specific message
6100 and a specific host). It was also confused by IPv6 addresses, because of
6101 the additional colons they contain. I have fixed the IPv4 problem, and
6102 patched it up to do a reasonable job for IPv6.
6104 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
6105 delivery, the log line now contains "pipelined" if PIPELINING was used.
6107 4. An SMTP transport process used to panic and die if the bind() call to set
6108 an explicit outgoing interface failed. This has been changed; it is now
6109 treated in the same way as a connect() failure.
6111 5. A reference to $sender_host_name in the part of a conditional expansion
6112 that was being skipped was still causing a DNS lookup. This no longer
6115 6. The def: expansion condition was not recognizing references to header lines
6116 that used bh_ and bheader_.
6118 7. Added the _cache feature to named lists.
6120 8. The code for checking quota_filecount in the appendfile transport was
6121 allowing one more file than it should have been.
6123 9. For compatibility with Sendmail, the command line option
6131 and sets the incoming protocol and host name (for trusted callers). The
6132 host name and its colon can be omitted when only the protocol is to be set.
6133 Note the Exim already has two private options, -pd and -ps, that refer to
6134 embedded Perl. It is therefore impossible to set a protocol value of "d" or
6135 "s", but I don't think that's a major issue.
6137 10. A number of refactoring changes to the code, none of which should affect
6140 (a) The number of logging options was getting close to filling up the
6141 32-bit word that was used as a bit map. I have split them into two classes:
6142 those that are passed in the argument to log_write(), and those that are
6143 only ever tested independently outside of that function. These are now in
6144 separate 32-bit words, so there is plenty of room for expansion again.
6145 There is no change in the user interface or the logging behaviour.
6147 (b) When building, for example, log lines, the code previously used a
6148 macro that called string_cat() twice, in order to add two strings. This is
6149 not really sufficiently general. Furthermore, there was one instance where
6150 it was actually wrong because one of the argument was used twice, and in
6151 one call a function was used. (As it happened, calling the function twice
6152 did not affect the overall behaviour.) The macro has been replaced by a
6153 function that can join an arbitrary number of extra strings onto a growing
6156 (c) The code for expansion conditions now uses a table and a binary chop
6157 instead of a serial search (which was left over from when there were very
6158 few conditions). Also, it now recognizes conditions like "pam" even when
6159 the relevant support is not compiled in: a suitably worded error message is
6160 given if an attempt is made to use such a condition.
6162 11. Added ${time_interval:xxxxx}.
6164 12. A bug was causing one of the ddress fields not to be passed back correctly
6165 from remote delivery subprocesses. The field in question was not being
6166 subsequently used, so this caused to problems in practice.
6168 13. Added new log selectors queue_time and deliver_time.
6170 14. Might have fixed a bug in maildirsizefile handling that threw up
6171 "unexpected character" debug warnings, and recalculated the data
6172 unnecessarily. In any case, I expanded the warning message to give more
6175 15. Added the message "Restricted characters in address" to the statements in
6176 the default ACL that block characters like @ and % in local parts.
6178 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
6179 Three changes have been made:
6181 (a) There was a serious bug; a negative response to MAIL caused the whole
6182 recipient domain to be cached as invalid, thereby blocking all messages
6183 to all local parts at the same domain, from all senders. This bug has
6184 been fixed. The domain is no longer cached after a negative response to
6185 MAIL if the sender used is not empty.
6187 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
6190 (c) A new callout option, "use_sender" has been added for people who want
6191 the modified behaviour.
6197 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
6200 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
6201 indeed breaks things for older releases.
6203 3. Added additional logging to the case where there is a problem reading data
6204 from a filter that is running in a subprocess using a pipe, in order to
6205 try to track down a specific problem.
6207 4. Testing facility fudge: when running in the test harness and attempting
6208 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
6209 getting "No route to host". Convert this to a timeout.
6211 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
6214 6. Some OS don't have socklen_t but use size_t instead. This affects the
6215 fifth argument of getsockopt() amongst other things. This is now
6216 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
6217 can be set for individual OS. I have set it for SunOS5, OSF1, and
6218 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
6219 some earlier ones do not.
6221 7. Change 4.30/15 was not doing the test caselessly.
6223 8. The standard form for an IPv6 address literal was being rejected by address
6224 parsing in, for example, MAIL and RCPT commands. An example of this kind of
6225 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6226 this, as well as the form without the "IPv6" on the front (but only when
6227 address literals are enabled, of course).
6229 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
6231 10. Exim crashed if a message with an empty sender address specified by -f
6232 encountered a router with an errors_to setting. This could be provoked only
6233 by a command such as
6237 where an empty string was supplied; "<>" did not hit this bug.
6239 11. Installed PCRE release 4.5.
6241 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
6242 remained set. It is now erased.
6244 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
6245 times from message ids (which are base 36 rather than the normal 62).
6247 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
6248 were being counted as actual protocol errors, and logged if the log
6249 selector +smtp_protocol_error was set. One cannot be perfect in this test,
6250 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
6251 and DATA following a set of rejected RCPTs do not count as protocol errors.
6252 In other words, Exim assumes they were pipelined, though this may not
6253 actually be the case. Of course, in all cases the client gets an
6254 appropriate error code.
6256 15. If a lookup fails in an ACL condition, a message about the failure may
6257 be available; it is used if testing the ACL cannot continue, because most
6258 such messages specify what the cause of the deferral is. However, some
6259 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
6260 that caused an old message to be retained and used if a later statement
6261 caused a defer, replacing the real cause of the deferral.
6263 16. If an IP address had so many PTR records that the DNS lookup buffer
6264 was not large enough to hold them, Exim could crash while trying to process
6265 the truncated data. It now detects and logs this case.
6267 17. Further to 4.21/58, another change has been made: if (and only if) the
6268 first line of a message (the first header line) ends with CRLF, a bare LF
6269 in a subsequent header line has a space inserted after it, so as not to
6270 terminate the header.
6272 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
6273 unnecessarily, used atoi() instead of strtol(), and didn't check the
6274 termination when getting file sizes from file names by regex.
6276 19. Completely re-implemented the support for maildirsize files, in the light
6277 of a number of problems with the previous contributed implementation
6278 (4.30/29). In particular:
6280 . If the quota is zero, the maildirsize file is maintained, but no quota is
6283 . If the maildir directory does not exist, it is created before any attempt
6284 to write a maildirsize file.
6286 . The quota value in the file is just a cache; if the quota is changed in
6287 the transport, the new value overrides.
6289 . A regular expression is available for excluding directories from the
6292 20. The autoreply transport checks the characters in options that define the
6293 message's headers; it allows continued headers, but it was checking with
6294 isspace() after an embedded newline instead of explicitly looking for a
6297 21. If all the "regular" hosts to which an address was routed had passed their
6298 expiry times, and had not reached their retry times, the address was
6299 bounced, even if fallback hosts were defined. Now Exim should go on to try
6302 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
6303 equivalent code in the SMTP transport. Some hosts send humungous responses
6304 to HELO/EHLO, more than 1024 it seems.
6306 23. Refactoring: code in filter.c used (void *) for "any old type" but this
6307 gives compiler warnings in some environments. I've now done it "properly",
6310 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
6311 (because of problems with the built-in one) was declared to return uschar *
6312 instead of char *, causing compiler failure.
6314 25. Fixed a file descriptor leak when processing alias/forward files.
6316 26. Fixed a minor format string issue in dbfn.c.
6318 27. Typo in exim.c: ("dmbnz" for "dbmnz").
6320 28. If a filter file refered to $h_xxx or $message_headers, and the headers
6321 contained RFC 2047 "words", Exim's memory could, under certain conditions,
6324 29. When a sender address is verified, it is cached, to save repeating the test
6325 when there is more than one recipient in a message. However, when the
6326 verification involves a callout, it is possible for different callout
6327 options to be set for different recipients. It is too complicated to keep
6328 track of this in the cache, so now Exim always runs a verification when a
6329 callout is required, relying on the callout cache for the optimization.
6330 The overhead is duplication of the address routing, but this should not be
6333 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
6334 to be verified with callout=postmaster, and the main callout worked but the
6335 postmaster check failed, the verification correctly failed. However, if a
6336 subsequent RCPT command asked for sender verification *without* the
6337 postmaster check, incorrect caching caused this verification also to fail,
6340 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
6341 it was not caching the DNS options (qualify_single, search_parents) that
6342 were used when the lookup failed. A subsequent lookup with different
6343 options therefore always gave the same answer, though there were cases
6344 where it should not have. (Example: a "domains = !$mx_any" option on a
6345 dnslookup router: the "domains" option is always processed without any
6346 widening, but the router might have qualify_single set.) Now Exim uses the
6347 cached value only when the same options are set.
6349 32. Added John Jetmore's "exipick" utility to the distribution.
6351 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
6352 than a timeout (e.g. a certificate is required, and is not provided), an
6353 Exim server now closes the connection immediately. Previously it waited for
6354 the client to close - but if the client is SSL, it seems that they each
6355 wait for each other, leading to a delay before one of them times out.
6357 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
6358 maintained 0.8.x compatibility because I don't think many are using it, and
6359 it is clearly obsolete.
6361 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
6364 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
6365 client certificate was expired. A simple patch fixes this, though I don't
6366 understand the full logic of why the verify callback is called multiple
6369 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
6370 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
6371 which causes problems with some clients (such as the Certicom SSL Plus
6372 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
6373 disables the coutermeasure allowing Eudora to connect."
6375 38. Exim was not checking that a write() to a log file succeeded. This could
6376 lead to Bad Things if a log got too big, in particular if it hit a file
6377 size limit. Exim now panics and dies if it cannot write to a log file, just
6378 as it does if it cannot open a log file.
6380 39. Modified OS/Makefile-Linux so that it now contains
6382 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
6384 The two -D definitions ensure that Exim is compiled with large file
6385 support, which makes it possible to handle log files that are bigger than
6388 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
6389 instance) a domain was checked against a named list that involved a lookup,
6390 causing $domain_data to be set, then another domain was checked against the
6391 same list, then the first domain was re-checked, the value of $domain_data
6392 after the final check could be wrong. In particular, if the second check
6393 failed, it could be set empty. This bug probably also applied to
6396 41. The strip_trailing_dot option was not being applied to the address given
6397 with the -f command-line option.
6399 42. The code for reading a message's header from the spool was incrementing
6400 $received_count, but never initializing it. This meant that the value was
6401 incorrect (doubled) while delivering a message in the same process in which
6402 it was received. In the most common configuration of Exim, this never
6403 happens - a fresh exec is done - but it can happen when
6404 deliver_drop_privilege is set.
6406 43. When Exim logs an SMTP synchronization error - client data sent too soon -
6407 it now includes up to 150 characters of the unexpected data in the log
6410 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6411 and building data strings. The size of both of these buffers was 10 000
6412 bytes - far larger than anybody would *ever* want, thought I. Needless to
6413 say, somebody hit the limit. I have increased the maximum line length to
6414 20 000 and the maximum data length of concatenated lines to 100 000. I have
6415 also fixed two bugs, because there was no checking on these buffers. Tsk,
6416 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6417 buffer is too small.
6419 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6420 lsearch lookups. Now it does.
6422 46. When parsing a route_list item in a manualroute router, a fixed-length
6423 buffer was used for the list of hosts. I made this 1024 bytes long,
6424 thinking that nobody would ever have a list of hosts that long. Wrong.
6425 Somebody had a whole pile of complicated expansion conditions, and the
6426 string was silently truncated, leading to an expansion error. It turns out
6427 that it is easier to change to an unlimited length (owing to other changes
6428 that have happened since this code was originally written) than to build
6429 structure for giving a limitation error. The length of the item that
6430 expands into the list of hosts is now unlimited.
6432 47. The lsearch lookup could not handle data where the length of text line was
6433 more than 4095 characters. Such lines were truncated, leading to shortened
6434 data being returned. It should now handle lines of any length.
6436 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6437 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6440 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6441 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6442 no longer happens in comment lines. A list of replacements is now placed
6443 at the head of all of the source files, except those whose only change is
6444 to replace PERL_COMMAND in the very first #! line.
6446 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6447 messages on the queue, with a bottom-up merge sort, using code contributed
6448 by Michael Haardt. This should make operations like -bp somewhat faster on
6449 large queues. It won't affect queue runners, except when queue_run_in_order
6452 51. Installed eximstats 1.31 in the distribution.
6454 52. Added support for SRV lookups to the dnslookup router.
6456 53. If an ACL referred to $message_body or $message_body_end, the value was not
6457 reset for any messages that followed in the same SMTP session.
6459 54. The store-handling optimization for building very long strings was not
6460 differentiating between the different store pools. I don't think this
6461 actually made any difference in practice, but I've tidied it.
6463 55. While running the routers to verify a sender address, $sender_address
6464 was still set to the sender address. This is wrong, because when routing to
6465 send a bounce to the sender, it would be empty. Therefore, I have changed
6466 it so that, while verifying a sender address, $sender_address is set to <>.
6467 (There is no change to what happens when verifying a recipient address.)
6469 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6470 target A or AAAA records (if not already returned) without resetting the
6471 qualify_single or search_parents options of the DNS resolver. These are
6472 inappropriate in this case because the targets of MX and SRV records must
6473 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6474 target that, when qualified, matched something in the local domain. These
6475 two options are now turned off when doing these lookups.
6477 57. It seems that at least some releases of Reiserfs (which does not have the
6478 concept of a fixed number of inodes) returns zero and not -1 for the
6479 number of available inodes. This interacted badly with check_spool_inodes,
6480 which assumed that -1 was the "no such thing" setting. What I have done is
6481 to check that the total number of inodes is greater than zero before doing
6482 the test of how many are available.
6484 58. When a "warn" ACL statement has a log_message modifier, the message is
6485 remembered, and not repeated. This is to avoid a lot of repetition when a
6486 message has many recipients that cause the same warning to be written.
6487 However, Exim was preserving the list of already written lines for an
6488 entire SMTP session, which doesn't seem right. The memory is now reset if a
6489 new message is started.
6491 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6492 debugging output unless log_rewrite was also set.
6494 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6495 of (int)(handle) when we know that handle contains (void *)(-1).
6497 61. The Exim daemon panic-logs an error return when it closes the incoming
6498 connection. However "connection reset by peer" seems to be common, and
6499 isn't really an error worthy of noting specially, so that particular error
6502 62. When Exim is trying to find all the local interfaces, it used to panic and
6503 die if the ioctl to get the interface flags failed. However, it seems that
6504 on at least one OS (Solaris 9) it is possible to have an interface that is
6505 included in the list of interfaces, but for which you get a failure error
6506 for this call. This happens when the interface is not "plumbed" into a
6507 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6508 failure of the "get flags" call assumes that the interface is down.
6510 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6511 makes life easier for people who are doing arithmetic on fields extracted
6512 from dates, where you often get leading zeros that should not be
6513 interpreted as octal.
6515 64. Added qualify_domain to the redirect router, to override the global
6518 65. If a pathologically long header line contained very many addresses (the
6519 report of this problem mentioned 10 000) and each of them was rewritten,
6520 Exim could use up a very large amount of memory. (It kept on making new
6521 copies of the header line as it rewrote, and never released the old ones.)
6522 At the expense of a bit more processing, the header rewriting function has
6523 been changed so that it no longer eats memory in this way.
6525 66. The generation of the Received: header has been moved from the time that a
6526 message starts to be received, to the time that it finishes. The timestamp
6527 in the Received: header should now be very close to that of the <= log
6528 line. There are two side-effects of this change:
6530 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6531 logged header lines no longer include the local Received: line, because
6532 it has not yet been created. The same applies to a copy of the message
6533 that is returned to a non-SMTP sender when a message is rejected.
6535 (b) When a filter file is tested using -bf, no additional Received: header
6536 is added to the test message. After some thought, I decided that this
6539 This change does not affect the value of $received_for. It is still set
6540 after address rewriting, but before local_scan() is called.
6542 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6544 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6545 gave an unhelpful panic error message, and a defer error. I have managed to
6546 change this behaviour so that it now rejects any supplied certificate,
6547 which seems right, as the list of acceptable certificates is empty.
6549 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6550 gave an unhelpful defer error. I have not managed to make this reject any
6551 supplied certificates, but the error message it gives is "no certificate
6552 supplied", which is not helpful.
6554 70. exigrep's output now also includes lines that are not associated with any
6555 message, but which match the given pattern. Implemented by a patch from
6556 Martin Sluka, which also tidied up the Perl a bit.
6558 71. Recipient callout verification, like sender verification, was using <> in
6559 the MAIL FROM command. This isn't really the right thing, since the actual
6560 sender may affect whether the remote host accepts the recipient or not. I
6561 have changed it to use the actual sender in the callout; this means that
6562 the cache record is now keyed on a recipient/sender pair, not just the
6563 recipient address. There doesn't seem to be a real danger of callout loops,
6564 since a callout by the remote host to check the sender would use <>.
6565 [SEE ABOVE: changed after hitting problems.]
6567 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6568 temporary errors. However, in the case of such a code being given after
6569 the end of a data transmission (i.e. after ".") Exim was failing to write
6570 a retry record for the message. (Yes, there was some broken host that was
6571 actually sending 8xx at this point.)
6573 73. An unknown lookup type in a host list could cause Exim to panic-die when
6574 the list was checked. (An example that provoked this was putting <; in the
6575 middle of a list instead of at the start.) If this happened during a DATA
6576 ACL check, a -D file could be left lying around. This kind of configuration
6577 error no longer causes Exim to die; instead it causes a defer error. The
6578 incident is still logged to the main and panic logs.
6580 74. Buglet left over from Exim 3 conversion. The message "too many messages
6581 in one connection" was written to the rejectlog but not the mainlog, except
6582 when address rewriting (yes!) was being logged.
6584 75. Added write_rejectlog option.
6586 76. When a system filter was run not as root (that is, when system_filter_user
6587 was set), the values of the $n variables were not being returned to the
6588 main process; thus, they were not subsequently available in the $sn
6591 77. Added +return_path_on_delivery log selector.
6593 78. A connection timeout was being treated differently from recipients deferred
6594 when testing hosts_max_try with a message that was older than the host's
6595 retry timeout. (The host should not be counted, thus allowing all hosts to
6596 be tried at least once before bouncing.) This may have been the cause of an
6597 occasionally reported bug whereby a message would remain on the queue
6598 longer than the retry timeout, but would be bounced if a delivery was
6599 forced. I say "may" because I never totally pinned down the problem;
6600 setting up timeout/retry tests is difficult. See also the next item.
6602 79. The ultimate address timeout was not being applied to errors that involved
6603 a combination of host plus message (for example, a timeout on a MAIL
6604 command). When an address resolved to a number of possible hosts, and they
6605 were not all tried for each delivery (e.g. because of hosts_max_try), a
6606 message could remain on the queue longer than the retry timeout.
6608 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6611 81. Fixed an obscure SMTP outgoing bug which required at least the following
6612 conditions: (a) there was another message waiting for the same server;
6613 (b) the server returned 5xx to all RCPT commands in the first message so
6614 that the message was not completed; (c) the server dropped the connection
6615 or gave a negative response to the RSET that Exim sends to abort the
6616 transaction. The observed case was a dropped connection after DATA that had
6617 been sent in pipelining mode. That is, the server had advertised PIPELINING
6618 but was not implementing it correctly. The effect of the bug was incorrect
6619 behaviour, such as trying another host, and this could lead to a crash.
6625 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6626 and daemon.c were passed as pointers to ints; they should have been
6627 pointers to socklen_t variables (which are typically unsigned ints).
6629 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6632 3. Fixed a really odd bug that affected only the testing scheme; patching a
6633 certain fixed string in the binary changed the value of another string that
6634 happened to be identical to the end of the original first string.
6636 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6637 name", it returns that address as the IP address. On some operating
6638 systems (e.g. Solaris), it also passes back the IP address string as the
6639 "host name". However, on others (e.g. Linux), it passes back an empty
6640 string. Exim wasn't checking for this, and was changing the host name to an
6641 empty string, assuming it had been canonicalized.
6643 5. Although rare, it is permitted to have more than one PTR record for a given
6644 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6645 all the names associated with an address, because they do in Solaris.
6646 However, it seems that they do not in Linux for data that comes from the
6647 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6648 I found this out when I moved to a new Linux workstation and tried to run
6649 the Exim test suite.
6651 To get round this problem I have changed the code so that it now does its
6652 own call to the DNS to look up PTR records when searching for a host name.
6653 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6654 addresses that are only in /etc/hosts are still found.
6656 This behaviour is, however, controlled by an option called host_lookup_
6657 order, which defaults to "bydns:byaddr". If people want to use the other
6658 order, or indeed, just use one or the other means of lookup, they can
6659 specify it in this variable.
6661 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6662 some operating systems, this comes back from gethostbyaddr() as an empty
6663 string, and this is what Exim used to test for. However, it seems that in
6664 other systems, "." is yielded. Exim now tests for this case too.
6666 7. The values of check_spool_space and check_log_space are now held internally
6667 as a number of kilobytes instead of an absolute number of bytes. If a
6668 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6669 kilobyte. This means that much larger values can be stored.
6671 8. Exim monitor: an attempt to get the action menu when not actually pointing
6672 at a message produces an empty menu entitled "No message selected". This
6673 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6674 no entries in it ("Shell widget menu has zero width and/or height"). So I
6675 have added a single, blank menu entry in this case.
6677 9. Added ${quote_local_part.
6679 10. MIME decoding is now applied to the contents of Subject: header lines when
6682 11. Now that a reference to $sender_host_address automatically causes a reverse
6683 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6684 host lookup before query-style lookups in lists that might use this
6685 variable. This has therefore been abolished, and the "net-" prefix is no
6686 longer necessary for query-style lookups.
6688 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6689 have been a typo for LFLAGS, so it has been changed.
6691 13. The install script calls Exim with "-C /dev/null" in order to find the
6692 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6693 to be output. However, since Exim outputs its version number before the
6694 error, it didn't break the script. It just looked ugly. I fixed this by
6695 always allowing "-C /dev/null" if the caller is root.
6697 14. Ignore overlarge ACL variable number when reading spool file - insurance
6698 against a later release with more variables having written the file.
6700 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6701 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6702 this, as well as the form without the "IPv6" on the front.
6704 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6705 OS/Makefile-Darwin file.
6707 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6708 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6710 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6711 variable. However, some OS use ints for inodes. I've added cast to long int
6712 to get rid of the compiler warning.
6714 19. I had forgotten to lock out "/../" in configuration file names when
6715 ALT_CONFIG_PREFIX was set.
6717 20. Routers used for verification do not need to specify transports. However,
6718 if such a router generated a host list, and callout was configured, Exim
6719 crashed, because it could not find a port number from the (non-existent)
6720 transport. It now assumes port 25 in this circumstance.
6722 21. Added the -t option to exigrep.
6724 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6725 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6726 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6727 them is set, LOOKUP_LSEARCH is forced.
6729 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6731 24. Added sender and host information to the "rejected by local_scan()" log
6732 line; previously there was no indication of these.
6734 25. Added .include_if_exists.
6736 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6737 Linux. It turns out that not all file systems support this. Apparently some
6738 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6739 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6740 not supported on the file descriptor, is now ignored when Exim is trying to
6741 sync a directory. This applies only to Linux.
6743 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6745 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6746 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6749 29. Added support for maildirsize files from supplied patch (modified a bit).
6751 30. The use of :fail: followed by an empty string could lead Exim to respond to
6752 sender verification failures with (e.g.):
6754 550 Verification failed for <xxx>
6755 550 Sender verify failed
6757 where the first response line was missing the '-' that indicates it is not
6758 the final line of the response.
6760 31. The loop for finding the name of the user that called Exim had a hardwired
6761 limit of 10; it now uses the value of finduser_retries, which is used for
6762 all other user lookups.
6764 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6767 33. Exim was neglecting to zero errno before one call of strtol() when
6768 expanding a string and expecting an integer value. On some systems this
6769 resulted in spurious "integer overflow" errors. Also, it was casting the
6770 result into an int without checking.
6772 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6773 did not work. The code looks as if it has *never* worked, though it appears
6774 to have been documented since at least release 1.62. I have made it work.
6776 35. The "timeout_DNS" error in retry rules, also documented since at least
6777 1.62, also never worked. As it isn't clear exactly what this means, and
6778 clearly it isn't a major issue, I have abolished the feature by treating it
6779 as "timeout", and writing a warning to the main and panic logs.
6781 36. The display of retry rules for -brt wasn't always showing the error code
6784 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6785 timeout_connect_A, timeout_connect_MX.
6787 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6788 to the empty sender.
6790 39. The daemon was not analysing the content of -oX till after it had closed
6791 stderr and disconnected from the controlling terminal. This meant that any
6792 syntax errors were only noted on the panic log, and the return code from
6793 the command was 0. By re-arranging the code a little, I've made the
6794 decoding happen first, so such errors now appear on stderr, and the return
6795 code is 1. However, the actual setting up of the sockets still happens in
6796 the disconnected process, so errors there are still only recorded on the
6799 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6800 connections (as happens on some IP stacks) was logged at start up time as
6801 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6802 it from "IPv6 and IPv4", which means that two separate sockets are being
6805 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6806 says whether AF_INET or AF_INET6 was passed as an argument.
6808 42. Exiwhat output was messed up when time zones were included in log
6811 43. Exiwhat now gives more information about the daemon's listening ports,
6812 and whether -tls-on-connect was used.
6814 44. The "port" option of the smtp transport is now expanded.
6816 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6817 silently ignored. Now an error message is written to the main and panic
6820 46. There's a new ACL modifier called "logwrite" which writes to a log file
6821 as soon as it is encountered.
6823 47. Added $local_user_uid and $local_user_gid at routing time.
6825 48. Exim crashed when trying to verify a sender address that was being
6828 49. Exim was recognizing only a space character after ".include". It now also
6829 recognizes a tab character.
6831 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6832 extracting the relevant information from the specification. The man page no
6833 longer contains scrambled data for the -d option, and I've added a section
6834 at the front about calling Exim under different names.
6836 51. Added "extra_headers" argument to the "mail" command in filter files.
6838 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6841 53. Installed eximstats 1.29.
6843 54. Added transport_filter_timeout as a generic transport option.
6845 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6846 Cc: header lines. This was required by RFC 822, but it not required by RFC
6849 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6850 incoming messages that did not have them. Now it does so only if the
6851 message originates locally, that is, if there is no associated remote host
6852 address. When Resent- header lines are present, this applies to the Resent-
6853 lines rather than the non-Resent- lines.
6855 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6856 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6858 58. Messages for configuration errors now include the name of the main
6859 configuration file - useful now that there may be more than one file in a
6860 list (.included file names were always shown).
6862 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6863 for those rare installations that do not start the daemon as root or run it
6864 setuid root. I've cut out the call to initgroups() if the daemon is not
6867 60. The Exim user and group can now be bound into the binary as text strings
6868 that are looked up at the start of Exim's processing.
6870 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6872 62. Added $mailstore_basename variable.
6874 63. Installed patch to sieve.c from Michael Haardt.
6876 64. When Exim failed to open the panic log after failing to open the main log,
6877 the original message it was trying to log was written to stderr and debug
6878 output, but if they were not available (the usual case in production), it
6879 was lost. Now it is written to syslog before the two lines that record the
6880 failures to open the logs.
6882 65. Users' Exim filters run in subprocesses under the user's uid. It is
6883 possible for a "deliver" command or an alias in a "personal" command to
6884 provoke an address rewrite. If logging of address rewriting is configured,
6885 this fails because the process is not running as root or exim. There may be
6886 a better way of dealing with this, but for the moment (because 4.30 needs
6887 to be released), I have disabled address rewrite logging when running a
6888 filter in a non-root, non-exim process.
6894 1. The buildconfig auxiliary program wasn't quoting the value set for
6895 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6896 not defined. This bug was masked in 4.22 by the effect that was fixed in
6899 2. Some messages that were rejected after a message id was allocated were
6900 shown as "incomplete" by exigrep. It no longer does this for messages that
6901 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6903 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6904 have allow_domain_literals set, the ID did not get logged in the <= line.
6905 Domain literals are now always recognized in Message-ID: header lines.
6907 4. The first argument for a ${extract expansion item is the key name or field
6908 number. Leading and trailing spaces in this item were not being ignored,
6909 causing some misleading effects.
6911 5. When deliver_drop_privilege was set, single queue runner processes started
6912 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6913 same command in the process it spins off) were not dropping privilege.
6915 6. When the daemon running as "exim" started a queue runner, it always
6916 re-executed Exim in the spun-off process. This is a waste of effort when
6917 deliver_drop_privilege is set. The new process now just calls the
6918 queue-runner function directly.
6924 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6927 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6928 ignored. Though the use of -oP was forcing the writing of a pid file, it
6929 was always written to the default place.
6931 3. If the message "no IP address found for host xxxx" is generated during
6932 incoming verification, it is now followed by identification of the incoming
6933 connection (so you can more easily find what provoked it).
6935 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6937 5. Added some features to "harden" Exim a bit more against certain attacks:
6939 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6940 be put in Local/Makefile. This is like the never_users runtime option,
6941 but it cannot be overridden. The default setting is "root".
6943 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6944 prefix string with which any file named in a -C command line option
6947 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6948 is retained for -C and -D only if the caller of Exim is root. Without
6949 it, the exim user may also use -C and -D and retain privilege.
6951 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6952 command line option is disabled.
6954 6. Macro names set by the -D option must start with an upper case letter, just
6955 like macro names defined in the configuration file.
6957 7. Added "dereference=" facility to LDAP.
6959 8. Two instances of the typo "uknown" in the source files are fixed.
6961 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6962 the Configure-Makefile script screwed up while processing it.
6964 10. Incorporated PCRE 4.4.
6966 11. The SMTP synchronization check was not operating right at the start of an
6967 SMTP session. For example, it could not catch a HELO sent before the client
6968 waited for the greeting. There is now a check for outstanding input at the
6969 point when the greeting is written. Because of the duplex, asynchronous
6970 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6971 way, but not yet received, when the check is performed.
6973 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6974 on TCP/IP sockets, because this apparently causes some broken clients to
6977 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6978 unchanged) from the Cygwin maintainer.
6980 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6981 was supported instead of testing for mbx. Effectively a typo.
6983 15. The spa authenticator server code was not checking that the input it
6984 received was valid base64.
6986 16. The debug output line for the "set" modifier in ACLs was not showing the
6987 name of the variable that was being set.
6989 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6991 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6992 lookup on its own. Something else was needed to trigger the lookup. For
6993 example, a match in host_lookup or the need for a host name in a host list.
6994 Now, if $sender_host_name is referenced and the host name has not yet been
6995 looked up, a lookup is performed. If the lookup fails, the variable remains
6996 empty, and $host_lookup_failed is set to "1".
6998 19. Added "eqi" as a case-independent comparison operator.
7000 20. The saslauthd authentication condition could segfault if neither service
7001 nor realm was specified.
7003 21. If an overflowing value such as "2048M" was set for message_size_limit, the
7004 error message that was logged was misleading, and incoming SMTP
7005 connections were dropped. The message is now more accurate, and temporary
7006 errors are given to SMTP connections.
7008 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
7009 (except RSET) with a 421 error, until QUIT is received. However, it was
7010 failing to send a response to QUIT.
7012 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
7013 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
7014 ACL is now run after the helo_try_verify_hosts code.
7016 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
7017 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
7018 case-independent, so other case variants are also recognized). Apparently
7019 some systems use these upper case variants.
7021 25. If more than two messages were waiting for the same host, and a transport
7022 filter was specified for the transport, Exim sent two messages over the
7023 same TCP/IP connection, and then failed with "socket operation on non-
7024 socket" when it tried to send the third.
7026 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
7028 27. The extern definition of crypt16() in expand.c was not being excluded when
7029 the OS had its own crypt16() function.
7031 28. Added bounce_return_body as a new option, and bounce_return_size_limit
7032 as a preferred synonym for return_size_limit, both as an option and as an
7035 29. Added LIBS=-liconv to OS/Makefile-OSF1.
7037 30. Changed the default configuration ACL to relax the local part checking rule
7038 for addresses that are not in any local domains. For these addresses,
7039 slashes and pipe symbols are allowed within local parts, but the sequence
7040 /../ is explicitly forbidden.
7042 31. SPA server authentication was not clearing the challenge buffer before
7045 32. log_message in a "warn" ACL statement was writing to the reject log as
7046 well as to the main log, which contradicts the documentation and doesn't
7047 seem right (because no rejection is happening). So I have stopped it.
7049 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
7050 However, I am unable to do any testing of this.
7052 34. Fixed an infelicity in the appendfile transport. When checking directories
7053 for a mailbox, to see if any needed to be created, it was accidentally
7054 using path names with one or more superfluous leading slashes; tracing
7055 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
7057 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
7058 contents are added to the log line that is written for every discarded
7059 recipient. (Previously a log_message setting was ignored.)
7061 36. The ${quote: operator now quotes the string if it is empty.
7063 37. The install script runs exim in order to find its version number. If for
7064 some reason other than non-existence or emptiness, which it checks, it
7065 could not run './exim', it was installing it with an empty version number,
7066 i.e. as "exim-". This error state is now caught, and the installation is
7069 38. An argument was missing from the function that creates an error message
7070 when Exim fails to connect to the socket for saslauthd authentication.
7071 This could cause Exim to crash, or give a corrupted message.
7073 39. Added isip, isip4, and isip6 to ${if conditions.
7075 40. The ACL variables $acl_xx are now saved with the message, and can be
7076 accessed later in routers, transports, and filters.
7078 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
7079 strings in the file are not string-expanded.
7081 42. If a MAIL command specified a SIZE value that was too large to fit into an
7082 int variable, the check against message_size_limit failed. Such values are
7083 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
7084 day this will have to be increased, but I don't think I want to be around
7085 when emails are that large.
7092 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
7093 iconv() is not standard in FreeBSD.
7095 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
7096 IPv6 enabled. The observed symptom was a segmentation fault on return from
7097 the function os_common_find_running_interfaces() in src/os.c.
7099 3. In the check_special_case() function in daemon.c I had used "errno" as an
7100 argument name, which causes warnings on some systems. This was basically a
7101 typo, since it was named "eno" in the comments!
7103 4. The code that waits for the clock to tick (at a resolution of some fraction
7104 of a second) so as to ensure message-id uniqueness was always waiting for
7105 at least one whole tick, when it could have waited for less. [This is
7106 almost certainly not relevant at current processor speeds, where it is
7107 unlikely to ever wait at all. But we try to future-proof.]
7109 5. The function that sleeps for a time interval that includes fractions of a
7110 second contained a race. It did not block SIGALRM between setting the
7111 timer, and suspending (a couple of lines later). If the interval was short
7112 and the sigsuspend() was delayed until after it had expired, the suspension
7113 never ended. On busy systems this could lead to processes getting stuck for
7116 6. Some uncommon configurations may cause a lookup to happen in a queue runner
7117 process, before it forks any delivery processes. The open lookup caching
7118 mechanism meant that the open file or database connection was passed into
7119 the delivery process. The problem was that delivery processes always tidy
7120 up cached lookup data. This could cause a problem for the next delivery
7121 process started by the queue runner, because the external queue runner
7122 process does not know about the closure. So the next delivery process
7123 still has data in the lookup cache. In the case of a file lookup, there was
7124 no problem because closing a file descriptor in a subprocess doesn't affect
7125 the parent. However, if the lookup was caching a connection to a database,
7126 the connection was closed, and the second delivery process was likely to
7127 see errors such as "PGSQL: query failed: server closed the connection
7128 unexpectedly". The problem has been fixed by closing all cached lookups
7129 in a queue runner before running a delivery process.
7131 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
7132 seem to have the "const" qualifier which it has on other OS. I've
7135 8. Change 4.21/2 was too strict. It is only if there are two authenticators
7136 *of the same type* (client or server) with the same public name that an
7137 error should be diagnosed.
7139 9. When Exim looked up a host name for an IP address, but failed to find the
7140 original IP address when looking up the host name (a safety check), it
7141 output the message "<ip address> does not match any IP for NULL", which was
7142 confusing, to say the least. The bug was that the host name should have
7143 appeared instead of "NULL".
7145 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
7146 user that is built into the binary, and the -C or -D options is used, root
7147 privilege is dropped before the configuration file is read. In addition,
7148 logging is switched to stderr instead of the normal log files. If the
7149 configuration then re-defines the Exim user, the unprivileged environment
7150 is probably not what is expected, so Exim logs a panic warning message (but
7153 However, if deliver_drop_privilege is set, the unprivileged state may well
7154 be exactly what is intended, so the warning has been cut out in that case,
7155 and Exim is allowed to try to write to its normal log files.
7161 1. smtp_return_error_details was not giving details for temporary sender
7162 or receiver verification errors.
7164 2. Diagnose a configuration error if two authenticators have the same public
7167 3. Exim used not to create the message log file for a message until the first
7168 delivery attempt. This could be confusing when incoming messages were held
7169 for policy or load reasons. The message log file is now created at the time
7170 the message is received, and an initial "Received" line is written to it.
7172 4. The automatically generated man page for command line options had a minor
7173 bug that caused no ill effects; however, a more serious problem was that
7174 the procedure for building the man page automatically didn't always
7175 operate. Consequently, release 4.20 contains an out-of-date version. This
7176 shouldn't happen again.
7178 5. When building Exim with embedded Perl support, the script that builds the
7179 Makefile was calling 'perl' to find its compile-time parameters, ignoring
7180 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
7182 6. The freeze_tell option was not being used for messages that were frozen on
7183 arrival, either by an ACL or by local_scan().
7185 7. Added the smtp_incomplete_transaction log selector.
7187 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
7188 was accepting AUTH without a new EHLO.
7190 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
7191 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
7194 10. Logging of TCP/IP connections (when configured) now happens in the main
7195 daemon process instead of the child process, so that the TCP/IP connection
7196 count is more accurate (but it can never be perfect).
7198 11. The use of "drop" in a nested ACL was not being handled correctly in the
7199 outer ACL. Now, if condition failure induced by the nested "drop" causes
7200 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
7201 or "require"), the connection is dropped.
7203 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
7204 that yield "discard" can now be used with an "accept" or a "discard" verb,
7205 but an error is generated for any others (because I can't see a useful way
7206 to define what should happen).
7208 13. When an ACL is read dynamically from a file (or anywhere else), the lines
7209 are now processed in the same way as lines in the Exim configuration file.
7210 In particular, continuation lines are supported.
7212 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
7214 15. Added -ti meaning -t -i.
7216 16. Check for letters, digits, hyphens, and dots in the names of dnslist
7217 domains, and warn by logging if others are found.
7219 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
7220 returned from GIFCONF when Exim is trying to find the list of interfaces on
7221 a host. The code in os.c has been modified to copy each ifreq to an aligned
7222 structure in all cases.
7224 Also, in some cases, the returned ifreq's were being copied to a 'struct
7225 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
7226 means the last couple of bytes of an IPv6 address could be chopped if the
7227 ifreq contained only a normal sockaddr (14 bytes storage).
7229 18. Named domain lists were not supported in the hosts_treat_as_local option.
7230 An entry such as +xxxx was not recognized, and was treated as a literal
7233 19. Ensure that header lines added by a DATA ACL are included in the reject log
7234 if the ACL subsequently rejects the message.
7236 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
7237 MD5 (which is deprecated).
7239 21. When testing a filter file using -bf, Exim was writing a message when it
7240 took the sender from a "From " line in the message, but it was not doing so
7241 when it took $return_path from a Return-Path: header line. It now does.
7243 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
7244 with a valid header line field name (a series of printing characters
7245 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
7247 23. Changed "disc" in the source to "disk" to conform to the documentation and
7248 the book and for uniformity.
7250 24. Ignore Sendmail's -Ooption=value command line item.
7252 25. When execve() failed while trying to run a command in a pipe transport,
7253 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
7254 could be confused with a return value of 69 from the command itself. This
7255 has been changed to 127, the value the shell returns if it is asked to run
7256 a non-existent command. The wording for the related log line suggests a
7257 non-existent command as the problem.
7259 26. If received_header_text expands to an empty string, do not add a Received:
7260 header line to the message. (Well, it adds a token one on the spool, but
7261 marks it "old" so that it doesn't get used or transmitted.)
7263 27. Installed eximstats 1.28 (addition of -nt option).
7265 28. There was no check for failure on the call to getsockname() in the daemon
7266 code. This can fail if there is a shortage of resources on the system, with
7267 ENOMEM, for example. A temporary error is now given on failure.
7269 29. Contrary to the C standard, it seems that in some environments, the
7270 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
7271 program. Exim now does this explicitly; it affects the formatting of
7272 timestamps using strftime().
7274 30. If exiqsumm was given junk data, it threw up some uninitialized variable
7275 complaints. I've now initialized all the variables, to avoid this.
7277 32. Header lines added by a system filter were not being "seen" during
7278 transport-time rewrites.
7280 33. The info_callback() function passed to OpenSSL is set up with type void
7281 (*)(SSL *, int, int), as described somewhere. However, when calling the
7282 function (actually a macro) that sets it up, the type void(*)() is
7283 expected. I've put in a cast to prevent warnings from picky compilers.
7285 34. If a DNS black list lookup found a CNAME record, but there were no A
7286 records associated with the domain it pointed at, Exim crashed.
7288 35. If a DNS black list lookup returned more than one A record, Exim ignored
7289 all but the first. It now scans all returned addresses if a particular IP
7290 value is being sought. In this situation, the contents of the
7291 $dnslist_value variable are a list of all the addresses, separated by a
7294 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
7295 to accept a host name and all its aliases if the forward lookup for any of
7296 them yielded the IP address of the incoming connection. Now it accepts only
7297 those names whose forward lookup yields the correct IP address. Any other
7298 names are discarded. This closes a loophole whereby a rogue DNS
7299 administrator could create reverse DNS records to break through a
7300 wildcarded host restriction in an ACL.
7302 37. If a user filter or a system filter that ran in a subprocess used any of
7303 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
7304 the wrong values were passed to the pipe command ($thisaddress had the
7305 value of $0, $0 had the value of $1, etc). This bug was introduced by
7306 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
7308 38. Improved the line breaking for long SMTP error messages from ACLs.
7309 Previously, if there was no break point between 40 and 75 characters, Exim
7310 left the rest of the message alone. Two changes have been made: (a) I've
7311 reduced the minimum length to 35 characters; (b) if it can't find a break
7312 point between 35 and 75 characters, it looks ahead and uses the first one
7313 that it finds. This may give the occasional overlong line, but at least the
7314 remaining text gets split now.
7316 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
7317 file descriptors might be low, and that setting 1000 would always raise it.
7318 It turns out that in some environments, the limit is already over 1000 and
7319 that lowering it causes trouble. So now Exim takes care not to decrease it.
7321 40. When delivering a message, the value of $return_path is set to $sender_
7322 address at the start of routing (routers may change the value). By an
7323 oversight, this default was not being set up when an address was tested by
7324 -bt or -bv, which affected the outcome if any router or filter referred to
7327 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
7328 log only when "message" or "log_message" are set. However, if one of the
7329 conditions was an address verification, or a call to a nested ACL, the
7330 messages generated by the underlying test were being passed through. This
7331 no longer happens. The underlying message is available in $acl_verify_
7332 message for both "message" and "log_message" expansions, so it can be
7333 passed through if needed.
7335 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
7336 new expansion $bh_ to give the encoded byte string without charset
7337 translation. Translation happens only if iconv() is available; HAVE_ICONV
7338 indicates this at build time. HEADERS_CHARSET gives the charset to
7339 translate to; headers_charset can change it in the configuration, and
7340 "headers charset" can change it in an individual filter file.
7342 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
7343 that creates RFC 2047 encoded "words" labels them as that charset instead
7344 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
7345 expansion operator; (ii) when Exim creates a From: line for a local
7346 message; (iii) when a header line is rewritten to include a "phrase" part.
7348 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
7349 buggy, causing it to skip the first lines of messages whose message ID
7350 ended in 'D'. This would not have bitten before Exim release 4.14, because
7351 message IDs were unlikely to end in 'D' before then. The effect was to have
7352 incorrect size information for certain domains.
7354 45. #include "config.h" was missing at the start of the crypt16.c module. This
7355 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
7358 46. If there was a timeout during a "random" callout check, Exim treated it as
7359 a failure of the random address, and carried on sending RSET and the real
7360 address. If the delay was just some slowness somewhere, the response to the
7361 original RCPT would be taken as a response to RSET and so on, causing
7362 mayhem of various kinds.
7364 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
7365 when I implemented it. It didn't allow for the fact that some option values
7366 may legitimately be negative (e.g. size_addition), and it didn't even do
7367 the right test for positive values.
7369 48. Domain names in DNS records are case-independent. Exim always looks them up
7370 in lower case. Some resolvers return domain names in exactly the case they
7371 appear in the zone file, that is, they may contain uppercase letters. Not
7372 all resolvers do this - some return always lower case. Exim was treating a
7373 change of case by a resolver as a change of domain, similar to a widening
7374 of a domain abbreviation. This triggered its re-routing code and so it was
7375 trying to route what was effectively the same domain again. This normally
7376 caused routing to fail (because the router wouldn't handle the domain
7377 twice). Now Exim checks for this case specially, and just changes the
7378 casing of the domain that it ultimately uses when it transmits the message
7381 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
7384 50. If a filter generated a file delivery with a non-absolute name (possible if
7385 no home directory exists for the router), the forbid_file option was not
7388 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
7389 the existing equality matching.
7391 52. Exim was using ints instead of ino_t variables in some places where it was
7392 dealing with inode numbers.
7394 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
7395 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
7396 called TMPDIR, and if it finds it is different, it changes its value.
7398 54. The smtp_printf() function is now made available to local_scan() so
7399 additional output lines can be written before returning. There is also an
7400 smtp_fflush() function to enable the detection of a dropped connection.
7401 The variables smtp_input and smtp_batched_input are exported to
7404 55. Changed the default runtime configuration: the message "Unknown user"
7405 has been removed from the ACL, and instead placed on the localuser router,
7406 using the cannot_route_message feature. This means that any verification
7407 failures that generate their own messages won't get overridden. Similarly,
7408 the "Unrouteable address" message that was in the ACL for unverifiable
7409 relay addresses has also been removed.
7411 56. Added hosts_avoid_esmtp to the smtp transport.
7413 57. The exicyclog script was not checking for the esoteric option
7414 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7415 will work only if exicyclog is run under the appropriate euid.
7417 58. Following a discussion on the list, the rules by which Exim recognises line
7418 endings on incoming messages have been changed. The -dropcr and drop_cr
7419 options are now no-ops, retained only for backwards compatibility. The
7420 following line terminators are recognized: LF CRLF CR. However, special
7421 processing applies to CR:
7423 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7424 nor a local message in the state where . is a terminator.
7426 (ii) If a bare CR is encountered in a header line, an extra space is added
7427 after the line terminator so as not to end the header. The reasoning
7428 behind this is that bare CRs in header lines are most likely either
7429 to be mistakes, or people trying to play silly games.
7431 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7432 was being incorrectly given as 18 bytes larger than it should have been.
7433 This is a VOB (very old bug).
7435 60. This may never have affected anything current, but just in case it has:
7436 When the local host is found other than at the start of a list of hosts,
7437 the local host, those with the same MX, and any that follow, are discarded.
7438 When the list in question was part of a longer list of hosts, the following
7439 hosts (not currently being processed) were also being discarded. This no
7440 longer happens. I'm not sure if this situation could ever has previously
7443 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7446 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7447 header. This is something that is recommended in a new Internet Draft, and
7448 is something that is documented as being done by Sendmail. There are two
7449 possible values. For messages generated by the autoreply transport, Exim
7452 Auto-Submitted: auto-replied
7454 whereas for all other generated messages (e.g. bounces) it adds
7456 Auto-Submitted: auto-generated
7458 63. The "personal" condition in filters now includes a test for the
7459 Auto-Submitted: header. If it contains the string "auto-" the message it
7460 not considered personal.
7462 64. Added rcpt_include_affixes as a generic transport option.
7464 65. Added queue_only_override (default true).
7466 66. Added the syslog_duplication option.
7468 67. If what should have been the first header line of a message consisted of
7469 a space followed by a colon, Exim was mis-interpreting it as a header line.
7470 It isn't of course - it is syntactically invalid and should therefore be
7471 treated as the start of the message body. The misbehaviour could have
7472 caused a number of strange effects, including loss of data in subsequent
7473 header lines, and spool format errors.
7475 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7476 client host had authenticated. This control can now be exercised by an ACL
7477 for more flexibility.
7479 69. By default, callouts do not happen when testing with -bh. There is now a
7480 variant, -bhc, which does actually run the callout code, including
7481 consulting and updating the callout cache.
7483 70. Added support for saslauthd authentication, courtesy of Alexander
7486 71. If statvfs() failed on the spool or log directories while checking their
7487 size for availability, Exim confusingly gave the error "space shortage".
7488 Furthermore, in debugging mode it crashed with a floating point exception.
7489 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7490 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7491 serious problem, Exim now writes to the main and panic logs when this
7492 happens, with details of the failure. It then refuses to accept the
7493 incoming message, giving the message "spool directory problem" or "log
7494 directory problem" with a 421 code for SMTP messages.
7496 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7497 0, 1, and 2 exist, because some OS complain for execs without them (see
7498 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7499 descriptors. However, the code omitted to check that the open succeeded,
7500 causing mysterious errors if for some reason the permissions on /dev/null
7501 got screwed. Now Exim writes a message to the main and panic logs, and
7502 bombs out if it can't open /dev/null.
7504 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7505 interact so that it is all more flexible. It is supposed to remain
7506 backwards compatible. Also added extra_local_interfaces.
7508 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7509 to bomb out with an assertion failure - to the client this appears as a
7510 connection drop. This problem occurs in the part of the code that was taken
7511 from the Samba project. Fortunately, the assertion is in a very simple
7512 function, so I have fixed this by reproducing the function inline in the
7513 one place where it is called, and arranging for authentication to fail
7514 instead of killing the process with assert().
7516 75. The SPA client code was not working when the server requested OEM rather
7517 than Unicode encoding.
7519 76. Added code to make require_files with a specific uid setting more usable in
7520 the case where statting the file as root fails - usually a non-root-mounted
7521 NFS file system. When this happens and the failure is EACCES, Exim now
7522 forks a subprocess and does the per-uid checking as the relevant uid.
7524 77. Added process_log_path.
7526 78. If log_file_path was not explicitly set, a setting of check_log_space or
7527 check_log_inodes was ignored.
7529 79. If a space check for the spool or log partitions fails, the incident is now
7530 logged. Of course, in the latter case the data may get lost...
7532 80. Added the %p formatting code to string_format() so that it can be used to
7533 print addresses in debug_print(). Adjusted all the address printing in the
7534 debugging in store.c to use %p rather than %d.
7536 81. There was a concern that a line of code in smtp_in.c could overflow a
7537 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7538 initially expressed, the concern was not well-founded, because trailing
7539 spaces are removed early. However, if the trailing spaces were followed by
7540 a NULL, they did not get removed, so the overflow was possible. Two fixes
7543 (a) I re-wrote the offending code in a cleaner fashion.
7544 (b) If an incoming SMTP command contains a NULL character, it is rejected
7547 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7548 runs initgroups(), so that if the Exim user is in any additional groups,
7549 they will be used during message reception.
7555 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / jgh / exim.git / blob