1 This document describes *changes* to previous versions, that might
2 affect Exim's operation, with an unchanged configuration file. For new
3 options, and new features, see the NewStuff file next to this ChangeLog.
6 Since Exim version 4.94
7 -----------------------
9 JH/02 Bug 2587: Fix pam expansion condition. Tainted values are commonly used
10 as arguments, so an implementation trying to copy these into a local
11 buffer was taking a taint-enforcement trap. Fix by using dynamically
12 created buffers. Similar fix for radius expansion condition.
14 JH/03 Bug 2586: Fix listcount expansion operator. Using tainted arguments is
15 reasonable, eg. to count headers. Fix by using dynamically created
16 buffers rather than a local. Do similar fixes for ACL actions "dcc",
17 "log_reject_target", "malware" and "spam"; the arguments are expanded
18 so could be handling tainted values.
20 JH/04 Bug 2590: Fix -bi (newaliases). A previous code rearrangement had
21 broken the (no-op) support for this sendmail command. Restore it
22 to doing nothing, silently, and returning good status.
24 JH/05 Bug 2593: Fix "vacation" in Exim filter. Previously, when a "once"
25 record path was given (or the default used) without a leading directory
26 path, an error occurred on trying to open it. Use the transport's working
29 JH/06 Bug 2594: Change the name used for certificate name checks in the smtp
30 transport. Previously it was the name on the DNS A-record; use instead
31 the head of the CNAME chain leading there (if there is one). This seems
32 to align better with RFC 6125.
34 JH/07 Bug 2597: Fix a resource leak. Using a lookup in obtaining a value for
35 smtp_accept_max_per_host allocated resources which were not released
36 when the limit was exceeded. This eventually crashed the daemon. Fix
37 by adding a relase action in that path.
39 JH/08 Bug 2598: Fix verify ACL condition. The options for the condition are
40 expanded; previously using tainted values was rejected. Fix by using
41 dynamically-created buffers.
43 JH/10 Bug 2603: Fix coding of string copying to only evaluate arguments once.
44 Previously a macro used one argument twice; when called with the
45 argument as an expression having side-effects, incorrect operation
46 resulted. Use an inlineable function.
48 JH/11 Bug 2604: Fix request to cutthrough-deliver when a connection is already
49 held open for a verify callout. Previously this wan not accounted for
50 and a corrupt onward SMTP conversation resulted.
52 JH/13 Fix dsearch "subdir" filter to ignore ".". Previously only ".." was
53 excluded, not matching the documentation.
55 JH/14 Bug 2606: Fix a segfault in sqlite lookups. When no, or a bad, filename
56 was given for the sqlite_dbfile a trap resulted.
58 JH/15 Bug 2620: Fix "spam" ACL condition. Previously, tainted values for the
59 "name" argument resulted in a trap. There is no reason to disallow such;
60 this was a coding error.
62 JH/16 Bug 2615: Fix pause during message reception, on systems that have been
63 suspended/resumed. The Linux CLOCK_MONOTONIC does not account for time
64 spent suspended, ignoring the Posix definition. Previously we assumed
65 it did and a constant offset from real time could be used as a correction.
66 Change to using the same clock source for the start-of-message and the
67 post-message next-tick-wait. Also change to using CLOCK_BOOTTIME if it
68 exists, just to get a clock slightly more aligned to reality.
70 JH/17 Bug 2295: Fix DKIM signing to always semicolon-terminate. Although the
71 RFC says it is optional some validators care. The missing char was not
72 intended but triggered by a line-wrap alignement. Discovery and fix by
73 Guillaume Outters, hacked on by JH.
75 JH/18 Bug 2617: Fix a taint trap in parse_fix_phrase(). Previously when the
76 name being quoted was tainted a trap would be taken. Fix by using
77 dynamicaly created buffers. The routine could have been called by a
78 rewrite with the "h" flag, by using the "-F" command-line option, or
79 by using a "name=" option on a control=submission ACL modifier.
81 JH/21 Bug 2630: Fix eol-replacement string for the ${readsocket } expansion.
82 Previously when a whitespace character was specified it was not inserted
83 after removing the newline.
85 JH/24 Bug 2634: Fix a taint trap seen on NetBSD: the testing coded for
86 is_tainted() had an off-by-one error in the overenthusiastic direction.
87 Find and fix by Gavan. Although NetBSD is not a supported platform for
88 4.94 this bug could affect other platforms.
90 JH/24 Bug 2634: Fix a taint trap seen on NetBSD: the testing coded for
91 is_tainted() had an off-by-one error in the overenthusiastic direction.
92 Find and fix by Gavan. Although NetBSD is not a supported platform for
93 4.94 this bug could affect other platforms.
94 JH/21 Bug 2630: Fix eol-replacement string for the ${readsocket } expansion.
95 Previously when a whitespace character was specified it was not inserted
96 after removing the newline.
98 JH/22 Bug 2265: Force SNI usage for smtp transport DANE'd connections, to be
99 the domain part of the recipient address. This overrides any tls_sni
100 option set, which was previously used.
102 JH/23 Logging: with the +tls_sni log_selector, do not wrap the received SNI
105 JH/26 Bug 2646: fix a memory usage issue in ldap lookups. Previously, when more
106 than one server was defined and depending on the platform memory layout
107 details, an internal consistency trap could be hit while walking the list
110 JH/27 Bug 2648: fix the passing of an authenticator public-name through spool
111 files. The value is used by the authresults expansion item. Previously
112 if this was used in a router or transport, a crash could result.
114 JH/30 Bug 2677: fix matching of long addresses. Since 4.93 a limit of 256 was
115 applied. This resulted, if any header-line rewrite rules were configured,
116 in a panic-log trigerrable by sending a message with a long address in
117 a header. Fix by increaing the arbitrary limit to larger than a single
118 (dewrapped) 5322 header line maximum size.
120 JH/31 The ESMTP option name advertised for the SUPPORT_EARLY_PIPE build option
121 is changed from X_PIPE_CONNECT to PIPE_CONNECT. This is in line with
122 RFC 6648 which deprecates X- options in protocols as a general practice.
123 Changeover between the implementations is handled by the mechanisms
126 JH/32 Bug 2599: fix delay of delivery to a local address where there is also
127 a remote which uses callout/hold. Previously the local was queued.
129 JH/33 Fix a taint trap in the ${listextract } expansion when the source data
132 JH/35 Bug 2343: Harden exim_tidydb against corrupt wait- files.
134 JH/36 Bug 2687: Fix interpretation of multiple ^ chars in a plaintext
135 authenticator client_send option. Previously the next char, after a pair
136 was collapsed, was taken verbatim (so ^^^foo became ^^foo; ^^^^foo became
137 ^^\x00foo). Fixed to get ^\x00foo and ^^foo respectively to match the
138 documentation. There is still no way to get a leading ^ immediately
139 after a NUL (ie. for the password of a PLAIN method authenticator.
141 JH/39 Bug 2691: fix $local_part_data. When the matching list element
142 referred to a file, bad data was returned. This likely also affected
145 JH/41 Fix daemon SIGHUP on FreeBSD. Previously, a named socket for IPC was
146 left undeleted; the attempt to re-create it then failed - resulting in
147 the usual "SIGHUP tp have daemon reload configuration" to not work.
148 This affected any platform not supporting "abstract" Unix-domain
149 sockets (i.e. not Linux).
151 JH/42 Bug 2692: Harden against a peer which reneges on a 452 "too many
152 recipients" response to RCPT in a later response, with a 250. The
153 previous coding assumed this would not happen, and under PIPELINING
154 would result in both lost and duplicate recipients for a message.
156 JH/43 Bug 2694: Fix weighted distribution of work to multiple spamd servers.
157 Previously the weighting was incorrectly applied. Similar fix for socks
158 proxies. Found and fixed by Heiko Schlichting.
160 JH/44 Bug 2701: Fix list-expansion of dns_ipv4_lookup. Previously, it did
161 not handle sub-lists included using the +namedlist syntax. While
162 investigating, the same found for dns_trust_aa, dns_again_means_nonexist,
163 dnssec_require_domains, dnssec_request_domains, srv_fail_domains,
166 HS/01 Enforce absolute PID file path name.
168 HS/02 Handle SIGINT as we handle SIGTERM: terminate the Exim process.
170 PP/01 Add a too-many-bad-recipients guard to the default config's RCPT ACL.
172 PP/02 Bug 2643: Correct TLS DH constants.
173 A missing NUL termination in our code-generation tool had led to some
174 incorrect Diffie-Hellman constants in the Exim source.
175 Reported by kylon94, code-gen tool fix by Simon Arlott.
177 PP/03 Impose security length checks on various command-line options.
178 Fixes CVE-2020-SPRSS reported by Qualys.
180 PP/04 Fix Linux security issue CVE-2020-SLCWD and guard against PATH_MAX
181 better. Reported by Qualys.
183 PP/05 Fix security issue CVE-2020-PFPSN and guard against cmdline invoker
184 providing a particularly obnoxious sender full name.
187 PP/06 Fix CVE-2020-28016 (PFPZA): Heap out-of-bounds write in parse_fix_phrase()
189 PP/07 Refuse to allocate too little memory, block negative/zero allocations.
192 PP/08 Change default for recipients_max from unlimited to 50,000.
194 PP/09 Fix security issue with too many recipients on a message (to remove a
195 known security problem if someone does set recipients_max to unlimited,
196 or if local additions add to the recipient list).
197 Fixes CVE-2020-RCPTL reported by Qualys.
199 PP/10 Fix security issue in SMTP verb option parsing
200 Fixes CVE-2020-EXOPT reported by Qualys.
202 PP/11 Fix security issue in BDAT state confusion.
203 Ensure we reset known-good where we know we need to not be reading BDAT
204 data, as a general case fix, and move the places where we switch to BDAT
205 mode until after various protocol state checks.
206 Fixes CVE-2020-BDATA reported by Qualys.
208 HS/03 Die on "/../" in msglog file names
210 QS/01 Creation of (database) files in $spool_dir: only uid=0 or the uid of
211 the Exim runtime user are allowed to create files.
213 QS/02 PID file creation/deletion: only possible if uid=0 or uid is the Exim
216 QS/03 When reading the output from interpreted forward files we do not
217 pass the pipe between the parent and the interpreting process to
218 executed child processes (if any).
220 QS/04 Always die if requested from internal logging, even is logging is
227 JH/01 Avoid costly startup code when not strictly needed. This reduces time
228 for some exim process initialisations. It does mean that the logging
229 of TLS configuration problems is only done for the daemon startup.
231 JH/02 Early-pipelining support code is now included unless disabled in Makefile.
233 JH/03 DKIM verification defaults no long accept sha1 hashes, to conform to
234 RFC 8301. They can still be enabled, using the dkim_verify_hashes main
237 JH/04 Support CHUNKING from an smtp transport using a transport_filter, when
238 DKIM signing is being done. Previously a transport_filter would always
239 disable CHUNKING, falling back to traditional DATA.
241 JH/05 Regard command-line receipients as tainted.
243 JH/06 Bug 340: Remove the daemon pid file on exit, whe due to SIGTERM.
245 JH/07 Bug 2489: Fix crash in the "pam" expansion condition. It seems that the
246 PAM library frees one of the arguments given to it, despite the
247 documentation. Therefore a plain malloc must be used.
249 JH/08 Bug 2491: Use tainted buffers for the transport smtp context. Previously
250 on-stack buffers were used, resulting in a taint trap when DSN information
251 copied from a received message was written into the buffer.
253 JH/09 Bug 2493: Harden ARC verify against Outlook, whick has been seen to mix
254 the ordering of its ARC headers. This caused a crash.
256 JH/10 Bug 2492: Use tainted memory for retry record when needed. Previously when
257 a new record was being constructed with information from the peer, a trap
260 JH/11 Bug 2494: Unset the default for dmarc_tld_file. Previously a naiive
261 installation would get error messages from DMARC verify, when it hit the
262 nonexistent file indicated by the default. Distros wanting DMARC enabled
263 should both provide the file and set the option.
264 Also enforce no DMARC verification for command-line sourced messages.
266 JH/12 Fix an uninitialised flag in early-pipelining. Previously connections
267 could, depending on the platform, hang at the STARTTLS response.
269 JH/13 Bug 2498: Reset a counter used for ARC verify before handling another
270 message on a connection. Previously if one message had ARC headers and
271 the following one did not, a crash could result when adding an
272 Authentication-Results: header.
274 JH/14 Bug 2500: Rewind some of the common-coding in string handling between the
275 Exim main code and Exim-related utities. The introduction of taint
276 tracking also did many adjustments to string handling. Since then, eximon
277 frequently terminated with an assert failure.
279 JH/15 When PIPELINING, synch after every hundred or so RCPT commands sent and
280 check for 452 responses. This slightly helps the inefficieny of doing
281 a large alias-expansion into a recipient-limited target. The max_rcpt
282 transport option still applies (and at the current default, will override
283 the new feature). The check is done for either cause of synch, and forces
284 a fast-retry of all 452'd recipients using a new MAIL FROM on the same
285 connection. The new facility is not tunable at this time.
287 JH/16 Fix the variables set by the gsasl authenticator. Previously a pointer to
288 library live data was being used, so the results became garbage. Make
289 copies while it is still usable.
291 JH/17 Logging: when the deliver_time selector ise set, include the DT= field
292 on delivery deferred (==) and failed (**) lines (if a delivery was
293 attemtped). Previously it was only on completion (=>) lines.
295 JH/18 Authentication: the gsasl driver not provides the $authN variables in time
296 for the expansion of the server_scram_iter and server_scram_salt options.
298 WB/01 SPF: DNS lookups for the obsolete SPF RR type done by the libspf2 library
299 are now specifically given a NO_DATA response without hitting the system
300 resolver. The library goes on to do the now-standard TXT lookup.
301 Use of dnsdb lookups is not affected.
303 JH/19 Bug 2507: Modules: on handling a dynamic-module (lookups) open failure,
304 only retrieve the errormessage once. Previously two calls to dlerror()
305 were used, and the second one (for mainlog/paniclog) retrieved null
308 JH/20 Taint checking: disallow use of tainted data for
309 - the appendfile transport file and directory options
310 - the pipe transport command
311 - the autoreply transport file, log and once options
312 - file names used by the redirect router (including filter files)
314 - paths used by single-key lookups
315 Previously this was permitted.
317 JH/21 Bug 2501: Fix init call in the heimdal authenticator. Previously it
318 adjusted the size of a major service buffer; this failed because the
319 buffer was in use at the time. Change to a compile-time increase in the
320 buffer size, when this authenticator is compiled into exim.
322 JH/22 Taint-checking: move to safe-mode taint checking on all platforms. The
323 previous fast-mode was untenable in the face of glibs using mmap to
324 support larger malloc requests.
326 PP/01 Update the openssl_options possible values through OpenSSL 1.1.1c.
327 New values supported, if defined on system where compiled:
328 allow_no_dhe_kex, cryptopro_tlsext_bug, enable_middlebox_compat,
329 no_anti_replay, no_encrypt_then_mac, prioritize_chacha, tlsext_padding
331 JH/23 Performance improvement in the initial phase of a two-pass queue run. By
332 running a limited number of proceses in parallel, a benefit is gained. The
333 amount varies with the platform hardware and load. The use of the option
334 queue_run_in_order means we cannot do this, as ordering becomes
337 JH/24 Bug 2524: fix the cyrus_sasl auth driver gssapi usage. A previous fix
338 had introduced a string-copy (for ensuring NUL-termination) which was not
339 appropriate for that case, which can include embedded NUL bytes in the
340 block of data. Investigation showed the copy to actually be needless, the
341 data being length-specified.
343 JH/25 Fix use of concurrent TLS connections under GnuTLS. When a callout was
344 done during a receiving connection, and both used TLS, global info was
345 used rather than per-connection info for tracking the state of data
346 queued for transmission. This could result in a connection hang.
348 JH/26 Fix use of the SIZE parameter on MAIL commands, on continued connections.
349 Previously, when delivering serveral messages down a single connection
350 only the first would provide a SIZE. This was due to the size information
351 not being properly tracked.
353 JH/27 Bug 2530: When operating in a timezone with sub-minute offset, such as
354 TAI (at 37 seconds currently), pretend to be in UTC for time-related
355 expansion and logging. Previously, spurious values such as a future
356 minute could be seen.
358 JH/28 Bug 2533: Fix expansion of ${tr } item. When called in some situations
359 it could crash from a null-deref. This could also affect the
360 ${addresses: } operator and ${readsock } item.
362 JH/29 Bug 2537: Fix $mime_part_count. When a single connection had a non-mime
363 message following a mime one, the variable was not reset.
365 JH/30 When an pipelined-connect fails at the first response, assume incorrect
366 cached capability (perhaps the peer reneged?) and immediately retry in
369 JH/31 Fix spurious detection of timeout while writing to transport filter.
371 JH/32 Bug 2541: Fix segfault on bad cmdline -f (sender) argument. Previously
372 an attempt to copy the string was made before checking it.
374 JH/33 Fix the dsearch lookup to return an untainted result. Previously the
375 taint of the lookup key was maintained; we now regard the presence in the
376 filesystem as sufficient validation.
378 JH/34 Fix the readsocket expansion to not segfault when an empty "options"
379 argument is supplied.
381 JH/35 The dsearch lookup now requires that the directory is an absolute path.
382 Previously this was not checked, and nonempty relative paths made an
383 access under Exim's current working directory.
385 JH/36 Bug 2554: Fix msg:defer event for the hosts_max_try_hardlimit case.
386 Previously no event was raised.
388 JH/37 Bug 2552: Fix the check on spool space during reception to use the SIZE
389 parameter supplied by the sender MAIL FROM command. Previously it was
390 ignored, and only the check_spool_space option value for the required
393 JH/38 Fix $dkim_key_length. This should, after a DKIM verification, present
394 the size of the signing public-key. Previously it was instead giving
395 the size of the signature hash.
397 JH/39 DKIM verification: the RFC 8301 restriction on sizes of RSA keys is now
398 the default. See the (new) dkim_verify_min_keysizes option.
400 JH/40 Fix a memory-handling bug: when a connection carried multiple messages
401 and an ACL use a lookup for checking either the local_part or domain,
402 stale data could be accessed. Ensure that variable references are
403 dropped between messages.
405 JH/41 Bug 2571: Fix SPA authenticator. Running as a server, an offset supplied
406 by the client was not checked as pointing within response data before
407 being used. A malicious client could thus cause an out-of-bounds read and
408 possibly gain authentication. Fix by adding the check.
410 JH/42 Internationalisation: change the default for downconversion in the smtp
411 transport to be "if needed". Previously it was "as previously set" for
412 the message, which usually meant "if needed" for message-submission but
413 "no" for everything else. However, MTAs have been seen using SMTPUTF8
414 even when the envelope addresses did not need it, resulting in forwarding
415 failures to non-supporting MTAs. A downconvert in such cases will be
416 a no-op on the addresses, merely dropping the use of SMTPUTF8 by the
417 transport. The change does mean that addresses needing conversion will
418 be converted when previously a delivery failure would occur.
420 JH/43 Fix possible long line in DSN. Previously when a very long SMTP error
421 response was received it would be used unchecked in a fail-DSN, violating
422 standards on line-length limits. Truncate if needed.
424 HS/01 Remove parameters of the link to www.open-spf.org. The linked form
425 doesn't work. (Additionally add a new main config option to configure the
432 JH/01 OpenSSL: With debug enabled output keying information sufficient, server
433 side, to decode a TLS 1.3 packet capture.
435 JH/02 OpenSSL: Suppress the sending of (stateful) TLS1.3 session tickets.
436 Previously the default library behaviour applied, sending two, each in
439 JH/03 Debug output for ACL now gives the config file name and line number for
442 JH/04 The default received_header_text now uses the RFC 8314 tls cipher clause.
444 JH/05 DKIM: ensure that dkim_domain elements are lowercased before use.
446 JH/06 Fix buggy handling of autoreply bounce_return_size_limit, and a possible
447 buffer overrun for (non-chunking) other transports.
449 JH/07 GnuTLS: Our use of late (post-handshake) certificate verification, under
450 TLS1.3, means that a server rejecting a client certificate is not visible
451 to the client until the first read of encrypted data (typically the
452 response to EHLO). Add detection for that case and treat it as a failed
453 TLS connection attempt, so that the normal retry-in-clear can work (if
454 suitably configured).
456 JB/01 Bug 2375: fix expansions of 822 addresses having comments in local-part
457 and/or domain. Found and fixed by Jason Betts.
459 JH/08 Add hardening against SRV & TLSA lookups the hit CNAMEs (a nonvalid
460 configuration). If a CNAME target was not a wellformed name pattern, a
463 JH/09 Logging: Fix initial listening-on line for multiple ports for an IP when
464 the OS reports them interleaved with other addresses.
466 JH/10 OpenSSL: Fix aggregation of messages. Previously, when PIPELINING was
467 used both for input and for a verify callout, both encrypted, SMTP
468 responses being sent by the server could be lost. This resulted in
469 dropped connections and sometimes bounces generated by a peer sending
472 JH/11 Harden plaintext authenticator against a badly misconfigured client-send
473 string. Previously it was possible to cause undefined behaviour in a
474 library routine (usually a crash). Found by "zerons".
476 JH/12 Bug 2384: fix "-bP smtp_receive_timeout". Previously it returned no
479 JH/13 Bug 2386: Fix builds with Dane under LibreSSL 2.9.0 onward. Some old
480 API was removed, so update to use the newer ones.
482 JH/14 Bug 1891: Close the log file if receiving a non-smtp message, without
483 any timeout set, is taking a long time. Previously we would hang on to a
484 rotated logfile "forever" if the input was arriving with long gaps
485 (a previous attempt to fix addressed lack, for a long time, of initial
488 HS/01 Bug 2390: Use message_id for tempfile creation to avoid races in a
489 shared (NFS) environment. The length of the tempfile name is now
490 4 + 16 ("hdr.$message_exim_id") which might break on file
491 systems which restrict the file name length to lower values.
494 HS/02 Bug 2390: Use message_id for tempfile creation to avoid races in a
495 shared (NFS) environment.
497 HS/03 Bug 2392: exigrep does case sensitive *option* processing (as it
498 did for all versions <4.90). Notably -M, -m, --invert, -I may be
501 JH/15 Use unsigned when creating bitmasks in macros, to avoid build errors
502 on some platforms for bit 31.
504 JH/16 GnuTLS: rework ciphersuite strings under recent library versions. Thanks
505 to changes apparently associated with TLS1.3 handling some of the APIs
506 previously used were either nonfunctional or inappropriate. Strings
507 like TLS1.3:ECDHE_SECP256R1__RSA_PSS_RSAE_SHA256__AES_256_GCM__AEAD:256
508 and TLS1.2:ECDHE_SECP256R1__RSA_SHA256__AES_128_CBC__SHA256:128 replace
509 the previous TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256 .
510 This affects log line X= elements, the $tls_{in,out}_cipher variables,
511 and the use of specific cipher names in the encrypted= ACL condition.
513 JH/17 OpenSSL: the default openssl_options now disables ssl_v3.
515 JH/18 GnuTLS: fix $tls_out_ocsp under hosts_request_ocsp. Previously the
516 verification result was not updated unless hosts_require_ocsp applied.
518 JH/19 Bug 2398: fix listing of a named-queue. Previously, even with the option
519 queue_list_requires_admin set to false, non-admin users were denied the
522 JH/20 Bug 2389: fix server advertising of usable certificates, under GnuTLS in
523 directory-of-certs mode. Previously they were advertised despite the
526 JH/21 The smtp transport option "hosts_noproxy_tls" is now unset by default.
527 A single TCP connection by a client will now hold a TLS connection open
528 for multiple message deliveries, by default. Previously the default was to
531 JH/22 The smtp transport option "hosts_try_dane" now enables all hosts by
532 default. If built with the facility, DANE will be used. The facility
533 SUPPORT_DANE is now enabled in the prototype build Makefile "EDITME".
535 JH/23 The build default is now for TLS to be included; the SUPPORT_TLS define
536 is replaced with DISABLE_TLS. Either USE_GNUTLS or (the new) USE_OPENSSL
537 must be defined and you must still, unless you define DISABLE_TLS, manage
538 the the include-dir and library-file requirements that go with that
539 choice. Non-TLS builds are still supported.
541 JH/24 Fix duplicated logging of peer name/address, on a transport connection-
544 JH/25 The smtp transport option "hosts_try_fastopen" now enables all hosts by
545 default. If the platform supports and has the facility enabled, it will
546 be requested on all coneections.
548 JH/26 The PIPE_CONNECT facility is promoted from experimental status and is now
549 controlled by the build-time option SUPPORT_PIPE_CONNECT.
551 PP/01 Unbreak heimdal_gssapi, broken in 4.92.
553 JH/27 Bug 2404: Use the main-section configuration option "dsn_from" for
554 success-DSN messages. Previously the From: header was always the default
555 one for these; the option was ignored.
557 JH/28 Fix the timeout on smtp response to apply to the whole response.
558 Previously it was reset for every read, so a teergrubing peer sending
559 single bytes within the time limit could extend the connection for a
560 long time. Credit to Qualsys Security Advisory Team for the discovery.
562 JH/29 Fix DSN Final-Recipient: field. Previously it was the post-routing
563 delivery address, which leaked information of the results of local
564 forwarding. Change to the original envelope recipient address, per
567 JH/30 Bug 2411: Fix DSN generation when RFC 3461 failure notification is
568 requested. Previously not bounce was generated and a log entry of
569 error ignored was made.
571 JH/31 Avoid re-expansion in ${sort } expansion. (CVE-2019-13917)
573 JH/32 Introduce a general tainting mechanism for values read from the input
574 channel, and values derived from them. Refuse to expand any tainted
575 values, to catch one form of exploit.
577 JH/33 Bug 2413: Fix dkim_strict option. Previously the expansion result
578 was unused and the unexpanded text used for the test. Found and
579 fixed by Ruben Jenster.
581 JH/34 Fix crash after TLS shutdown. When the TCP/SMTP channel was left open,
582 an attempt to use a TLS library read routine dereffed a nul pointer,
585 JH/35 Bug 2409: filter out-of-spec chars from callout response before using
586 them in our smtp response.
588 JH/36 Have the general router option retry_use_local_part default to true when
589 any of the restrictive preconditions are set (to anything). Previously it
590 was only for check_local user. The change removes one item of manual
591 configuration which is required for proper retries when a remote router
592 handles a subset of addresses for a domain.
594 JH/37 Appendfile: when evaluating quota use (non-quota_size_regex) take the file
595 link count into consideration.
597 HS/04 Fix handling of very log lines in -H files. If a -<key> <value> line
598 caused the extension of big_buffer, the following lines were ignored.
600 JH/38 Bug 1395: Teach the DNS negative-cache about TTL value from the SOA in
601 accordance with RFC 2308. Previously there was no expiry, so a longlived
602 receive process (eg. due to ACL delays) versus a short SOA value could
605 HS/05 Handle trailing backslash gracefully. (CVE-2019-15846)
607 JH/39 Promote DMARC support to mainline.
609 JH/40 Bug 2452: Add a References: header to DSNs.
611 JH/41 With GnuTLS 3.6.0 (and later) do not attempt to manage Diffie-Hellman
612 parameters. The relevant library call is documented as "Deprecated: This
613 function is unnecessary and discouraged on GnuTLS 3.6.0 or later. Since
614 3.6.0, DH parameters are negotiated following RFC7919."
616 HS/06 Change the default of dnssec_request_domains to "*"
618 JH/42 Bug 2545: Fix CHUNKING for all RCPT commands rejected. Previously we
619 carried on and emitted a BDAT command, even when PIPELINING was not
622 JH/43 Bug 2465: Fix taint-handling in dsearch lookup. Previously a nontainted
623 buffer was used for the filename, resulting in a trap when tainted
624 arguments (eg. $domain) were used.
626 JH/44 With OpenSSL 1.1.1 (onwards) disable renegotiation for TLS1.2 and below;
627 recommended to avoid a possible server-load attack. The feature can be
628 re-enabled via the openssl_options main cofiguration option.
630 JH/45 local_scan API: documented the current smtp_printf() call. This changed
631 for version 4.90 - adding a "more data" boolean to the arguments.
632 Bumped the ABI version number also, this having been missed previously;
633 release versions 4.90 to 4.92.3 inclusive were effectively broken in
634 respect of usage of smtp_printf() by either local_scan code or libraries
635 accessed via the ${dlfunc } expansion item. Both will need coding
636 adjustment for any calls to smtp_printf() to match the new function
637 signature; a FALSE value for the new argument is always safe.
639 JH/46 FreeBSD: fix use of the sendfile() syscall. The shim was not updating
640 the file-offset (which the Linux syscall does, and exim expects); this
641 resulted in an indefinite loop.
643 JH/47 ARC: fix crash in signing, triggered when a configuration error failed
644 to do ARC verification. The Authentication-Results: header line added
645 by the configuration then had no ARC item.
651 JH/01 Remove code calling the customisable local_scan function, unless a new
652 definition "HAVE_LOCAL_SCAN=yes" is present in the Local/Makefile.
654 JH/02 Bug 1007: Avoid doing logging from signal-handlers, as that can result in
655 non-signal-safe functions being used.
657 JH/03 Bug 2269: When presented with a received message having a stupidly large
658 number of DKIM-Signature headers, disable DKIM verification to avoid
659 a resource-consumption attack. The limit is set at twenty.
661 JH/04 Add variables $arc_domains, $arc_oldest_pass for ARC verify. Fix the
662 report of oldest_pass in ${authres } in consequence, and separate out
663 some descriptions of reasons for verification fail.
665 JH/05 Bug 2273: Cutthrough delivery left a window where the received messsage
666 files in the spool were present and unlocked. A queue-runner could spot
667 them, resulting in a duplicate delivery. Fix that by doing the unlock
668 after the unlink. Investigation by Tim Stewart. Take the opportunity to
669 add more error-checking on spoolfile handling while that code is being
672 PP/01 Refuse to open a spool data file (*-D) if it's a symlink.
673 No known attacks, no CVE, this is defensive hardening.
675 JH/06 Bug 2275: The MIME ACL unlocked the received message files early, and
676 a queue-runner could start a delivery while other operations were ongoing.
677 Cutthrough delivery was a common victim, resulting in duplicate delivery.
678 Found and investigated by Tim Stewart. Fix by using the open message data
679 file handle rather than opening another, and not locally closing it (which
680 releases a lock) for that case, while creating the temporary .eml format
681 file for the MIME ACL. Also applies to "regex" and "spam" ACL conditions.
683 JH/07 Bug 177: Make a random-recipient callout success visible in ACL, by setting
684 $sender_verify_failure/$recipient_verify_failure to "random".
686 JH/08 When generating a selfsigned cert, use serial number 1 since zero is not
689 JH/09 Bug 2274: Fix logging of cmdline args when starting in an unlinked cwd.
690 Previously this would segfault.
692 JH/10 Fix ARC signing for case when DKIM signing failed. Previously this would
695 JH/11 Bug 2264: Exim now only follows CNAME chains one step by default. We'd
696 like zero, since the resolver should be doing this for us, But we need one
697 as a CNAME but no MX presence gets the CNAME returned; we need to check
698 that doesn't point to an MX to declare it "no MX returned" rather than
699 "error, loop". A new main option is added so the older capability of
700 following some limited number of chain links is maintained.
702 JH/12 Add client-ip info to non-pass iprev ${authres } lines.
704 JH/13 For receent Openssl versions (1.1 onward) use modern generic protocol
705 methods. These should support TLS 1.3; they arrived with TLS 1.3 and the
706 now-deprecated earlier definitions used only specified the range up to TLS
707 1.2 (in the older-version library docs).
709 JH/14 Bug 2284: Fix DKIM signing for body lines starting with a pair of dots.
711 JH/15 Rework TLS client-side context management. Stop using a global, and
712 explicitly pass a context around. This enables future use of TLS for
713 connections to service-daemons (eg. malware scanning) while a client smtp
714 connection is using TLS; with cutthrough connections this is quite likely.
716 JH/16 Fix ARC verification to do AS checks in reverse order.
718 JH/17 Support a "tls" option on the ${readsocket } expansion item.
720 JH/18 Bug 2287: Fix the protocol name (eg utf8esmtp) for multiple messages
721 using the SMTPUTF8 option on their MAIL FROM commands, in one connection.
722 Previously the "utf8" would be re-prepended for every additional message.
724 JH/19 Reject MAIL FROM commands with SMTPUTF8 when the facility was not advertised.
725 Previously thery were accepted, resulting in issues when attempting to
726 forward messages to a non-supporting MTA.
728 PP/02 Let -n work with printing macros too, not just options.
730 JH/20 Bug 2296: Fix cutthrough for >1 address redirection. Previously only
731 one parent address was copied, and bogus data was used at delivery-logging
732 time. Either a crash (after delivery) or bogus log data could result.
733 Discovery and analysis by Tim Stewart.
735 PP/03 Make ${utf8clean:} expansion operator detect incomplete final character.
736 Previously if the string ended mid-character, we did not insert the
737 promised '?' replacement.
739 PP/04 Documentation: current string operators work on bytes, not codepoints.
741 JH/21 Change as many as possible of the global flags into one-bit bitfields; these
742 should pack well giving a smaller memory footprint so better caching and
743 therefore performance. Group the declarations where this can't be done so
744 that the byte-sized flag variables are not interspersed among pointer
745 variables, giving a better chance of good packing by the compiler.
747 JH/22 Bug 1896: Fix the envelope from for DMARC forensic reports to be possibly
748 non-null, to avoid issues with sites running BATV. Previously reports were
749 sent with an empty envelope sender so looked like bounces.
751 JH/23 Bug 2318: Fix the noerror command within filters. It wasn't working.
752 The ignore_error flag wasn't being returned from the filter subprocess so
753 was not set for later routers. Investigation and fix by Matthias Kurz.
755 JH/24 Bug 2310: Raise a msg:fail:internal event for each undelivered recipient,
756 and a msg:complete for the whole, when a message is manually removed using
757 -Mrm. Developement by Matthias Kurz, hacked on by JH.
759 JH/25 Avoid fixed-size buffers for pathnames in DB access. This required using
760 a "Gnu special" function, asprintf() in the DB utility binary builds; I
761 hope that is portable enough.
763 JH/26 Bug 2311: Fix DANE-TA verification under GnuTLS. Previously it was also
764 requiring a known-CA anchor certificate; make it now rely entirely on the
765 TLSA as an anchor. Checking the name on the leaf cert against the name
766 on the A-record for the host is still done for TA (but not for EE mode).
768 JH/27 Fix logging of proxy address. Previously, a pointless "PRX=[]:0" would be
769 included in delivery lines for non-proxied connections, when compiled with
770 SUPPORT_SOCKS and running with proxy logging enabled.
772 JH/28 Bug 2314: Fire msg:fail:delivery event even when error is being ignored.
773 Developement by Matthias Kurz, tweaked by JH. While in that bit of code,
774 move the existing event to fire before the normal logging of message
775 failure so that custom logging is bracketed by normal logging.
777 JH/29 Bug 2322: A "fail" command in a non-system filter (file) now fires the
778 msg:fail:internal event. Developement by Matthias Kurz.
780 JH/30 Bug 2329: Increase buffer size used for dns lookup from 2k, which was
781 far too small for todays use of crypto signatures stored there. Go all
782 the way to the max DNS message size of 64kB, even though this might be
783 overmuch for IOT constrained device use.
785 JH/31 Fix a bad use of a copy function, which could be used to pointlessly
786 copy a string over itself. The library routine is documented as not
787 supporting overlapping copies, and on MacOS it actually raised a SIGABRT.
789 JH/32 For main options check_spool_space and check_inode_space, where the
790 platform supports 64b integers, support more than the previous 2^31 kB
791 (i.e. more than 2 TB). Accept E, P and T multipliers in addition to
792 the previous G, M, k.
794 JH/33 Bug 2338: Fix the cyrus-sasl authenticator to fill in the
795 $authenticated_fail_id variable on authentication failure. Previously
798 JH/34 Increase RSA keysize of autogen selfsign cert from 1024 to 2048. RHEL 8.0
799 OpenSSL didn't want to use such a weak key. Do for GnuTLS also, and for
800 more-modern GnuTLS move from GNUTLS_SEC_PARAM_LOW to
801 GNUTLS_SEC_PARAM_MEDIUM.
803 JH/35 OpenSSL: fail the handshake when SNI processing hits a problem, server
804 side. Previously we would continue as if no SNI had been received.
806 JH/36 Harden the handling of string-lists. When a list consisted of a sole
807 "<" character, which should be a list-separator specification, we walked
808 off past the nul-terimation.
810 JH/37 Bug 2341: Send "message delayed" warning MDNs (restricted to external
811 causes) even when the retry time is not yet met. Previously they were
812 not, meaning that when (say) an account was over-quota and temp-rejecting,
813 and multiple senders' messages were queued, only one sender would get
814 notified on each configured delay_warning cycle.
816 JH/38 Bug 2351: Log failures to extract envelope addresses from message headers.
818 JH/39 OpenSSL: clear the error stack after an SSL_accept(). With anon-auth
819 cipher-suites, an error can be left on the stack even for a succeeding
820 accept; this results in impossible error messages when a later operation
823 AM/01 Bug 2359: GnuTLS: repeat lowlevel read and write operations while they
824 return error codes indicating retry. Under TLS1.3 this becomes required.
826 JH/40 Fix the feature-cache refresh for EXPERIMENTAL_PIPE_CONNECT. Previously
827 it only wrote the new authenticators, resulting in a lack of tracking of
828 peer changes of ESMTP extensions until the next cache flush.
830 JH/41 Fix the loop reading a message header line to check for integer overflow,
831 and more-often against header_maxsize. Previously a crafted message could
832 induce a crash of the recive process; now the message is cleanly rejected.
834 JH/42 Bug 2366: Fix the behaviour of the dkim_verify_signers option. It had
835 been totally disabled for all of 4.91. Discovery and fix by "Mad Alex".
841 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
842 When redis_servers is set to a list of > 1 element, and the Redis servers
843 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
844 case of MOVED into a DEFER case instead, thus moving the query onto the
845 next server in the list. For a cluster of N elements, all N servers must
846 be defined in redis_servers.
848 GF/02 Catch and remove uninitialized value warning in exiqsumm
849 Check for existence of @ARGV before looking at $ARGV[0]
851 JH/01 Replace the store_release() internal interface with store_newblock(),
852 which internalises the check required to safely use the old one, plus
853 the allocate and data copy operations duplicated in both (!) of the
854 extant use locations.
856 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
857 modifier. This matches the restriction on the commandline.
859 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
860 Previously only the last row was returned.
862 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
863 we assumed that tags in the header were well-formed, and parsed the
864 element content after inspecting only the first char of the tag.
865 Assumptions at that stage could crash the receive process on malformed
868 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
869 While running the DKIM ACL we operate on the Permanent memory pool so that
870 variables created with "set" persist to the DATA ACL. Also (at any time)
871 DNS lookups that fail create cache records using the Permanent pool. But
872 expansions release any allocations made on the current pool - so a dnsdb
873 lookup expansion done in the DKIM ACL releases the memory used for the
874 DNS negative-cache, and bad things result. Solution is to switch to the
875 Main pool for expansions.
876 While we're in that code, add checks on the DNS cache during store_reset,
877 active in the testsuite.
878 Problem spotted, and debugging aided, by Wolfgang Breyha.
880 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
881 When none of the hosts presented to a transport match an already-open
882 connection, close it and proceed with the list. Previously we would
883 queue the message. Spotted by Lena with Yahoo, probably involving
886 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
887 Previously a spurious "250 OK id=" response was appended to the proper
890 JH/08 The "support for" informational output now, which built with Content
891 Scanning support, has a line for the malware scanner interfaces compiled
892 in. Interface can be individually included or not at build time.
894 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
895 by the template makefile "src/EDITME". The "STREAM" support for an older
896 ClamAV interface method is removed.
898 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
899 rows affected is given instead).
901 JH/11 The runtime Berkeley DB library version is now additionally output by
902 "exim -d -bV". Previously only the compile-time version was shown.
904 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
905 SMTP connection. Previously, when one had more receipients than the
906 first, an abortive onward connection was made. Move to full support for
907 multiple onward connections in sequence, handling cutthrough connection
908 for all multi-message initiating connections.
910 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
911 routers. Previously, a multi-recipient message would fail to match the
912 onward-connection opened for the first recipient, and cause its closure.
914 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
915 a timeout on read on a GnuTLS initiating connection, resulting in the
916 initiating connection being dropped. This mattered most when the callout
917 was marked defer_ok. Fix to keep the two timeout-detection methods
920 JH/15 Relax results from ACL control request to enable cutthrough, in
921 unsupported situations, from error to silently (except under debug)
922 ignoring. This covers use with PRDR, frozen messages, queue-only and
925 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
927 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
928 metadata, resulting in a crash in free().
930 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
931 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
932 Broken also in d185889f4, with init system revamp.
934 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
935 Previously we abruptly closed the connection after reading a malware-
936 found indication; now we go on to read the "scan ok" response line,
939 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
940 ACL. Previously, a crash would result.
942 JH/19 Speed up macro lookups during configuration file read, by skipping non-
943 macro text after a replacement (previously it was only once per line) and
944 by skipping builtin macros when searching for an uppercase lead character.
946 JH/20 DANE support moved from Experimental to mainline. The Makefile control
947 for the build is renamed.
949 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
950 was allocated for every new TLS startup, meaning one per message. Fix
951 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
953 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
954 reported the original. Fix to report (as far as possible) the ACL
955 result replacing the original.
957 JH/23 Fix memory leak during multi-message connections using STARTTLS under
958 OpenSSL. Certificate information is loaded for every new TLS startup,
959 and the resources needed to be freed.
961 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
963 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
966 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
967 DATA response info to the (existing) per-recipient response info for
968 the "C=" log element. It can have useful tracking info from the
969 destination system. Patch from Simon Arlott.
971 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
972 length value. Previously this would segfault.
974 HS/02 Support Avast multiline protoocol, this allows passing flags to
975 newer versions of the scanner.
977 JH/28 Ensure that variables possibly set during message acceptance are marked
978 dead before release of memory in the daemon loop. This stops complaints
979 about them when the debug_store option is enabled. Discovered specifically
980 for sender_rate_period, but applies to a whole set of variables.
981 Do the same for the queue-runner and queue-list loops, for variables set
982 from spool message files. Do the same for the SMTP per-message loop, for
983 certain variables indirectly set in ACL operations.
985 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
986 as a multi-recipient message from a mailinglist manager). The coding had
987 an arbitrary cutoff number of characters while checking for more input;
988 enforced by writing a NUL into the buffer. This corrupted long / fast
989 input. The problem was exposed more widely when more pipelineing of SMTP
990 responses was introduced, and one Exim system was feeding another.
991 The symptom is log complaints of SMTP syntax error (NUL chars) on the
992 receiving system, and refused recipients seen by the sending system
993 (propating to people being dropped from mailing lists).
994 Discovered and pinpointed by David Carter.
996 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
997 replaced by the ${authresults } expansion.
999 JH/31 Bug 2257: Fix pipe transport to not use a socket-only syscall.
1001 HS/03 Set a handler for SIGTERM and call exit(3) if running as PID 1. This
1002 allows proper process termination in container environments.
1004 JH/32 Bug 2258: Fix spool_wireformat in combination with LMTP transport.
1005 Previously the "final dot" had a newline after it; ensure it is CR,LF.
1007 JH/33 SPF: remove support for the "spf" ACL condition outcome values "err_temp"
1008 and "err_perm", deprecated since 4.83 when the RFC-defined words
1009 "temperror" and "permerror" were introduced.
1011 JH/34 Re-introduce enforcement of no cutthrough delivery on transports having
1012 transport-filters or DKIM-signing. The restriction was lost in the
1013 consolidation of verify-callout and delivery SMTP handling.
1014 Extend the restriction to also cover ARC-signing.
1016 JH/35 Cutthrough: for a final-dot response timeout (and nonunderstood responses)
1017 in defer=pass mode supply a 450 to the initiator. Previously the message
1020 PP/02 DANE: add dane_require_tls_ciphers SMTP Transport option; if unset,
1021 tls_require_ciphers is used as before.
1023 HS/03 Malware Avast: Better match the Avast multiline protocol. Add
1024 "pass_unscanned". Only tmpfails from the scanner are written to
1025 the paniclog, as they may require admin intervention (permission
1026 denied, license issues). Other scanner errors (like decompression
1027 bombs) do not cause a paniclog entry.
1029 JH/36 Fix reinitialisation of DKIM logging variable between messages.
1030 Previously it was possible to log spurious information in receive log
1033 JH/37 Bug 2255: Revert the disable of the OpenSSL session caching. This
1034 triggered odd behaviour from Outlook Express clients.
1036 PP/03 Add util/renew-opendmarc-tlds.sh script for safe renewal of public
1039 JH/38 DKIM: accept Ed25519 pubkeys in SubjectPublicKeyInfo-wrapped form,
1040 since the IETF WG has not yet settled on that versus the original
1041 "bare" representation.
1043 JH/39 Fix syslog logging for syslog_timestamp=no and log_selector +millisec.
1044 Previously the millisecond value corrupted the output.
1045 Fix also for syslog_pid=no and log_selector +pid, for which the pid
1046 corrupted the output.
1052 JH/01 Rework error string handling in TLS interface so that the caller in
1053 more cases is responsible for logging. This permits library-sourced
1054 string to be attached to addresses during delivery, and collapses
1055 pairs of long lines into single ones.
1057 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
1058 during configuration. Wildcards are allowed and expanded.
1060 JH/02 Rework error string handling in DKIM to pass more info back to callers.
1061 This permits better logging.
1063 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
1064 do not close it down and have the child transport start it up again on
1065 the passed-on TCP connection. Instead, proxy the child (and any
1066 subsequent ones) for TLS via a unix-domain socket channel. Logging is
1067 affected: the continued delivery log lines do not have any DNSSEC, TLS
1068 Certificate or OCSP information. TLS cipher information is still logged.
1070 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
1071 identical IP addresses on different listening ports. Will also affect
1074 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
1075 add noisy ifdef guards to special-case this sillyness.
1076 Patch from Bernd Kuhls.
1078 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
1079 than 255 are no longer allowed.
1081 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
1082 Disable the session-cache too, which might reduce our load. Since we
1083 currrectly use a new context for every connection, both as server and
1084 client, there is no benefit for these.
1085 GnuTLS appears to not support tickets server-side by default (we don't
1086 call gnutls_session_ticket_enable_server()) but client side is enabled
1087 by default on recent versions (3.1.3 +) unless the PFS priority string
1090 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
1091 <https://reproducible-builds.org/specs/source-date-epoch/>.
1093 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
1094 the check for any unsuccessful recipients did not notice the limit, and
1095 erroneously found still-pending ones.
1097 JH/08 Pipeline CHUNKING command and data together, on kernels that support
1098 MSG_MORE. Only in-clear (not on TLS connections).
1100 JH/09 Avoid using a temporary file during transport using dkim. Unless a
1101 transport-filter is involved we can buffer the headers in memory for
1102 creating the signature, and read the spool data file once for the
1103 signature and again for transmission.
1105 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
1106 4.77 as the kernel support then wasn't solid, having issues in 64bit
1107 mode. Now, it's been long enough. Add support for FreeBSD also.
1109 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
1110 case where the routing stage had gathered several addresses to send to
1111 a host before calling the transport for the first, we previously failed
1112 to close down TLS in the old transport process before passing the TCP
1113 connection to the new process. The new one sent a STARTTLS command
1114 which naturally failed, giving a failed delivery and bloating the retry
1115 database. Investigation and fix prototype from Wolfgang Breyha.
1117 JH/12 Fix check on SMTP command input synchronisation. Previously there were
1118 false-negatives in the check that the sender had not preempted a response
1119 or prompt from Exim (running as a server), due to that code's lack of
1120 awareness of the SMTP input buffering.
1122 PP/04 Add commandline_checks_require_admin option.
1123 Exim drops privileges sanely, various checks such as -be aren't a
1124 security problem, as long as you trust local users with access to their
1125 own account. When invoked by services which pass untrusted data to
1126 Exim, this might be an issue. Set this option in main configuration
1127 AND make fixes to the calling application, such as using `--` to stop
1130 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
1131 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
1132 responses to those, into a single TLS record each way (this usually means
1133 a single packet). As a side issue, smtp_enforce_sync now works on TLS
1136 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
1137 affects you only if you're dancing at the edge of the param size limits.
1138 If you are, and this message makes sense to you, then: raise the
1139 configured limit or use OpenSSL 1.1. Nothing we can do for older
1142 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
1143 cmdline element to get the documented default one. Previously it was
1146 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
1147 are unsuable for non-message-specific reasons - eg. connection timeout,
1148 banner-time rejection.
1150 JH/16 Fix logging of delivery remote port, when specified by router, under
1153 PP/06 Repair manualroute's ability to take options in any order, even if one
1154 is the name of a transport.
1157 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
1159 JH/17 Change the list-building routines interface to use the expanding-string
1160 triplet model, for better allocation and copying behaviour.
1162 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
1163 Previously it was constructed the first time a possibly-matching string
1164 was met in the configuration file input during startup; now it is done
1167 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
1168 compatible one, to avoid the (poorly documented) possibility of a config
1169 file in the working directory redirecting the DB files, possibly correpting
1170 some existing file. CVE-2017-10140 assigned for BDB.
1172 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
1173 cache-hot. Previously, although the result was properly cached, the
1174 initial verify call returned a defer.
1176 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
1177 the main verify for receipient in uncached-mode.
1179 JH/22 Retire historical build files to an "unsupported" subdir. These are
1180 defined as "ones for which we have no current evidence of testing".
1182 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
1183 if present. Previously it was ignored.
1185 JH/24 Start using specified-initialisers in C structure init coding. This is
1186 a C99 feature (it's 2017, so now considered safe).
1188 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
1189 if was a fixed-sized field and bitmask ops via macros; it is now more
1192 PP/07 GitHub PR 56: Apply MariaDB build fix.
1193 Patch provided by Jaroslav Škarvada.
1195 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
1196 during Coverity cleanups [4.87 JH/47]
1197 Diagnosis and fix provided by Michael Fischer v. Mollard.
1199 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
1200 the right size to place the terminating semicolon on its own folded
1201 line, the header hash was calculated to an incorrect value thanks to
1202 the (relaxed) space the fold became.
1204 HS/02 Fix Bug 2130: large writes from the transport subprocess were chunked
1205 and confused the parent.
1207 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
1208 which could crash as a result. This could lead to undeliverable messages.
1210 JH/28 Logging: "next input sent too soon" now shows where input was truncated
1213 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
1214 matters on fast-turnover and PID-randomising systems, which were getting
1215 out-of-order delivery.
1217 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
1218 a possibly-overlapping copy. The symptom was that "Remote host closed
1219 connection in response to HELO" was logged instead of the actual 4xx
1222 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
1223 Previously only that bufferd was discarded, resulting in SYMTP command
1226 JH/32 DKIM: when a message has multiple signatures matching an identity given
1227 in dkim_verify_signers, run the dkim acl once for each. Previously only
1228 one run was done. Bug 2189.
1230 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
1231 "panic the current process" to "deliberately defer". The panic log is
1232 still written with the problem list name; the mail and reject logs now
1233 get a temp-reject line for the message that was being handled, saying
1234 something like "domains check lookup or other defer". The SMTP 451
1235 message is still "Temporary local problem".
1237 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
1238 A crafted sequence of BDAT commands could result in in-use memory beeing
1239 freed. CVE-2017-16943.
1241 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
1242 from SMTP input. Previously it was always done; now only done for DATA
1243 and not BDAT commands. CVE-2017-16944.
1245 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
1246 to the message (such as an overlong header line). Previously this was
1247 not done and we did not exit BDAT mode. Followon from the previous item
1248 though a different problem.
1254 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
1255 than -2003 did; needs libidn2 in addition to libidn.
1257 JH/02 The path option on a pipe transport is now expanded before use.
1259 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
1260 Patch provided by "Björn", documentation fix added too.
1262 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
1263 missing a wire-to-host endian conversion.
1265 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
1266 close after a BDAT command line could be taken as a following command,
1267 giving a synch failure. Fix by only checking for synch immediately
1268 before acknowledging the chunk.
1270 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
1271 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
1272 macro. Patches provided by Josh Soref.
1274 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
1275 Previously we did not; the RFC seems ambiguous and VRFY is not listed
1276 by IANA as a service extension. However, John Klensin suggests that we
1279 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
1280 the dkim code may be unix-mode line endings rather than smtp wire-format
1281 CRLF, so prepend a CR to any bare LF.
1283 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
1284 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
1286 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
1287 the first were themselves being wrongly included in the feed into dkim
1288 processing; with most chunk sizes in use this resulted in an incorrect
1289 body hash calculated value.
1291 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
1292 DKIM signature block, for verification. Although advised against by
1293 standards it is specifically not ruled illegal.
1295 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
1297 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
1298 missing a body hash (the bh= tag).
1300 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
1301 It seems that HAProxy sends the Proxy Protocol information in clear and
1302 only then does a TLS startup, so do the same.
1304 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
1305 TCP connections (such as for Spamd) unless the daemon successfully set
1306 Fast Open mode on its listening sockets. This fixes breakage seen on
1307 too-old kernels or those not configured for Fast Open, at the cost of
1308 requiring both directions being enabled for TFO, and TFO never being used
1309 by non-daemon-related Exim processes.
1311 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
1312 endings, at least on the first header line. Try to canonify any that get
1313 past that check, despite the cost.
1315 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
1316 now limited to an arbitrary five deep, while parsing addresses with the
1317 strip_excess_angle_brackets option enabled.
1319 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
1320 instead leave the unprompted TLS handshake in socket buffer for the
1321 TLS library to consume.
1323 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
1325 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
1327 JH/16 Drop variables when they go out of scope. Memory management drops a whole
1328 region in one operation, for speed, and this leaves assigned pointers
1329 dangling. Add checks run only under the testsuite which checks all
1330 variables at a store-reset and panics on a dangling pointer; add code
1331 explicitly nulling out all the variables discovered. Fixes one known
1332 bug: a transport crash, where a dangling pointer for $sending_ip_address
1333 originally assigned in a verify callout, is re-used.
1335 PP/06 Drop '.' from @INC in various Perl scripts.
1337 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
1339 PP/08 Reduce a number of compilation warnings under clang; building with
1340 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
1341 should be warning-free.
1343 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
1345 HS/01 Fix portability problems introduced by PP/08 for platforms where
1346 realloc(NULL) is not equivalent to malloc() [SunOS et al].
1348 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
1349 chunk. This allows us to accept broken chunked messages. We need a more
1350 general solution here.
1352 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
1353 already-broken messages in the queue.
1355 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
1357 JH/19 Fix reference counting bug in routing-generated-address tracking.
1363 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
1364 supports it and a size is available (ie. the sending peer gave us one).
1366 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
1367 years of being deprecated). The replacements are the ACLs
1368 acl_smtp_mime and acl_not_smtp_mime.
1370 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
1371 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
1372 or even an in-clear connection were permitted. Now, if the host lookup
1373 was dnssec and dane was requested then the host is only used if the
1374 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
1375 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
1376 if one fails this test.
1377 This means that a poorly-configured remote DNS will make it incommunicado;
1378 but it protects against a DNS-interception attack on it.
1380 JH/04 Bug 1810: make continued-use of an open smtp transport connection
1381 non-noisy when a race steals the message being considered.
1383 JH/05 If main configuration option tls_certificate is unset, generate a
1384 self-signed certificate for inbound TLS connections.
1386 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
1387 in rewrites and routers.
1389 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
1390 and logged a warning sing 4.83; now they are a configuration file error.
1392 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
1393 (lacking @domain). Apply the same qualification processing as RCPT.
1395 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
1397 JH/10 Support ${sha256:} applied to a string (as well as the previous
1400 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
1401 a cutthrough deliver is pending, as we always want to make a connection.
1402 This also avoids re-routing the message when later placing the cutthrough
1403 connection after a verify cache hit.
1404 Do not update it with the verify result either.
1406 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
1407 when routing results in more than one destination address.
1409 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
1410 signing (which inhibits the cutthrough capability). Previously only
1411 the presence of an option was tested; now an expansion evaluating as
1412 empty is permissible (obviously it should depend only on data available
1413 when the cutthrough connection is made).
1415 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
1416 the relevant preceding SMTP command did not note the pipelining mode.
1418 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
1419 Previously they were not counted.
1421 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
1422 as one having no matching records. Previously we deferred the message
1423 that needed the lookup.
1425 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
1426 distinguished as "(=".
1428 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
1429 for missing MX records. Previously it only worked for missing A records.
1431 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
1433 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
1434 after the data-go-ahead and data-ack. Patch from Jason Betts.
1436 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
1437 even for a "none" policy. Patch from Tony Meyer.
1439 JH/22 Fix continued use of a connection for further deliveries. If a port was
1440 specified by a router, it must also match for the delivery to be
1443 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
1444 When one of the recipients of a message was unsuitable for the connection
1445 (has no matching addresses), we lost track of needing to mark it
1446 deferred. As a result mail would be lost.
1448 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
1450 JH/25 Decoding ACL controls is now done using a binary search; the source code
1451 takes up less space and should be simpler to maintain. Merge the ACL
1452 condition decode tables also, with similar effect.
1454 JH/26 Fix problem with one_time used on a redirect router which returned the
1455 parent address unchanged. A retry would see the parent address marked as
1456 delivered, so not attempt the (identical) child. As a result mail would
1459 JH/27 Fix a possible security hole, wherein a process operating with the Exim
1460 UID can gain a root shell. Credit to http://www.halfdog.net/ for
1461 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
1464 JH/28 Enable {spool,log} filesystem space and inode checks as default.
1465 Main config options check_{log,spool}_{inodes,space} are now
1466 100 inodes, 10MB unless set otherwise in the configuration.
1468 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
1469 Previously it only applied to the main-section connection policy
1472 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
1474 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
1475 by me. Added RFC7919 DH primes as an alternative.
1477 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
1478 are not in the system include path.
1480 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
1481 GnuTLS, when a session startup failed (eg because the client disconnected)
1482 Exim did stdio operations after fclose. This was exposed by a recent
1483 change which nulled out the file handle after the fclose.
1485 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
1486 signed directly by the cert-signing cert, rather than an intermediate
1487 OCSP-signing cert. This is the model used by LetsEncrypt.
1489 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
1491 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
1492 an incoming connection.
1494 HS/02 Bug 1802: Do not half-close the connection after sending a request
1497 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
1498 fallback to "prime256v1".
1500 JH/34 SECURITY: Use proper copy of DATA command in error message.
1501 Could leak key material. Remotely exploitable. CVE-2016-9963.
1507 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
1508 and 3.4.4 - once the server is enabled to respond to an OCSP request
1509 it does even when not requested, resulting in a stapling non-aware
1510 client dropping the TLS connection.
1512 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
1513 support variable-length bit vectors. No functional change.
1515 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
1516 The I= interface field on outgoing lines is now after the H= remote
1517 host field, same as incoming lines. There is a separate
1518 outgoing_interface log selector which allows you to disable the
1521 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
1522 If not running log_selector +smtp_connection the mainlog would be held
1523 open indefinitely after a "too many connections" event, including to a
1524 deleted file after a log rotate. Leave the per net connection logging
1525 leaving it open for efficiency as that will be quickly detected by the
1526 check on the next write.
1528 HS/01 Bug 1671: Fix post transport crash.
1529 Processing the wait-<transport> messages could crash the delivery
1530 process if the message IDs didn't exist for some reason. When
1531 using 'split_spool_directory=yes' the construction of the spool
1532 file name failed already, exposing the same netto behaviour.
1534 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
1535 mime_regex ACL conditions.
1537 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
1538 to DSN fail messages (bounces): remote IP, remote greeting, remote response
1539 to HELO, local diagnostic string.
1541 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
1542 log line to debug. Even when configured with a tls authenticator many
1543 client connections are expected to not authenticate in this way, so
1544 an authenticate fail is not an error.
1546 HS/02 Add the Exim version string to the process info. This way exiwhat
1547 gives some more detail about the running daemon.
1549 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
1550 matter for fast-change records such as DNSBLs.
1552 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
1553 retry record, even if constant. There may be multiple transports with
1554 different interface settings and the retry behaviour needs to be kept
1557 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
1559 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
1561 JH/10 Bug 840: fix log_defer_output option of pipe transport
1563 JH/11 Bug 830: use same host for all RCPTS of a message, even under
1564 hosts_randomize. This matters a lot when combined with mua_wrapper.
1566 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
1567 ${quote_pgsql:<string>} operator.
1569 JH/13 Bug 1708: avoid misaligned access in cached lookup.
1571 JH/14 Change header file name for freeradius-client. Relevant if compiling
1572 with Radius support; from the Gentoo tree and checked under Fedora.
1574 JH/15 Bug 1712: Introduce $prdr_requested flag variable
1576 JH/16 Bug 1714: Permit an empty string as expansion result for transport
1577 option transport_filter, meaning no filtering.
1579 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
1581 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
1582 defaults to "*" (all hosts). The variable is now available when not built
1583 with TLS, default unset, mainly to enable keeping the testsuite sane.
1584 If a server certificate is not supplied (via tls_certificate) an error is
1585 logged, and clients will find TLS connections fail on startup. Presumably
1586 they will retry in-clear.
1587 Packagers of Exim are strongly encouraged to create a server certificate
1588 at installation time.
1590 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
1591 with the $config_file variable.
1593 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
1594 in transport context, after the attempt, and per-recipient. The latter type
1595 is per host attempted. The event data is the error message, and the errno
1596 information encodes the lookup type (A vs. MX) used for the (first) host,
1597 and the trailing two digits of the smtp 4xx response.
1599 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
1600 to write to mainlog (or rejectlog, paniclog) in the window between file
1601 creation and permissions/ownership being changed. Particularly affects
1602 installations where exicyclog is run as root, rather than exim user;
1603 result is that the running daemon panics and dies.
1605 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
1607 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
1608 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
1609 "pri" and "weight". Note that the previous implicit priority given by the
1610 list order is no longer honoured.
1612 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
1613 for DKIM processing.
1615 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
1616 by defining SUPPORT_SOCKS.
1618 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
1619 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
1620 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
1621 variables are renamed to proxy_{local,external}_{address,port}.
1623 JH/27 Move Internationalisation support from Experimental to mainline, enabled
1624 for a build by defining SUPPORT_I18N
1626 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
1627 of the query string, and make ${quote_redis:} do that quoting.
1629 JH/29 Move Events support from Experimental to mainline, enabled by default
1630 and removable for a build by defining DISABLE_EVENT.
1632 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
1634 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
1635 cached by the daemon.
1637 JH/32 Move Redis support from Experimental to mainline, enabled for a build
1638 by defining LOOKUP_REDIS. The libhiredis library is required.
1640 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
1641 keys are given for lookup.
1643 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
1644 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
1645 only supported when built with TLS support. The PolarSSL SHA routines
1646 are still used when the TLS library is too old for convenient support.
1648 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
1649 openssl_options), for security. OpenSSL forces this from version 1.1.0
1650 server-side so match that on older versions.
1652 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
1653 allocation for $value could be released as the expansion processing
1654 concluded, but leaving the global pointer active for it.
1656 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
1657 and to use the domains and local_parts ACL conditions.
1659 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
1660 incorrectly not doubled on cutthrough transmission, hence seen as a
1661 body-termination at the receiving system - resulting in truncated mails.
1662 Commonly the sender saw a TCP-level error, and retransmitted the message
1663 via the normal store-and-forward channel. This could result in duplicates
1664 received - but deduplicating mailstores were liable to retain only the
1665 initial truncated version.
1667 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
1669 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
1671 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
1672 we're in there, support oversigning also; bug 1309.
1674 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
1676 HS/04 Add support for keep_environment and add_environment options.
1678 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
1679 either intentional arithmetic overflow during PRNG, or testing config-
1682 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
1683 delivery resulted in actual delivery. Cancel cutthrough before DATA
1686 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
1687 rejects a recipient: pass the reject to the originator.
1689 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
1690 Many were false-positives and ignorable, but it's worth fixing the
1693 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
1694 for the new environment-manipulation done at startup. Move the routines
1695 from being local to tls.c to being global via the os.c file.
1697 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
1698 an extract embedded as result-arg for a map, the first arg for extract
1699 is unavailable so we cannot tell if this is a numbered or keyed
1700 extraction. Accept either.
1706 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
1709 JH/02 The smtp transport option "multi_domain" is now expanded.
1711 JH/03 The smtp transport now requests PRDR by default, if the server offers
1714 JH/04 Certificate name checking on server certificates, when exim is a client,
1715 is now done by default. The transport option tls_verify_cert_hostnames
1716 can be used to disable this per-host. The build option
1717 EXPERIMENTAL_CERTNAMES is withdrawn.
1719 JH/05 The value of the tls_verify_certificates smtp transport and main options
1720 default to the word "system" to access the system default CA bundle.
1721 For GnuTLS, only version 3.0.20 or later.
1723 JH/06 Verification of the server certificate for a TLS connection is now tried
1724 (but not required) by default. The verification status is now logged by
1725 default, for both outbound TLS and client-certificate supplying inbound
1728 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
1731 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
1732 Status Notification (bounce) messages are now MIME format per RFC 3464.
1733 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
1734 under the control of the dsn_advertise_hosts option, and routers may
1735 have a dsn_lasthop option.
1737 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
1738 default, modifiable by a malware= option. The list separator for
1739 the options can now be changed in the usual way. Bug 68.
1741 JH/10 The smtp_receive_timeout main option is now expanded before use.
1743 JH/11 The incoming_interface log option now also enables logging of the
1744 local interface on delivery outgoing connections.
1746 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
1747 if the interface and destination host and port all match.
1749 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
1752 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
1753 Patch from Andrew Lewis.
1755 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
1756 now supports optional time-restrictions, weighting, and priority
1757 modifiers per server. Patch originally by <rommer@active.by>.
1759 JH/16 The spamd_address main option now supports a mixed list of local
1760 and remote servers. Remote servers can be IPv6 addresses, and
1761 specify a port-range.
1763 JH/17 Bug 68: The spamd_address main option now supports an optional
1764 timeout value per server.
1766 JH/18 Bug 1581: Router and transport options headers_add/remove can
1767 now have the list separator specified.
1769 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
1772 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
1775 JH/21 Support for the A6 type of dns record is withdrawn.
1777 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
1778 rather than the verbs used.
1780 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
1781 from 255 to 1024 chars.
1783 JH/24 Verification callouts now attempt to use TLS by default.
1785 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
1786 are generic router options now. The defaults didn't change.
1788 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
1789 Original patch from Alexander Shikoff, worked over by JH.
1791 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
1792 files if ZCAT_COMMAND is not executable.
1794 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
1796 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
1798 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
1799 Normally benign, it bites when the pair was led to by a CNAME;
1800 modern usage is to not canonicalize the domain to a CNAME target
1801 (and we were inconsistent anyway for A-only vs AAAA+A).
1803 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
1805 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
1806 when evaluating $sender_host_dnssec.
1808 JH/31 Check the HELO verification lookup for DNSSEC, adding new
1809 $sender_helo_dnssec variable.
1811 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
1813 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
1815 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
1817 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
1818 documented as working, but never had. Support all but $spam_report.
1820 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
1821 added for tls authenticator.
1823 HS/03 Add perl_taintmode main config option
1829 TL/01 When running the test suite, the README says that variables such as
1830 no_msglog_check are global and can be placed anywhere in a specific
1831 test's script, however it was observed that placement needed to be near
1832 the beginning for it to behave that way. Changed the runtest perl
1833 script to read through the entire script once to detect and set these
1834 variables, reset to the beginning of the script, and then run through
1835 the script parsing/test process like normal.
1837 TL/02 The BSD's have an arc4random API. One of the functions to induce
1838 adding randomness was arc4random_stir(), but it has been removed in
1839 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
1840 function when detected.
1842 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
1843 cause callback expansion.
1845 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
1846 syntax errors in an expansion can be treated as a string instead of
1847 logging or causing an error, due to the internal use of bool_lax
1848 instead of bool when processing it.
1850 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
1851 server certificates when making smtp deliveries.
1853 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
1855 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
1857 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
1859 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
1860 Merged patch from Sebastian Wiedenroth.
1862 JH/05 Fix results-pipe from transport process. Several recipients, combined
1863 with certificate use, exposed issues where response data items split
1864 over buffer boundaries were not parsed properly. This eventually
1865 resulted in duplicates being sent. This issue only became common enough
1866 to notice due to the introduction of connection certificate information,
1867 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1869 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1870 size buffer was used, resulting in syntax errors when an expansion
1873 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1874 version 3.3.6 or later.
1876 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1877 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1878 both become "event_action", the variables become $event_name, $event_data
1879 and $event_defer_errno. There is a new variable $verify_mode, usable in
1880 routers, transports and related events. The tls:cert event is now also
1881 raised for inbound connections, if the main configuration event_action
1884 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1885 early OCSP support, but no stapling (appears to be less than 1.0.0).
1887 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1888 server certificate names available under the smtp transport option
1889 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1892 JH/10 Time-related extraction expansions from certificates now use the main
1893 option "timezone" setting for output formatting, and are consistent
1894 between OpenSSL and GnuTLS compilations. Bug 1541.
1896 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1897 encoded parameter in the incoming message. Bug 1558.
1899 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1900 include certificate info, eximon was claiming there were spoolfile
1903 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1905 JH/14 Log delivery-related information more consistently, using the sequence
1906 "H=<name> [<ip>]" wherever possible.
1908 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1909 are problematic for Debian distribution, omit them from the release
1912 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1914 JH/16 Fix string representation of time values on 64bit time_t architectures.
1917 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1918 output list separator was used.
1923 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1924 checkers that were complaining about end of non-void function with no
1927 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1928 This was a regression introduced in 4.83 by another bugfix.
1930 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1932 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1933 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1939 TF/01 Correctly close the server side of TLS when forking for delivery.
1941 When a message was received over SMTP with TLS, Exim failed to clear up
1942 the incoming connection properly after forking off the child process to
1943 deliver the message. In some situations the subsequent outgoing
1944 delivery connection happened to have the same fd number as the incoming
1945 connection previously had. Exim would try to use TLS and fail, logging
1946 a "Bad file descriptor" error.
1948 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1949 utilities have not been installed.
1951 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1952 temporary space as the ACL may create new global variables.
1954 TL/01 LDAP support uses per connection or global context settings, depending
1955 upon the detected version of the libraries at build time.
1957 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1958 to extract and use the src ip:port in logging and expansions as if it
1959 were a direct connection from the outside internet. PPv2 support was
1960 updated based on HAProxy spec change in May 2014.
1962 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1964 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1965 Properly escape header and check for NULL return.
1967 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1970 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1972 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1973 characters in header names, implemented as a verify condition.
1974 Contributed by Michael Fischer v. Mollard.
1976 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1977 results permerror and temperror. Previous values are deprecated but
1978 still accepted. In a future release, err_perm and err_temp will be
1979 completely removed, which will be a backward incompatibility if the
1980 ACL tests for either of these two old results. Patch contributed by
1981 user bes-internal on the mailing list.
1983 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1985 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1986 selectors, in both main and reject logs.
1988 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1991 JH/07 Add malware type "sock" for talking to simple daemon.
1993 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1995 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1996 routers/transports under cutthrough routing.
1998 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1999 numbers. Touch up "bool" conditional to keep the same definition.
2001 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
2003 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
2005 JH/12 Expand items in router/transport headers_add or headers_remove lists
2006 individually rather than the list as a whole. Bug 1452.
2008 Required for reasonable handling of multiple headers_ options when
2009 they may be empty; requires that headers_remove items with embedded
2010 colons must have them doubled (or the list-separator changed).
2012 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
2013 view the policy declared in the DMARC record. Currently, $dmarc_status
2014 is a combined value of both the record presence and the result of the
2017 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
2019 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
2020 dnslookup router and the smtp transport (applying to the forward
2023 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
2024 of ldap servers used for a specific lookup. Patch provided by Heiko
2027 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
2028 New variable $lookup_dnssec_authenticated for observability.
2030 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
2031 Patch submitted by Lars Timman.
2033 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
2035 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
2036 Requires trusted mode and valid format message id, aborts otherwise.
2037 Patch contributed by Heiko Schlichting.
2039 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
2040 certextract with support for various fields. Bug 1358.
2042 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
2043 is requested by default, modifiable by smtp transport option
2046 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
2047 operate on certificate variables to give certificate fingerprints
2048 Also new ${sha256:cert_variable}.
2050 JH/23 The PRDR feature is moved from being Experimental into the mainline.
2052 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
2053 Christian Aistleitner.
2055 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
2057 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
2058 file. Patch from Wolfgang Breyha.
2060 JH/25 Expand the coverage of the delivery $host and $host_address to
2061 client authenticators run in verify callout. Bug 1476.
2063 JH/26 Port service names are now accepted for tls_on_connect_ports, to
2064 align with daemon_smtp_ports. Bug 72.
2066 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
2067 support and error reporting did not work properly.
2069 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
2070 and is readable. Patch from Andrew Colin Kissa.
2072 TL/14 Enhance documentation of ${run expansion and how it parses the
2073 commandline after expansion, particularly in the case when an
2074 unquoted variable expansion results in an empty value.
2076 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
2078 PP/02 Fix internal collision of T_APL on systems which support RFC3123
2079 by renaming away from it. Addresses GH issue 15, reported by
2082 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
2084 TL/15 SECURITY: prevent double expansion in math comparison functions
2085 (can expand unsanitized data). Not remotely exploitable.
2092 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
2094 PP/02 Make -n do something, by making it not do something.
2095 When combined with -bP, the name of an option is not output.
2097 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
2100 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
2101 $sender_host_name and config options to manage this, and basic check
2104 PP/05 DSCP support for outbound connections and control modifier for inbound.
2106 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
2107 (Only plugin which currently uses this is kerberos4, which nobody should
2108 be using, but we should make it available and other future plugins might
2109 conceivably use it, even though it would break NAT; stuff *should* be
2110 using channel bindings instead).
2112 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
2113 name; added for Sendmail compatibility; requires admin caller.
2114 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
2115 just ignore it); requires trusted caller.
2116 Also parse but ignore: -Ac -Am -X<logfile>
2119 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
2121 TL/02 Add +smtp_confirmation as a default logging option.
2123 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
2124 Patch by Magnus Holmgren from 2007-02-20.
2126 TL/04 Bugzilla 1281 - Spec typo.
2127 Bugzilla 1283 - Spec typo.
2128 Bugzilla 1290 - Spec grammar fixes.
2130 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
2132 TL/06 Add Experimental DMARC support using libopendmarc libraries.
2134 TL/07 Fix an out of order global option causing a segfault. Reported to dev
2135 mailing list by by Dmitry Isaikin.
2137 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
2139 JH/02 Support "G" suffix to numbers in ${if comparisons.
2141 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
2143 NM/01 Bugzilla 1197 - Spec typo
2144 Bugzilla 1196 - Spec examples corrections
2146 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
2148 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
2149 gnutls_enable_pkcs11, but renamed to more accurately indicate its
2152 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
2153 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
2155 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
2156 "acl {{name}{arg}...}", and optional args on acl condition
2159 JH/05 Permit multiple router/transport headers_add/remove lines.
2161 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
2163 JH/07 Avoid using a waiting database for a single-message-only transport.
2164 Performance patch from Paul Fisher. Bugzilla 1262.
2166 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
2169 JH/09 Add $headers_added variable, with content from use of ACL modifier
2170 add_header (but not yet added to the message). Bugzilla 199.
2172 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
2173 Pulled from Bugzilla 817 by Wolfgang Breyha.
2175 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
2177 (nb: this is the same fix as in Exim 4.80.1)
2179 JH/11 Add A= logging on delivery lines, and a client_set_id option on
2182 JH/12 Add optional authenticated_sender logging to A= and a log_selector
2185 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
2187 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
2188 advertise SMTP AUTH mechanism to us, instead of a generic
2189 protocol violation error. Also, make Exim more robust to bad
2190 data from the Dovecot auth socket.
2192 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
2194 When a queue runner is handling a message, Exim first routes the
2195 recipient addresses, during which it prunes them based on the retry
2196 hints database. After that it attempts to deliver the message to
2197 any remaining recipients. It then updates the hints database using
2200 So if a recipient address works intermittently, it can get repeatedly
2201 deferred at routing time. The retry hints record remains fresh so the
2202 address never reaches the final cutoff time.
2204 This is a fairly common occurrence when a user is bumping up against
2205 their storage quota. Exim had some logic in its local delivery code
2206 to deal with this. However it did not apply to per-recipient defers
2207 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
2209 This change adds a proper retry rule check during routing so that the
2210 final cutoff time is checked against the message's age. We only do
2211 this check if there is an address retry record and there is not a
2212 domain retry record; this implies that previous attempts to handle
2213 the address had the retry_use_local_parts option turned on. We use
2214 this as an approximation for the destination being like a local
2215 delivery, as in LMTP.
2217 I suspect this new check makes the old local delivery cutoff check
2218 redundant, but I have not verified this so I left the code in place.
2220 TF/02 Correct gecos expansion when From: is a prefix of the username.
2222 Test 0254 submits a message to Exim with the header
2226 When I ran the test suite under the user fanf2, Exim expanded
2227 the header to contain my full name, whereas it should have added
2228 a Resent-Sender: header. It erroneously treats any prefix of the
2229 username as equal to the username.
2231 This change corrects that bug.
2233 GF/01 DCC debug and logging tidyup
2234 Error conditions log to paniclog rather than rejectlog.
2235 Debug lines prefixed by "DCC: " to remove any ambiguity.
2237 TF/03 Avoid unnecessary rebuilds of lookup-related code.
2239 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
2240 Bug spotted by Jeremy Harris; was flawed since initial commit.
2241 Would have resulted in OCSP responses post-SNI triggering an Exim
2242 NULL dereference and crash.
2244 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
2246 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
2247 Bug detection, analysis and fix by Samuel Thibault.
2248 Bugzilla 1331, Debian bug #698092.
2250 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
2252 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
2253 Server implementation by Todd Lyons, client by JH.
2254 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
2255 config variable "prdr_enable" controls whether the server
2256 advertises the facility. If the client requests PRDR a new
2257 acl_data_smtp_prdr ACL is called once for each recipient, after
2258 the body content is received and before the acl_smtp_data ACL.
2259 The client is controlled by both of: a hosts_try_prdr option
2260 on the smtp transport, and the server advertisement.
2261 Default client logging of deliveries and rejections involving
2262 PRDR are flagged with the string "PRDR".
2264 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
2265 fclose(). Diagnosis by Todd Lyons.
2267 PP/17 Update configure.default to handle IPv6 localhost better.
2268 Patch by Alain Williams (plus minor tweaks).
2271 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
2272 This is now consistent with GnuTLS, and is now documented: the
2273 previous undocumented portable approach to treating the option as
2274 unset was to force an expansion failure. That still works, and
2275 an empty string is now equivalent.
2277 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
2278 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
2279 not performing validation itself.
2281 PP/20 Added force_command boolean option to pipe transport.
2282 Patch from Nick Koston, of cPanel Inc.
2284 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
2287 TF/04 Added udpsend ACL modifier and hexquote expansion operator
2289 PP/21 Fix eximon continuous updating with timestamped log-files.
2290 Broken in a format-string cleanup in 4.80, missed when I repaired the
2291 other false fix of the same issue.
2292 Report and fix from Heiko Schlichting.
2295 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
2296 Report from Prashanth Katuri.
2298 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
2299 It's SecureTransport, so affects any MacOS clients which use the
2300 system-integrated TLS libraries, including email clients.
2302 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
2303 using a MIME ACL for non-SMTP local injection.
2304 Report and assistance in diagnosis by Warren Baker.
2306 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
2308 JH/16 Fix comparisons for 64b. Bugzilla 1385.
2310 TL/09 Add expansion variable $authenticated_fail_id to keep track of
2311 last id that failed so it may be referenced in subsequent ACL's.
2313 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
2316 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
2317 ldap library initialization, allowing self-signed CA's to be
2318 used. Also properly sets require_cert option later in code by
2319 using NULL (global ldap config) instead of ldap handle (per
2320 session). Bug diagnosis and testing by alxgomz.
2322 TL/12 Enhanced documentation in the ratelimit.pl script provided in
2323 the src/util/ subdirectory.
2325 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
2326 renamed to Transport Post Delivery Action by Jeremy Harris, as
2329 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
2330 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
2331 redis_servers = needs to be configured which will be used by the redis
2332 lookup. Patch from Warren Baker, of The Packet Hub.
2334 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
2336 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
2337 hostname or reverse DNS when processing a host list. Used suggestions
2338 from multiple comments on this bug.
2340 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
2342 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
2343 Missed a few lines, added it to make the runtest require no keyboard
2346 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
2347 contains upper case chars. Make router use caseful_local_part.
2349 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
2350 support when GnuTLS has been built with p11-kit.
2356 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
2358 This, or similar/improved, will also be change PP/11 of 4.82.
2364 PP/01 Handle short writes when writing local log-files.
2365 In practice, only affects FreeBSD (8 onwards).
2366 Bugzilla 1053, with thanks to Dmitry Isaikin.
2368 NM/01 Bugzilla 949 - Documentation tweak
2370 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
2373 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
2375 PP/02 Implemented gsasl authenticator.
2377 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
2379 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
2380 `pkg-config foo` for cflags/libs.
2382 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
2383 with rest of GSASL and with heimdal_gssapi.
2385 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
2386 `pkg-config foo` for cflags/libs for the TLS implementation.
2388 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
2389 properties get this fed in as external SSF. A number of robustness
2390 and debugging improvements to the cyrus_sasl authenticator.
2392 PP/08 cyrus_sasl server now expands the server_realm option.
2394 PP/09 Bugzilla 1214 - Log authentication information in reject log.
2395 Patch by Jeremy Harris.
2397 PP/10 Added dbmjz lookup type.
2399 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
2401 PP/12 MAIL args handles TAB as well as SP, for better interop with
2402 non-compliant senders.
2403 Analysis and variant patch by Todd Lyons.
2405 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
2406 Bug report from Lars Müller <lars@samba.org> (via SUSE),
2407 Patch from Dirk Mueller <dmueller@suse.com>
2409 PP/13 tls_peerdn now print-escaped for spool files.
2410 Observed some $tls_peerdn in wild which contained \n, which resulted
2411 in spool file corruption.
2413 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
2414 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
2415 or write after TLS renegotiation, which otherwise led to messages
2418 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
2419 as a tracking header (ie: a signed header comes before the signature).
2420 Patch from Wolfgang Breyha.
2422 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
2423 comma-sep list; embedded commas doubled.
2425 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
2427 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
2429 Report and patch from Dmitry Banschikov.
2431 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
2432 Removed SSL_clear() after SSL_new() which led to protocol negotiation
2433 failures. We appear to now support TLS1.1+ with Exim.
2435 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
2436 lets Exim select keys and certificates based upon TLS SNI from client.
2437 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
2438 before an outbound SMTP session. New log_selector, +tls_sni.
2440 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
2441 NULL dereference. Report and patch from Alun Jones.
2443 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
2444 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
2445 Not seeing resolver debug output on NetBSD, but suspect this is a
2446 resolver implementation change.
2448 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
2449 Left warnings. Added "eximon gdb" invocation mode.
2451 PP/21 Defaulting "accept_8bitmime" to true, not false.
2453 PP/22 Added -bw for inetd wait mode support.
2455 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
2456 locate the relevant includes and libraries. Made this the default.
2458 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
2459 Bugzilla 1246, report and most of solution from Tomasz Kusy.
2461 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
2462 This may cause build issues on older platforms.
2464 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
2465 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
2466 gnutls_require_mac & gnutls_require_protocols (no longer supported).
2467 Added SNI support via GnuTLS too.
2468 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
2470 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
2472 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
2473 Applied second patch from Janne, implementing suggestion to default
2474 multiple-strings-in-record handling to match SPF spec.
2476 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
2478 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
2479 read-only, out of scope).
2480 Patch from Wolfgang Breyha, report from Stuart Northfield.
2482 PP/29 Fix three issues highlighted by clang analyser static analysis.
2483 Only crash-plausible issue would require the Cambridge-specific
2484 iplookup router and a misconfiguration.
2485 Report from Marcin Mirosław.
2487 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
2489 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
2490 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
2491 As part of this, removing so much warning spew let me fix some minor
2492 real issues in debug logging.
2494 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
2495 assignment on my part. Fixed.
2497 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
2498 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
2499 Janne Snabb (who went above and beyond: thank you).
2501 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
2502 string otherwise requires a connection and a bunch more work and it's
2503 relatively easy to get wrong. Should also expose TLS library linkage
2506 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
2507 64-bit ${eval} (JH/03).
2509 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
2510 GNU libc to support some of the 64-bit stuff, should not lead to
2511 conflicts. Defined before os.h is pulled in, so if a given platform
2512 needs to override this, it can.
2514 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
2515 protection layer was required, which is not implemented.
2516 Bugzilla 1254, patch from Wolfgang Breyha.
2518 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
2519 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
2520 tls_dhparam take prime identifiers. Also unbreak combination of
2521 OpenSSL+DH_params+TLSSNI.
2523 PP/39 Disable SSLv2 by default in OpenSSL support.
2529 PP/01 Solaris build fix for Oracle's LDAP libraries.
2530 Bugzilla 1109, patch from Stephen Usher.
2532 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
2534 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
2537 TF/02 Fix a couple more cases where we did not log the error message
2538 when unlink() failed. See also change 4.74-TF/03.
2540 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
2541 lock up or crash if it happened to be inside a call to libc when it
2542 got a SIGUSR1 from exiwhat.
2544 The SIGUSR1 handler appends the current process status to the process
2545 log which is later printed by exiwhat. It used to use the general
2546 purpose logging code to do this, but several functions it calls are
2547 not safe for signals.
2549 The new output code in the SIGUSR1 handler is specific to the process
2550 log, and simple enough that it's easy to inspect for signal safety.
2551 Removing some special cases also simplifies the general logging code.
2552 Removing the spurious timestamps from the process log simplifies
2555 TF/04 Improved ratelimit ACL condition.
2557 The /noupdate option has been deprecated in favour of /readonly which
2558 has clearer semantics. The /leaky, /strict, and /readonly update modes
2559 are mutually exclusive. The update mode is no longer included in the
2560 database key; it just determines when the database is updated. (This
2561 means that when you upgrade Exim will forget old rate measurements.)
2563 Exim now checks that the per_* options are used with an update mode that
2564 makes sense for the current ACL. For example, when Exim is processing a
2565 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
2566 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
2567 must specify per_mail/readonly. If you omit the update mode it defaults to
2568 /leaky where that makes sense (as before) or /readonly where required.
2570 The /noupdate option is now undocumented but still supported for
2571 backwards compatibility. It is equivalent to /readonly except that in
2572 ACLs where /readonly is required you may specify /leaky/noupdate or
2573 /strict/noupdate which are treated the same as /readonly.
2575 A useful new feature is the /count= option. This is a generalization
2576 of the per_byte option, so that you can measure the throughput of other
2577 aggregate values. For example, the per_byte option is now equivalent
2578 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
2580 The per_rcpt option has been generalized using the /count= mechanism
2581 (though it's more complicated than the per_byte equivalence). When it is
2582 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
2583 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
2584 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
2585 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
2586 Note that using per_rcpt with a non-readonly update mode in more than
2587 one ACL will cause the recipients to be double-counted. (The per_mail
2588 and per_byte options don't have this problem.)
2590 The handling of very low rates has changed slightly. If the computed rate
2591 is less than the event's count (usually one) then this event is the first
2592 after a long gap. In this case the rate is set to the same as this event's
2593 count, so that the first message of a spam run is counted properly.
2595 The major new feature is a mechanism for counting the rate of unique
2596 events. The new per_addr option counts the number of different
2597 recipients that someone has sent messages to in the last time period. It
2598 behaves like per_rcpt if all the recipient addresses are different, but
2599 duplicate recipient addresses do not increase the measured rate. Like
2600 the /count= option this is a general mechanism, so the per_addr option
2601 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
2602 example, measure the rate that a client uses different sender addresses
2603 with the options per_mail/unique=$sender_address. There are further
2604 details in the main documentation.
2606 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
2608 TF/06 Removed a few PCRE remnants.
2610 TF/07 Automatically extract Exim's version number from tags in the git
2611 repository when doing development or release builds.
2613 PP/02 Raise smtp_cmd_buffer_size to 16kB.
2614 Bugzilla 879. Patch from Paul Fisher.
2616 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
2617 Heavily based on revision 40f9a89a from Simon Arlott's tree.
2620 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
2622 PP/05 Variable $av_failed, true if the AV scanner deferred.
2623 Bugzilla 1078. Patch from John Horne.
2625 PP/06 Stop make process more reliably on build failure.
2626 Bugzilla 1087. Patch from Heiko Schlittermann.
2628 PP/07 Make maildir_use_size_file an _expandable_ boolean.
2629 Bugzilla 1089. Patch from Heiko Schlittermann.
2631 PP/08 Handle ${run} returning more data than OS pipe buffer size.
2632 Bugzilla 1131. Patch from Holger Weiß.
2634 PP/09 Handle IPv6 addresses with SPF.
2635 Bugzilla 860. Patch from Wolfgang Breyha.
2637 PP/10 GnuTLS: support TLS 1.2 & 1.1.
2639 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
2642 PP/11 match_* no longer expand right-hand-side by default.
2643 New compile-time build option, EXPAND_LISTMATCH_RHS.
2644 New expansion conditions, "inlist", "inlisti".
2646 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
2648 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
2650 PP/14 fix log_write() format string regression from TF/03.
2651 Bugzilla 1152. Patch from Dmitry Isaikin.
2657 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
2659 PP/02 Harmonised TLS library version reporting; only show if debugging.
2660 Layout now matches that introduced for other libraries in 4.74 PP/03.
2662 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
2664 PP/04 New "dns_use_edns0" global option.
2666 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
2669 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
2670 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
2672 TK/01 Updated PolarSSL code to 0.14.2.
2673 Bugzilla 1097. Patch from Andreas Metzler.
2675 PP/07 Catch divide-by-zero in ${eval:...}.
2676 Fixes bugzilla 1102.
2678 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
2681 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
2682 format-string attack -- SECURITY: remote arbitrary code execution.
2684 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
2685 time unintentionally subject to list matching rules, letting the header
2686 cause arbitrary Exim lookups (of items which can occur in lists, *not*
2687 arbitrary string expansion). This allowed for information disclosure.
2689 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
2690 INT_MIN/-1 -- value coerced to INT_MAX.
2696 NM/01 Workaround for PCRE version dependency in version reporting
2699 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
2700 This fixes portability to compilers other than gcc, notably
2701 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
2703 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
2704 makefiles for portability to HP-UX and POSIX correctness.
2706 PP/01 Permit LOOKUP_foo enabling on the make command-line.
2707 Also via indented variable definition in the Makefile.
2708 (Debugging by Oliver Heesakkers).
2710 PP/02 Restore caching of spamd results with expanded spamd_address.
2711 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
2713 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
2714 Improves build reliability. Fix from: Frank Elsner
2716 NM/02 Fix wide character breakage in the rfc2047 coding
2717 Fixes bug 1064. Patch from Andrey N. Oktyabrski
2719 NM/03 Allow underscore in dnslist lookups
2720 Fixes bug 1026. Patch from Graeme Fowler
2722 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
2723 Code patches from Adam Ciarcinski of NetBSD.
2725 NM/04 Fixed exiqgrep to cope with mailq missing size issue
2728 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
2729 is logged, to avoid truncation. Patch from John Horne.
2731 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
2732 Patch from Jakob Hirsch.
2734 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
2735 SQL string expansion failure details.
2736 Patch from Andrey Oktyabrski.
2738 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
2739 Patch from Simon Arlott.
2741 PP/09 New lookups functionality failed to compile on old gcc which rejects
2742 extern declarations in function scope.
2743 Patch from Oliver Fleischmann
2745 PP/10 Use sig_atomic_t for flags set from signal handlers.
2746 Check getgroups() return and improve debugging.
2747 Fixed developed for diagnosis in bug 927 (which turned out to be
2750 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
2751 Patch from Mark Zealey.
2753 PP/12 Bugzilla 1056: Improved spamd server selection.
2754 Patch from Mark Zealey.
2756 PP/13 Bugzilla 1086: Deal with maildir quota file races.
2757 Based on patch from Heiko Schlittermann.
2759 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
2760 Patch from Uwe Doering, sign-off by Michael Haardt.
2762 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
2763 variable declaration deep within a block. Bug and patch from
2766 PP/15 lookups-Makefile IRIX compatibility coercion.
2768 PP/16 Make DISABLE_DKIM build knob functional.
2770 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
2771 Patch by Simon Arlott
2773 TF/03 Fix valgrind.h portability to C89 compilers that do not support
2774 variable argument macros. Our copy now differs from upstream.
2780 TF/01 Failure to get a lock on a hints database can have serious
2781 consequences so log it to the panic log.
2783 TF/02 Log LMTP confirmation messages in the same way as SMTP,
2784 controlled using the smtp_confirmation log selector.
2786 TF/03 Include the error message when we fail to unlink a spool file.
2788 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
2789 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
2790 for maintaining out-of-tree patches for some time.
2792 PP/01 Bugzilla 139: Documentation and portability issues.
2793 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
2794 Handle per-OS dynamic-module compilation flags.
2796 PP/02 Let /dev/null have normal permissions.
2797 The 4.73 fixes were a little too stringent and complained about the
2798 permissions on /dev/null. Exempt it from some checks.
2799 Reported by Andreas M. Kirchwitz.
2801 PP/03 Report version information for many libraries, including
2802 Exim version information for dynamically loaded libraries. Created
2803 version.h, now support a version extension string for distributors
2804 who patch heavily. Dynamic module ABI change.
2806 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
2807 privilege escalation vulnerability whereby the Exim run-time user
2808 can cause root to append content of the attacker's choosing to
2811 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
2814 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
2815 If dropping privileges for untrusted macros, we disabled normal logging
2816 on the basis that it would fail; for the Exim run-time user, this is not
2817 the case, and it resulted in successful deliveries going unlogged.
2818 Fixed. Reported by Andreas Metzler.
2824 PP/01 Date: & Message-Id: revert to normally being appended to a message,
2825 only prepend for the Resent-* case. Fixes regression introduced in
2826 Exim 4.70 by NM/22 for Bugzilla 607.
2828 PP/02 Include check_rfc2047_length in configure.default because we're seeing
2829 increasing numbers of administrators be bitten by this.
2831 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
2833 PP/03 Bugzilla 994: added openssl_options main configuration option.
2835 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
2837 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
2839 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
2841 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
2842 without a peer certificate, leading to a segfault because of an
2843 assumption that peers always have certificates. Be a little more
2844 paranoid. Problem reported by Martin Tscholak.
2846 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
2847 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
2848 NB: ClamAV planning to remove STREAM in "middle of 2010".
2849 CL also introduces -bmalware, various -d+acl logging additions and
2850 more caution in buffer sizes.
2852 PP/09 Implemented reverse_ip expansion operator.
2854 PP/10 Bugzilla 937: provide a "debug" ACL control.
2856 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
2858 PP/12 Bugzilla 973: Implement --version.
2860 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2862 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2864 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2866 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2867 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2868 ignore trailing whitespace.
2870 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2872 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2875 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2876 Notification from Dr Andrew Aitchison.
2878 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2879 ExtendedDetectionInfo response format.
2880 Notification from John Horne.
2882 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2885 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2886 XSL and documented dependency on system catalogs, with examples of how
2889 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2892 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2893 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2894 configuration file which is writeable by the Exim user or group.
2896 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2897 of configuration files to cover files specified with the -C option if
2898 they are going to be used with root privileges, not just the default
2901 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2902 option (effectively making it always true).
2904 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2905 files to be used while preserving root privileges.
2907 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2908 that rogue child processes cannot use them.
2910 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2911 run-time user, instead of root.
2913 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2914 Exim run-time user without dropping privileges.
2916 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2917 result string, instead of calling string_vformat() twice with the same
2920 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2921 for other users. Others should always drop root privileges if they use
2922 -C on the command line, even for a whitelisted configure file.
2924 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2926 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2932 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2933 $data_path, and $header_path variables; fixed documentation bugs and
2936 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2937 exipick to access non-standard spools, including the "frozen" queue
2940 NM/01 Bugzilla 965: Support mysql stored procedures.
2941 Patch from Alain Williams
2943 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2945 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2946 Patch from Andreas Metzler
2948 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2949 Patch from Kirill Miazine
2951 NM/05 Bugzilla 671: Added umask to procmail example.
2953 JJ/03 installed exipick 20100323.0, fixing doc bug
2955 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2956 directory. Notification and patch from Dan Rosenberg.
2958 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2960 TK/02 Improve log output when DKIM signing operation fails.
2962 MH/01 Treat the transport option dkim_domain as a colon separated
2963 list, not as a single string, and sign the message with each element,
2964 omitting multiple occurences of the same signer.
2966 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2967 Bugzilla 985, 986. Patch by Simon Arlott
2969 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2970 Patch by Simon Arlott
2972 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2973 MBX locking. Notification from Dan Rosenberg.
2979 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2981 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2983 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2985 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2987 NM/04 Bugzilla 915: Flag broken perl installation during build.
2993 TK/01 Added patch by Johannes Berg that expands the main option
2994 "spamd_address" if it starts with a dollar sign.
2996 TK/02 Write list of recipients to X-Envelope-Sender header when building
2997 the mbox-format spool file for content scanning (suggested by Jakob
3000 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
3001 (http://www.dcc-servers.net/) support via dccifd. Activated by
3002 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
3004 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
3005 by Mark Daniel Reidel <mr@df.eu>.
3007 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
3008 When building exim an external PCRE library is now needed -
3009 PCRE is a system library on the majority of modern systems.
3010 See entry on PCRE_LIBS in EDITME file.
3012 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
3013 conversation. Added nologin parameter to request.
3014 Patch contributed by Kirill Miazine.
3016 TF/01 Do not log submission mode rewrites if they do not change the address.
3018 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
3020 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
3021 log files in place. Contributed by Roberto Lima.
3023 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
3025 TF/03 Bugzilla 615: When checking the local_parts router precondition
3026 after a local_part_suffix or local_part_prefix option, Exim now
3027 does not use the address's named list lookup cache, since this
3028 contains cached lookups for the whole local part.
3030 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
3031 Robert Millan. Documentation is in experimental-spec.txt.
3033 TF/04 Bugzilla 668: Fix parallel build (make -j).
3035 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
3037 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
3038 Patch provided by Jan Srzednicki.
3040 TF/05 Leading white space used to be stripped from $spam_report which
3041 wrecked the formatting. Now it is preserved.
3043 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
3044 that they are available at delivery time.
3046 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
3048 TF/08 TLS error reporting now respects the incoming_interface and
3049 incoming_port log selectors.
3051 TF/09 Produce a more useful error message if an SMTP transport's hosts
3052 setting expands to an empty string.
3054 NM/06 Bugzilla 744: EXPN did not work under TLS.
3055 Patch provided by Phil Pennock.
3057 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
3058 Patch provided by Richard Godbee.
3060 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
3061 acl_smtp_notquit, added index entry.
3063 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
3064 Patch provided by Eugene Bujak.
3066 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
3067 accept(). Patch provided by Maxim Dounin.
3069 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
3070 Patch provided by Phil Pennock.
3072 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
3074 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
3075 Patch provided by Brad "anomie" Jorsch.
3077 NM/14 Bugzilla 622: Added timeout setting to transport filter.
3078 Patch provided by Dean Brooks.
3080 TK/05 Add native DKIM support (does not depend on external libraries).
3082 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
3083 Patch provided by Graeme Fowler.
3085 NM/16 Bugzilla 851: Documentation example syntax fix.
3087 NM/17 Changed NOTICE file to remove references to embedded PCRE.
3089 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
3092 NM/19 Bugzilla 745: TLS version reporting.
3093 Patch provided by Phil Pennock.
3095 NM/20 Bugzilla 167: bool: condition support.
3096 Patch provided by Phil Pennock.
3098 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
3099 clients. Patch provided by Phil Pennock.
3101 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
3102 Patch provided by Brad "anomie" Jorsch.
3104 NM/23 Bugzilla 687: Fix misparses in eximstats.
3105 Patch provided by Heiko Schlittermann.
3107 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
3108 Patch provided by Heiko Schlittermann.
3110 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
3111 plus update to original patch.
3113 NM/26 Bugzilla 799: Documentation correction for ratelimit.
3115 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
3116 Patch provided by David Brownlee.
3118 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
3120 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
3122 NM/30 Bugzilla 888: TLS documentation bugfixes.
3124 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
3126 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
3127 Unlike the original bugzilla I have changed all shell scripts in src tree.
3129 NM/33 Bugzilla 898: Transport filter timeout fix.
3130 Patch by Todd Rinaldo.
3132 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
3133 Patch by Serge Demonchaux.
3135 NM/35 Bugzilla 39: Base64 decode bug fixes.
3136 Patch by Jakob Hirsch.
3138 NM/36 Bugzilla 909: Correct connect() call in dcc code.
3140 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
3142 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
3144 NM/39 Bugzilla 911: Fixed MakeLinks build script.
3150 TK/01 Add preliminary DKIM support. Currently requires a forked version of
3151 ALT-N's libdkim that I have put here:
3152 http://duncanthrax.net/exim-experimental/
3154 Note to Michael Haardt: I had to rename some vars in sieve.c. They
3155 were called 'true' and it seems that C99 defines that as a reserved
3156 keyword to be used with 'bool' variable types. That means you could
3157 not include C99-style headers which use bools without triggering
3158 build errors in sieve.c.
3160 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
3161 as mailq or other aliases. Changed the --help handling significantly
3162 to do whats expected. exim_usage() emits usage/help information.
3164 SC/01 Added the -bylocaldomain option to eximstats.
3166 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
3168 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
3170 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
3176 PH/01 Another patch from the Sieve maintainer.
3178 PH/02 When an IPv6 address is converted to a string for single-key lookup
3179 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
3180 dots are used instead of colons so that keys in lsearch files need not
3181 contain colons. This was done some time before quoting was made available
3182 in lsearch files. However, iplsearch files do require colons in IPv6 keys
3183 (notated using the quote facility) so as to distinguish them from IPv4
3184 keys. This meant that lookups for IP addresses in host lists did not work
3185 for iplsearch lookups.
3187 This has been fixed by arranging for IPv6 addresses to be expressed with
3188 colons if the lookup type is iplsearch. This is not incompatible, because
3189 previously such lookups could never work.
3191 The situation is now rather anomalous, since one *can* have colons in
3192 ordinary lsearch keys. However, making the change in all cases is
3193 incompatible and would probably break a number of configurations.
3195 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
3198 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
3199 conversion specification without a maximum field width, thereby enabling
3200 a rogue spamd server to cause a buffer overflow. While nobody in their
3201 right mind would setup Exim to query an untrusted spamd server, an
3202 attacker that gains access to a server running spamd could potentially
3203 exploit this vulnerability to run arbitrary code as the Exim user.
3205 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
3206 $primary_hostname instead of what libspf2 thinks the hosts name is.
3208 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
3209 a directory entry by the name of the lookup key. Previously, if a
3210 symlink pointed to a non-existing file or a file in a directory that
3211 Exim lacked permissions to read, a lookup for a key matching that
3212 symlink would fail. Now it is enough that a matching directory entry
3213 exists, symlink or not. (Bugzilla 503.)
3215 PH/03 The body_linecount and body_zerocount variables are now exported in the
3218 PH/04 Added the $dnslist_matched variable.
3220 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
3221 This means they are set thereafter only if the connection becomes
3224 PH/06 Added the client_condition to authenticators so that some can be skipped
3225 by clients under certain conditions.
3227 PH/07 The error message for a badly-placed control=no_multiline_responses left
3228 "_responses" off the end of the name.
3230 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
3232 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
3233 (without spaces) instead of just copying the configuration text.
3235 PH/10 Added the /noupdate option to the ratelimit ACL condition.
3237 PH/11 Added $max_received_linelength.
3239 PH/12 Added +ignore_defer and +include_defer to host lists.
3241 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
3242 way in which PCRE > 7.0 is built.
3244 PH/14 Implemented queue_only_load_latch.
3246 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
3247 MAIL command. The effect was to mangle the value on 64-bit systems.
3249 PH/16 Another patch from the Sieve maintainer.
3251 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
3253 PH/18 If a system quota error occurred while trying to create the file for
3254 a maildir delivery, the message "Mailbox is full" was not appended to the
3255 bounce if the delivery eventually timed out. Change 4.67/27 below applied
3256 only to a quota excession during the actual writing of the file.
3258 PH/19 It seems that peer DN values may contain newlines (and other non-printing
3259 characters?) which causes problems in log lines. The DN values are now
3260 passed through string_printing() before being added to log lines.
3262 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
3263 and InterBase are left for another time.)
3265 PH/21 Added message_body_newlines option.
3267 PH/22 Guard against possible overflow in moan_check_errorcopy().
3269 PH/23 POSIX allows open() to be a macro; guard against that.
3271 PH/24 If the recipient of an error message contained an @ in the local part
3272 (suitably quoted, of course), incorrect values were put in $domain and
3273 $local_part during the evaluation of errors_copy.
3279 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
3280 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
3283 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
3284 issue a MAIL command.
3286 PH/02 In an ACL statement such as
3288 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
3290 if a client was not listed at all, or was listed with a value other than
3291 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
3292 the condition was not true (as it should be), so access was not denied.
3293 The bug was that the ! inversion was incorrectly passed on to the second
3294 item. This has been fixed.
3296 PH/03 Added additional dnslists conditions == and =& which are different from
3297 = and & when the dns lookup returns more than one IP address.
3299 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
3300 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
3302 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
3303 FSYNC, which compiles an option called disable_fsync that allows for
3304 bypassing fsync(). The documentation is heavily laced with warnings.
3306 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
3308 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
3309 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
3310 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
3311 including adding "make clean"; (3) Added -fPIC when compiling the test
3312 dynamically loaded module, to get rid of a warning.
3314 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
3315 message fails, move_frozen_messages = true and ignore_bounce_errors_after
3316 = 0s. The bug is otherwise harmless.
3318 PH/07 There was a bug in the dovecot authenticator such that the value of
3319 $auth1 could be overwritten, and so not correctly preserved, after a
3320 successful authentication. This usually meant that the value preserved by
3321 the server_setid option was incorrect.
3323 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
3325 PH/09 Installed PCRE release 7.0.
3327 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
3328 run for batched SMTP input. It is now run at the start of every message
3329 in the batch. While fixing this I discovered that the process information
3330 (output by running exiwhat) was not always getting set for -bs and -bS
3331 input. This is fixed, and it now also says "batched" for BSMTP.
3333 PH/11 Added control=no_pipelining.
3335 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
3336 patch, slightly modified), and move the expansion of helo_data till after
3337 the connection is made in the smtp transport (so it can use these
3340 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
3342 PH/14 Added log_selector = +pid.
3344 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
3346 PH/16 Add ${if forany and ${if forall.
3348 PH/17 Added dsn_from option to vary the From: line in DSNs.
3350 PH/18 Flush SMTP output before performing a callout, unless control =
3351 no_callout_flush is set.
3353 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
3354 was true (the default) a successful delivery failed to delete the retry
3355 item, thus causing premature timeout of the address. The bug is now
3358 PH/20 Added hosts_avoid_pipelining to the smtp transport.
3360 PH/21 Long custom messages for fakedefer and fakereject are now split up
3361 into multiline responses in the same way that messages for "deny" and
3362 other ACL rejections are.
3364 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
3365 with slight modification.
3367 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
3368 draft, changing the syntax and factoring some duplicate code".
3370 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
3371 for deliveries of the second and subsequent messages over the same SMTP
3374 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
3375 ${reduce, with only minor "tidies".
3377 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
3379 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
3380 expansion side effects.
3382 PH/27 When a message times out after an over-quota error from an Exim-imposed
3383 quota, the bounce message says "mailbox is full". This message was not
3384 being given when it was a system quota that was exceeded. It now should
3387 MH/03 Made $recipients available in local_scan(). local_scan() already has
3388 better access to the recipient list through recipients_list[], but
3389 $recipients can be useful in postmaster-provided expansion strings.
3391 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
3392 in the case of a MAIL command with additional options following the
3393 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
3394 were accidentally chopped off.
3396 PH/29 SMTP synchronization checks are implemented when a command is read -
3397 there is a check that no more input is waiting when there shouldn't be
3398 any. However, for some commands, a delay in an ACL can mean that it is
3399 some time before the response is written. In this time, more input might
3400 arrive, invalidly. So now there are extra checks after an ACL has run for
3401 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
3402 pipelining has not been advertised.
3404 PH/30 MH's patch to allow iscntrl() characters to be list separators.
3406 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
3407 returned in the SMTP response when smtp_return_error_details was false.
3408 This has been fixed.
3410 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
3411 instead of the C I/O that was originally supplied, because problems were
3412 reported on Solaris.
3414 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
3415 Exim which did not show up earlier: it was assuming that a call to
3416 SSL_CTX_set_info_callback() might give an error value. In fact, there is
3417 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
3418 was a macro that became an assignment, so it seemed to work. This has
3419 changed to a proper function call with a void return, hence the compile
3420 error. Exim's code has been fixed.
3422 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
3425 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
3427 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
3429 PH/37 If a message is not accepted after it has had an id assigned (e.g.
3430 because it turns out to be too big or there is a timeout) there is no
3431 "Completed" line in the log. When some messages of this type were
3432 selected by exigrep, they were listed as "not completed". Others were
3433 picked up by some special patterns. I have improved the selection
3434 criteria to be more general.
3436 PH/38 The host_find_failed option in the manualroute router can now be set
3437 to "ignore", to completely ignore a host whose IP address cannot be
3438 found. If all hosts are ignored, the behaviour is controlled by the new
3439 host_all_ignored option.
3441 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
3442 homing or because of multiple MX records with /mx) generated more than
3443 one IP address, and the following item turned out to be the local host,
3444 all the secondary addresses of the first item were incorrectly removed
3445 from the list, along with the local host and any following hosts (which
3446 is what is supposed to happen).
3448 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
3449 whoever called Exim into the -H file. In the case of the daemon it was
3450 behaving confusingly. When first started, it used values for whoever
3451 started the daemon, but after a SIGHUP it used the Exim user (because it
3452 calls itself on a restart). I have changed the code so that it now always
3455 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
3456 message are rejected with the same error (e.g. no authentication or bad
3457 sender address), and a DATA command is nevertheless sent (as can happen
3458 with PIPELINING or a stupid MUA), the error message that was given to the
3459 RCPT commands is included in the rejection of the DATA command. This is
3460 intended to be helpful for MUAs that show only the final error to their
3463 PH/42 Another patch from the Sieve maintainer.
3465 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
3466 Eximstats - Fixed some broken HTML links and added missing column headers
3468 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
3469 columns for Rejects, Temp Rejects, Ham, and Spam rows.
3471 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
3473 PH/43 Yet another patch from the Sieve maintainer.
3475 PH/44 I found a way to check for a TCP/IP connection going away before sending
3476 the response to the final '.' that terminates a message, but only in the
3477 case where the client has not sent further data following the '.'
3478 (unfortunately, this is allowed). However, in many cases there won't be
3479 any further data because there won't be any more messages to send. A call
3480 to select() can be used: if it shows that the input is "ready", there is
3481 either input waiting, or the socket has been closed. An attempt to read
3482 the next input character can distinguish the two cases. Previously, Exim
3483 would have sent an OK response which the client would never have see.
3484 This could lead to message repetition. This fix should cure that, at
3485 least in a lot of common cases.
3487 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
3488 advertised in response to EHLO.
3494 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
3495 fixed by 4.65/MH/01 (is this a record?) are fixed:
3497 (i) An empty string was always treated as zero by the numeric comparison
3498 operators. This behaviour has been restored.
3500 (ii) It is documented that the numeric comparison operators always treat
3501 their arguments as decimal numbers. This was broken in that numbers
3502 starting with 0 were being interpreted as octal.
3504 While fixing these problems I realized that there was another issue that
3505 hadn't been noticed. Values of message_size_limit (both the global option
3506 and the transport option) were treated as octal if they started with 0.
3507 The documentation was vague. These values are now always treated as
3508 decimal, and I will make that clear in the documentation.
3514 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
3515 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
3518 MH/01 Don't check that the operands of numeric comparison operators are
3519 integers when their expansion is in "skipping" mode (fixes bug
3520 introduced by 4.64-PH/07).
3522 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
3523 child addresses, Exim now panics and dies. Previously, because the count
3524 is held in a short int, deliveries were likely to be lost. As such a
3525 large number of recipients for a single message is ridiculous
3526 (performance will be very, very poor), I have chosen to impose a limit
3527 rather than extend the field.
3533 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
3534 leftover -K file (the existence of which was triggered by #402).
3535 While we were at it, introduced process PID as part of the -K
3536 filename. This should rule out race conditions when creating
3539 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
3540 processing considerably. Previous code took too long for large mails,
3541 triggering a timeout which in turn triggers #401.
3543 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
3544 in the DK code in transports.c. sendfile() is not really portable,
3545 hence the _LINUX specificness.
3547 TF/01 In the add_headers option to the mail command in an Exim filter,
3548 there was a bug that Exim would claim a syntax error in any
3549 header after the first one which had an odd number of characters
3552 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
3553 callout verification, Exim cached a "reject" for the entire domain. This
3554 is correct for most verifications, but it is not correct for a recipient
3555 verification with use_sender or use_postmaster set, because in that case
3556 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
3557 case of MAIL FROM:<> rejection from other early rejections (e.g.
3558 rejection of HELO). When verifying a recipient using a non-null MAIL
3559 address, the cache is ignored if it shows MAIL FROM:<> rejection.
3560 Whatever the result of the callout, the value of the domain cache is
3561 left unchanged (for any other kind of callout, getting as far as trying
3562 RCPT means that the domain itself is ok).
3564 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
3567 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
3568 manifest itself as EPIPE rather than ECONNECT. When tidying away a
3569 session, the daemon ignores ECONNECT errors and logs others; it now
3570 ignores EPIPE as well.
3572 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
3573 (quoted-printable decoding).
3575 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
3576 later the small subsequent patch to fix an introduced bug.
3578 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
3580 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
3582 PH/08 An error is now given if message_size_limit is specified negative.
3584 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
3585 to be given (somewhat) arbitrary names.
3587 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
3590 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
3591 miscellaneous code fixes
3593 PH/10 Added the log_reject_target ACL modifier to specify where to log
3596 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
3597 hostname. This is wrong, because it relates to the incoming message (and
3598 probably the interface on which it is arriving) and not to the outgoing
3599 callout (which could be using a different interface). This has been
3600 changed to use the value of the helo_data option from the smtp transport
3601 instead - this is what is used when a message is actually being sent. If
3602 there is no remote transport (possible with a router that sets up host
3603 addresses), $smtp_active_hostname is used.
3605 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
3606 tweaks were necessary in order to get it to work (see also 21 below):
3607 (a) The code assumed that strncpy() returns a negative number on buffer
3608 overflow, which isn't the case. Replaced with Exim's string_format()
3610 (b) There were several signed/unsigned issues. I just did the minimum
3611 hacking in of casts. There is scope for a larger refactoring.
3612 (c) The code used strcasecmp() which is not a standard C function.
3613 Replaced with Exim's strcmpic() function.
3614 (d) The code set only $1; it now sets $auth1 as well.
3615 (e) A simple test gave the error "authentication client didn't specify
3616 service in request". It would seem that Dovecot has changed its
3617 interface. Fortunately there's a specification; I followed it and
3618 changed what the client sends and it appears to be working now.
3620 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
3623 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
3624 address A is aliased to B and C, where B exists and C does not. Without
3625 -v the output is "A verified" because verification stops after a
3626 successful redirection if more than one address is generated. However,
3627 with -v the child addresses are also verified. Exim was outputting "A
3628 failed to verify" and then showing the successful verification for C,
3629 with its parentage. It now outputs "B failed to verify", showing B's
3630 parentage before showing the successful verification of C.
3632 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
3633 look up a TXT record in a specific list after matching in a combined
3636 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
3637 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
3638 they consult the DNS. I had assumed they would set it the way they
3639 wanted; and indeed my experiments on Linux seem to show that in some
3640 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
3641 To be on the safe side, however, I have now made the interface to
3642 host_find_byname() similar to host_find_bydns(), with an argument
3643 containing the DNS resolver options. The host_find_byname() function now
3644 sets these options at its start, just as host_find_bydns() does. The smtp
3645 transport options dns_qualify_single and dns_search_parents are passed to
3646 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
3647 of host_find_byname() use the default settings of RES_DEFNAMES
3648 (qualify_single) but not RES_DNSRCH (search_parents).
3650 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
3651 spool_read_header() do less string testing, by means of a preliminary
3652 switch on the second character of optional "-foo" lines. (This is
3653 overdue, caused by the large number of possibilities that now exist.
3654 Originally there were few.) While I was there, I also converted the
3655 str(n)cmp tests so they don't re-test the leading "-" and the first
3656 character, in the hope this might squeeze out yet more improvement.
3658 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
3659 flag allowing group syntax was set by the header_syntax check but not
3660 turned off, possible causing trouble later; (2) The flag was not being
3661 set at all for the header_verify test, causing "group"-style headers to
3662 be rejected. I have now set it in this case, and also caused header_
3663 verify to ignore an empty address taken from a group. While doing this, I
3664 came across some other cases where the code for allowing group syntax
3665 while scanning a header line wasn't quite right (mostly, not resetting
3666 the flag correctly in the right place). These bugs could have caused
3667 trouble for malformed header lines. I hope it is now all correct.
3669 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
3670 with the "reply" argument non-NULL. The code, however (which originally
3671 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
3672 but it didn't always do it. This confused somebody who was copying the
3673 code for some other use. I have removed all the tests.
3675 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
3676 feature that was used to support insecure browsers during the U.S. crypto
3677 embargo. It requires special client support, and Exim is probably the
3678 only MTA that supported it -- and would never use it because real RSA is
3679 always available. This code has been removed, because it had the bad
3680 effect of slowing Exim down by computing (never used) parameters for the
3681 RSA_EXPORT functionality.
3683 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
3684 authenticator to fail if there's a tab character in the incoming data
3685 (there should never be unless someone is messing about, as it's supposed
3686 to be base64-encoded). Also added, on Timo's advice, the "secured" option
3687 if the connection is using TLS or if the remote IP is the same as the
3688 local IP, and the "valid-client-cert option" if a client certificate has
3691 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
3692 authenticators. This can be used for authorization after authentication
3693 succeeds. (In the case of plaintext, it servers for both authentication
3696 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
3697 if any retry times were supplied.
3699 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
3700 connection, where there is no client IP address to check. In this
3701 situation, the verify now always succeeds.
3703 PH/25 Applied John Jetmore's -Mset patch.
3705 PH/26 Added -bem to be like -Mset, but loading a message from a file.
3707 PH/27 In a string expansion for a processed (not raw) header when multiple
3708 headers of the same name were present, leading whitespace was being
3709 removed from all of them, but trailing whitespace was being removed only
3710 from the last one. Now trailing whitespace is removed from each header
3711 before concatenation. Completely empty headers in a concatenation (as
3712 before) are ignored.
3714 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
3715 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
3717 PH/29 [Removed. This was a change that I later backed out, and forgot to
3718 correct the ChangeLog entry (that I had efficiently created) before
3719 committing the later change.]
3721 PH/30 Exim was sometimes attempting to deliver messages that had suffered
3722 address errors (4xx response to RCPT) over the same connection as other
3723 messages routed to the same hosts. Such deliveries are always "forced",
3724 so retry times are not inspected. This resulted in far too many retries
3725 for the affected addresses. The effect occurred only when there were more
3726 hosts than the hosts_max_try setting in the smtp transport when it had
3727 the 4xx errors. Those hosts that it had tried were not added to the list
3728 of hosts for which the message was waiting, so if all were tried, there
3729 was no problem. Two fixes have been applied:
3731 (i) If there are any address or message errors in an SMTP delivery, none
3732 of the hosts (tried or untried) are now added to the list of hosts
3733 for which the message is waiting, so the message should not be a
3734 candidate for sending over the same connection that was used for a
3735 successful delivery of some other message. This seems entirely
3736 reasonable: after all the message is NOT "waiting for some host".
3737 This is so "obvious" that I'm not sure why it wasn't done
3738 previously. Hope I haven't missed anything, but it can't do any
3739 harm, as the worst effect is to miss an optimization.
3741 (ii) If, despite (i), such a delivery is accidentally attempted, the
3742 routing retry time is respected, so at least it doesn't keep
3743 hammering the server.
3745 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
3746 in ${readsocket because some servers need this prod.
3748 PH/32 Added some extra debug output when updating a wait-xxx database.
3750 PH/33 The hint "could be header name not terminated by colon", which has been
3751 given for certain expansion errors for a long time, was not being given
3752 for the ${if def:h_colon_omitted{... case.
3754 PH/34 The spec says: "With one important exception, whenever a domain list is
3755 being scanned, $domain contains the subject domain." There was at least
3756 one case where this was not true.
3758 PH/35 The error "getsockname() failed: connection reset by peer" was being
3759 written to the panic log as well as the main log, but it isn't really
3760 panic-worthy as it just means the connection died rather early on. I have
3761 removed the panic log writing for the ECONNRESET error when getsockname()
3764 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
3765 runs only) independently of the message's sender address. This meant
3766 that, if the 4xx error was in fact related to the sender, a different
3767 message to the same recipient with a different sender could confuse
3768 things. In particular, this can happen when sending to a greylisting
3769 server, but other circumstances could also provoke similar problems.
3770 I have changed the default so that the retry time for these errors is now
3771 based a combination of the sender and recipient addresses. This change
3772 can be overridden by setting address_retry_include_sender=false in the
3775 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
3776 remote server are returned as part of bounce messages. This was not
3777 happening for LMTP over a pipe (the lmtp transport), but now it is the
3778 same for both kinds of LMTP.
3780 PH/38 Despite being documented as not happening, Exim was rewriting addresses
3781 in header lines that were in fact CNAMEs. This is no longer the case.
3783 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
3784 and queue runs started by the daemon processed all messages. This has
3785 been fixed so that -R and -S can now usefully be given with -q<time>.
3787 PH/40 Import PCRE release 6.7 (fixes some bugs).
3789 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
3791 PH/42 Give an error if -q is specified more than once.
3793 PH/43 Renamed the variables $interface_address and $interface_port as
3794 $received_ip_address and $received_port, to make it clear that these
3795 values apply to message reception, and not to the outgoing interface when
3796 a message is delivered. (The old names remain recognized, of course.)
3798 PH/44 There was no timeout on the connect() call when using a Unix domain
3799 socket in the ${readsocket expansion. There now is.
3801 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
3802 be meaningful with "accept".
3804 SC/01 Eximstats V1.43
3805 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
3807 SC/02 Eximstats V1.44
3808 Use a glob alias rather than an array ref in the generated
3809 parser. This improves both readability and performance.
3811 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
3812 Collect SpamAssassin and rejection statistics.
3813 Don't display local sender or destination tables unless
3814 there is data to show.
3815 Added average volumes into the top table text output.
3817 SC/04 Eximstats V1.46
3818 Collect data on the number of addresses (recipients)
3819 as well as the number of messages.
3821 SC/05 Eximstats V1.47
3822 Added 'Message too big' to the list of mail rejection
3823 reasons (thanks to Marco Gaiarin).
3825 SC/06 Eximstats V1.48
3826 Mainlog lines which have GMT offsets and are too short to
3827 have a flag are now skipped.
3829 SC/07 Eximstats V1.49 (Alain Williams)
3830 Added the -emptyok flag.
3832 SC/08 Eximstats V1.50
3833 Fixes for obtaining the IP address from reject messages.
3835 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
3836 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
3837 whitespace changes from 4.64-PH/27
3839 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
3842 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
3843 are found, allow negative numbers in numeric criteria)
3845 JJ/06 exipick.20061117.2, added new $message_body_missing variable
3847 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
3848 to match changes made in 4.64-PH/43
3850 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
3852 PH/47 Put in an explicit test for a DNS lookup of an address record where the
3853 "domain" is actually an IP address, and force a failure. This locks out
3854 those revolvers/nameservers that support "A-for-A" lookups, in
3855 contravention of the specifications.
3857 PH/48 When a host name was looked up from an IP address, and the subsequent
3858 forward lookup of the name timed out, the host name was left in
3859 $sender_host_name, contrary to the specification.
3861 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3862 restricted to single-key lookups, Exim was not diagnosing an error if
3863 * or *@ was used with a query-style lookup.
3865 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3867 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3868 long ago, but noone interested enough thought of it. Let's just say that
3869 the "1.1" means that there are some new functions that weren't there at
3870 some point in the past.
3872 PH/51 Error processing for expansion failure of helo_data from an smtp
3873 transport during callout processing was broken.
3875 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3876 tested/used via the -bh/-bhc/-bs options.
3878 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3879 bug, fixed in subsequent PCRE releases).
3881 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3882 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3884 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3890 SC/01 Use a glob alias rather than an array ref in eximstats generated
3891 parser. This improves both readability and performance.
3893 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3894 Don't display local sender or destination tables in eximstats unless
3895 there is data to show.
3896 Added average volumes into the eximstats top table text output.
3898 SC/03 Collect data on the number of addresses (recipients) as well
3899 as the number of messages in eximstats.
3901 TF/01 Correct an error in the documentation for the redirect router. Exim
3902 does (usually) call initgroups() when daemonizing.
3904 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3905 with consistent privilege compared to when running as a daemon.
3907 TF/03 Note in the spec that $authenticated_id is not set for local
3908 submissions from trusted users.
3910 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3911 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3913 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3914 by adding some example configuration directives to the default
3915 configuration file. A little bit of work is required to uncomment the
3916 directives and define how usernames and passwords are checked, but
3917 there is now a framework to start from.
3919 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3920 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3921 without this. I don't know how relevant this is to other LDAP libraries.
3923 PH/02 Add the verb name to the "unknown ACL verb" error.
3925 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3927 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3929 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3930 directory not expanded when it should be if an expanded home directory
3931 was set for the address (which is overridden by the transport).
3933 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3936 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3937 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3938 because it is too late at that time, and has no effect.
3940 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3941 security issue with \' (bugzilla #107). I could not use the
3942 PQescapeStringConn() function, because it needs a PGconn value as one of
3945 PH/08 When testing addresses using -bt, indicate those final addresses that
3946 are duplicates that would not cause an additional delivery. At least one
3947 person was confused, thinking that -bt output corresponded to deliveries.
3948 (Suppressing duplicates isn't a good idea as you lose the information
3949 about possibly different redirections that led to the duplicates.)
3951 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3952 systems where poll() doesn't work, in particular OS X.
3954 PH/10 Added more information to debugging output for retry time not reached.
3956 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3957 operations in malware.c.
3959 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3962 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3963 syslog_duplication set false, log lines that would normally be written
3964 both the the main log and to the reject log were not written to syslog at
3967 PH/14 In the default configuration, change the use of "message" in ACL warn
3968 statements to "add_header".
3970 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3971 not followed by a command (e.g. "seen endif").
3973 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3974 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3977 PH/17 Added extra conditions to the default value of delay_warning_condition
3981 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3982 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3983 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3986 The Auto-Submitted: and various List- headers are standardised, whereas I
3987 don't think Precedence: ever was.
3989 PH/18 Refactored debugging code in route_finduser() to show more information,
3990 in particular, the error code if getpwnam() issues one.
3992 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3993 This is apparently needed in addition to the PH/07 change above to avoid
3994 any possible encoding problems.
3996 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3997 but not after initializing Perl.
3999 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
4000 output them only if debugging. By default they are written stderr,
4001 apparently, which is not desirable.
4003 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
4006 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
4009 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
4011 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
4012 authenticated or an ident call has been made. Suppress the default
4013 values for $authenticated_id and $authenticated_sender (but permit -oMai
4014 and -oMas) when testing with -bh.
4016 PH/24 Re-jigged the order of the tests in the default configuration so that the
4017 tests for valid domains and recipients precede the DNS black list and CSA
4018 tests, on the grounds that those ones are more expensive.
4020 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
4021 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
4022 This bug exists in every version of Exim that I still have, right back to
4025 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
4026 However, an attempt to turn on case-sensitivity in a regex key by
4027 including (?-i) didn't work because the subject string was already
4028 lowercased, and the effects were non-intuitive. It turns out that a
4029 one-line patch can be used to allow (?-i) to work as expected.
4035 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
4036 other effects) broke the use of negated acl sub-conditions.
4038 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
4041 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
4042 "Deny" causes Exim to reject the incoming connection with a 554 error.
4043 Unfortunately, if there is a major crisis, such as a disk failure,
4044 tcp-wrappers gives "deny", whereas what one would like would be some
4045 kind of temporary error. A kludge has been added to help with this.
4046 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
4047 554 error is used if errno is still zero or contains ENOENT (which occurs
4048 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
4051 PH/03 Add -lutil to the default FreeBSD LIBS setting.
4053 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
4054 errors. Otherwise a message that provokes a temporary error (when other
4055 messages do not) can cause a whole host to time out.
4057 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
4058 addresses were routed directly to files or pipes from a redirect router.
4059 File deliveries just didn't batch; pipe deliveries might have suffered
4062 PH/06 A failure to get a lock for a hints database would erroneously always say
4063 "Failed to get write lock", even when it was really a read lock.
4065 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
4066 of 0600. This has been changed to use the value of the lockfile_mode
4067 option (which defaults to 0600).
4069 PH/08 Applied small patch from the Sieve maintainer.
4071 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
4072 folder from quota calculations, a direct delivery into this folder messed
4073 up the contents of the maildirsize file. This was because the regex was
4074 used only to exclude .Trash (or whatever) when the size of the mailbox
4075 was calculated. There was no check that a delivery was happening into an
4076 excluded directory. This bug has been fixed by ignoring all quota
4077 processing for deliveries into excluded directories.
4079 PH/10 Added the maildirfolder_create_regex option to appendfile.
4085 PH/01 The code for finding all the local interface addresses on a FreeBSD
4086 system running IPv6 was broken. This may well have applied to all BSD
4087 systems, as well as to others that have similar system calls. The broken
4088 code found IPv4 interfaces correctly, but gave incorrect values for the
4089 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
4090 that it would not match correctly against @[] and not recognize the IPv6
4093 PH/02 The ipliteral router was not recognizing addresses of the form user@
4094 [ipv6:....] because it didn't know about the "ipv6:" prefix.
4096 PH/03 Added disable_ipv6.
4098 PH/04 Changed $reply_address to use the raw form of the headers instead of the
4099 decoded form, because it is most often used to construct To: headers
4100 lines in autoreplies, and the decoded form may well be syntactically
4101 invalid. However, $reply_address has leading white space removed, and all
4102 newlines turned into spaces so that the autoreply transport does not
4105 PH/05 If group was specified without a user on a router, and no group or user
4106 was specified on a transport, the group from the router was ignored.
4108 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
4109 for visible compile-time settings that can be used to change these
4110 numbers, for those that want even more. Backwards compatibility with old
4111 spool files has been maintained. However, going back to a previous Exim
4112 release will lost any variables that are in spool files.
4114 PH/07 Two small changes when running in the test harness: increase delay when
4115 passing a TCP/IP connection to a new process, in case the original
4116 process has to generate a bounce, and remove special handling of
4117 127.0.0.2 (sic), which is no longer necessary.
4119 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
4120 be the same on different OS.
4122 PH/09 Moved a debug statement in filter processing to avoid a race problem when
4125 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
4126 whether --show-vars was specified or not
4128 JJ/02 exipick: Added support for new ACL variable spool format introduced
4131 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
4132 syntactically invalid From: or Reply-to: line, and a filter used this to
4133 generate an autoreply, and therefore failed to obtain an address for the
4134 autoreply, Exim could try to deliver to a non-existent relative file
4135 name, causing unrelated and misleading errors. What now happens is that
4136 it logs this as a hard delivery error, but does not attempt to create a
4139 PH/11 The exinext utility has a -C option for testing purposes, but although
4140 the given file was scanned by exinext itself; it wasn't being passed on
4141 when Exim was called.
4143 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
4144 an end-of-file indication when reading a command response.
4146 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
4147 compiled. In many other places in Exim, IPv6 addresses are always
4148 recognized, so I have changed this. It also means that IPv4 domain
4149 literals of the form [IPV4:n.n.n.n] are now always recognized.
4151 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
4152 used if the router is not running as root, for example, when verifying at
4153 ACL time, or when using -bh. The debugging output from this situation was
4154 non-existent - all you got was a failure to exec. I have made two
4157 (a) Failures to set uid/gid, the current directory, or a process leader
4158 in a subprocess such as that created by queryprogram now generate
4159 suitable debugging output when -d is set.
4161 (b) The queryprogram router detects when it is not running as root,
4162 outputs suitable debugging information if -d is set, and then runs
4163 the subprocess without attempting to change uid/gid.
4165 PH/15 Minor change to Makefile for building test_host (undocumented testing
4168 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
4169 additional section of a DNS packet that returns MX or SRV records.
4170 Instead, it always explicitly searches for A/AAAA records. This avoids
4171 major problems that occur when a DNS server includes only records of one
4172 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
4173 fixed another bug: if SRV records were looked up and the corresponding
4174 address records were *not* found in the additional section, the port
4175 values from the SRV records were lost.
4177 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
4178 using the correct key (the original address) when searching the retry
4179 rules in order to find which one to use for generating the retry hint.
4181 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
4182 adding the default one. Similarly, if it contains a Reply-To: header, the
4183 errors_reply_to option, if set, is not used.
4185 PH/19 When calculating a retry time, Exim used to measure the "time since
4186 failure" by looking at the "first failed" field in the retry record. Now
4187 it does not use this if it is later than than the arrival time of the
4188 message. Instead it uses the arrival time. This makes for better
4189 behaviour in cases where some deliveries succeed, thus re-setting the
4190 "first failed" field. An example is a quota failure for a huge message
4191 when small messages continue to be delivered. Without this change, the
4192 "time since failure" will always be short, possible causing more frequent
4193 delivery attempts for the huge message than are intended.
4194 [Note: This change was subsequently modified - see PH/04 for 4.62.]
4196 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
4197 $1, $2, $3) because the numerical variables can be reset during some
4198 expansion items (e.g. "match"), thereby losing the authentication data.
4200 PH/21 Make -bV show the size of off_t variables so that the test suite can
4201 decide whether to run tests for quotas > 2G.
4203 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
4204 mailbox_size, and mailbox_filecount in the appendfile transport. If a
4205 filecount value is greater than 2G or if a quota value is greater than 2G
4206 on a system where the size of off_t is not greater than 4, a panic error
4209 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
4210 never match. The debug and -bh output now contains an explicit error
4211 message indicating a malformed IPv4 address or mask.
4213 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
4214 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
4215 PH/23 above applies.
4217 PH/25 Do not write to syslog when running in the test harness. The only
4218 occasion when this arises is a failure to open the main or panic logs
4219 (for which there is an explicit test).
4221 PH/26 Added the /no_tell option to "control=freeze".
4223 PH/27 If a host name lookup failed very early in a connection, for example, if
4224 the IP address matched host_lookup and the reverse lookup yielded a name
4225 that did not have a forward lookup, an error message of the form "no IP
4226 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
4227 could be logged. Now it outputs the IP address instead of "NULL".
4229 PH/28 An enabling patch from MH: add new function child_open_exim2() which
4230 allows the sender and the authenticated sender to be set when
4231 submitting a message from within Exim. Since child_open_exim() is
4232 documented for local_scan(), the new function should be too.
4234 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
4235 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
4236 results in an empty string is now treated as unset.
4238 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
4240 PH/31 Added sender_verify_fail logging option.
4242 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
4243 needed by RFC 822 but not by RFC 2822 was commented out. I have now
4244 tidied the source and removed it altogether.
4246 PH/33 When a queue run was abandoned because the load average was too high, a
4247 log line was always written; now it is written only if the queue_run log
4248 selector is set. In addition, the log line for abandonment now contains
4249 information about the queue run such as the pid. This is always present
4250 in "start" and "stop" lines but was omitted from the "abandon" line.
4252 PH/34 Omit spaces between a header name and the colon in the error message that
4253 is given when verify = headers_syntax fails (if there are lots of them,
4254 the message gets confusing).
4256 PH/35 Change the default for dns_check_names_pattern to allow slashes within
4257 names, as there are now some PTR records that contain slashes. This check
4258 is only to protect against broken name servers that fall over on strange
4259 characters, so the fact that it applies to all lookups doesn't matter.
4261 PH/36 Now that the new test suite is complete, we can remove some of the
4262 special code in Exim that was needed for the old test suite. For example,
4263 sorting DNS records because real resolvers return them in an arbitrary
4264 order. The new test suite's fake resolver always returns records in the
4267 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
4268 bounces) except when queue_only is set, to avoid logging races between
4269 the different processes.
4271 PH/38 Panic-die if .include specifies a non-absolute path.
4273 PH/39 A tweak to the "H" retry rule from its user.
4275 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
4276 a label. They prevented compilation on older perls.
4278 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
4279 a warning to be raised on newish perls.
4281 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
4282 on queue. Changes to match documented behaviour of showing count of
4283 messages matching specified criteria.
4285 PH/40 Changed the default ident timeout from 30s to 5s.
4287 PH/41 Added support for the use of login_cap features, on those BSD systems
4288 that have them, for controlling the resources used by pipe deliveries.
4290 PH/42 The content-scanning code uses fopen() to create files in which to put
4291 message data. Previously it was not paying any attention to the mode of
4292 the files. Exim runs with umask(0) because the rest of the code creates
4293 files with open(), and sets the required mode explicitly. Thus, these
4294 files were ending up world-writeable. This was not a big issue, because,
4295 being within the spool directory, they were not world-accessible. I have
4296 created a function called modefopen, which takes an additional mode
4297 argument. It sets umask(777), creates the file, chmods it to the required
4298 mode, then resets the umask. All the relevant calls to fopen() in the
4299 content scanning code have been changed to use this function.
4301 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
4302 to 24 hours. This avoids potential overflow problems when processing G
4303 and H retry rules. I suspect nobody ever tinkers with this value.
4305 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
4307 PH/45 When the plaintext authenticator is running as a client, the server's
4308 challenges are checked to ensure they are valid base64 strings. By
4309 default, the authentication attempt is cancelled if an invalid string is
4310 received. Setting client_ignore_invalid_base64 true ignores these errors.
4311 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
4312 they are received. Thus, the responses can be made to depend on the
4313 challenges. If an invalid string is ignored, an empty string is placed in
4316 PH/46 Messages that are created by the autoreply transport now contains a
4317 References: header, in accordance with RFCs 2822 and 3834.
4319 PH/47 Added authenticated_sender_force to the smtp transport.
4321 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
4323 PH/49 Installed latest patch from the Sieve maintainer.
4325 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
4326 was also set, the appendfile transport was unnecessarily scanning a
4327 directory of message files (e.g. for maildir delivery) to find the count
4328 of files (along with the size), even though it did not need this
4329 information. It now does the scan only if it needs to find either the
4330 size of the count of files.
4332 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
4334 PH/52 Two bugs concerned with error handling when the smtp transport is
4337 (i) Exim was not creating retry information for temporary errors given
4338 for individual recipients after the DATA command when the smtp transport
4339 was used in LMTP mode. This meant that they could be retried too
4340 frequently, and not timed out correctly.
4342 (ii) Exim was setting the flag that allows error details to be returned
4343 for LMTP errors on RCPT commands, but not for LMTP errors for individual
4344 recipients that were returned after the DATA command.
4346 PH/53 This is related to PH/52, but is more general: for any failing address,
4347 when detailed error information was permitted to be returned to the
4348 sender, but the error was temporary, then after the final timeout, only
4349 "retry timeout exceeded" was returned. Now it returns the full error as
4350 well as "retry timeout exceeded".
4352 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
4353 do this, and (what is worse) MTAs that accept it.
4355 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
4356 will now be deprecated.
4358 PH/56 New os.c-cygwin from the Cygwin maintainer.
4360 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
4361 formats (previously only available in exim formats via -bpr, -bpru,
4362 and -bpra. Now also available in native and exiqgrep formats)
4364 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
4365 with very large, slow to parse queues
4367 JJ/08 exipick: added ! as generic prefix to negate any criteria format
4369 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
4371 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
4372 responses to authentication challenges, though it was showing the
4373 challenges; (ii) I've removed the CR characters from the debug output for
4376 PH/58 Allow for the insertion of a newline as well as a space when a string
4377 is turned into more than one encoded-word during RFC 2047 encoding. The
4378 Sieve code now uses this.
4380 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
4381 data_4xx, lost_connection, tls_required.
4383 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
4384 message was being sent as an SMTP response.
4386 PH/61 Add -l and -k options to exicyclog.
4388 PH/62 When verifying, if an address was redirected to one new address, so that
4389 verification continued, and the new address failed or deferred after
4390 having set something in $address_data, the value of $address_data was not
4391 passed back to the ACL. This was different to the case when no
4392 redirection occurred. The value is now passed back in both cases.
4394 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
4395 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
4396 use login_cap.h, so on its own it isn't the distinguishing feature. The
4397 new name refers directly to the setclassresources() function.
4399 PH/65 Added configuration files for NetBSD3.
4401 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
4403 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
4404 is preferred over IPv4.
4406 PH/68 The bounce_return_message and bounce_return_body options were not being
4407 honoured for bounces generated during the reception of non-SMTP messages.
4408 In particular, this applied to messages rejected by the ACL. This bug has
4409 been fixed. However, if bounce_return_message is true and bounce_return_
4410 body is false, the headers that are returned for a non-SMTP message
4411 include only those that have been read before the error was detected.
4412 (In the case of an ACL rejection, they have all been read.)
4414 PH/69 The HTML version of the specification is now built in a directory called
4415 spec_html instead of spec.html, because the latter looks like a path with
4416 a MIME-type, and this confuses some software.
4418 PH/70 Catch two compiler warnings in sieve.c.
4420 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
4421 function verify_get_ident() calls ip_connect() to connect a socket, but
4422 if the "connect()" function timed out, ip_connect() used to close the
4423 socket. However, verify_get_ident() also closes the socket later, and in
4424 between Exim writes to the log, which may get opened at this point. When
4425 the socket was closed in ip_connect(), the log could get the same file
4426 descriptor number as the socket. This naturally causes chaos. The fix is
4427 not to close the socket in ip_connect(); the socket should be closed by
4428 the function that creates it. There was only one place in the code where
4429 this was missing, in the iplookup router, which I don't think anybody now
4430 uses, but I've fixed it anyway.
4432 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
4433 well as to direct DNS lookups. Otherwise the handling of names in host
4434 lists is inconsistent and therefore confusing.
4440 PH/01 Two changes to the default runtime configuration:
4442 (1) Move the checks for relay_from_hosts and authenticated clients from
4443 after to before the (commented out) DNS black list checks.
4445 (2) Add control=submission to the relay_from_hosts and authenticated
4446 clients checks, on the grounds that messages accepted by these
4447 statements are most likely to be submissions.
4449 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
4451 (1) Generate an error if the third argument for the ${prvs expansion is
4454 (2) Treat a missing third argument of ${prvscheck as if it were an empty
4457 (3) Reset the variables that are obtained from the first argument of
4458 ${prvscheck and used in the second argument before leaving the code,
4459 because their memory is reclaimed, so using them afterwards may do
4462 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
4463 one (it's much easier than Tom thought :-).
4465 (5) Because of (4), we can now allow for the use of $prvscheck_result
4466 inside the third argument.
4468 PH/03 For some reason, the default setting of PATH when running a command from
4469 a pipe transport was just "/usr/bin". I have changed it to
4472 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
4473 anything to be listed in the output from -bV.
4475 PH/05 When a filter generated an autoreply, the entire To: header line was
4476 quoted in the delivery log line, like this:
4478 => >A.N.Other <ano@some.domain> <original@ddress> ...
4480 This has been changed so that it extracts the operative address. There
4481 may be more than one such address. If so, they are comma-separated, like
4484 => >ano@some.domain,ona@other.domain <original@ddress> ...
4486 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
4487 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
4488 not being looked up in the rDNS to get a host name, Exim was showing the
4489 IP address twice in Received: lines, even though the IP addresses were
4490 identical. For example:
4492 Received: from [1.2.3.4] (helo=[1.2.3.4])
4494 However, if the real host name was known, it was omitting the HELO data
4495 if it matched the actual IP address. This has been tidied up so that it
4496 doesn't show the same IP address twice.
4498 PH/07 When both +timestamp and +memory debugging was on, the value given by
4499 $tod_xxx expansions could be wrong, because the tod_stamp() function was
4500 called by the debug printing, thereby overwriting the timestamp buffer.
4501 Debugging no longer uses the tod_stamp() function when +timestamp is set.
4503 PH/08 When the original message was included in an autoreply transport, it
4504 always said "this is a copy of the message, including all the headers",
4505 even if body_only or headers_only was set. It now gives an appropriate
4508 PH/09 Applied a patch from the Sieve maintainer which:
4510 o fixes some comments
4511 o adds the (disabled) notify extension core
4512 o adds some debug output for the result of if/elsif tests
4513 o points to the current vacation draft in the documentation
4514 and documents the missing references header update
4518 o fixes a bug in processing the envelope test (when testing
4519 multiple envelope elements, the last element determined the
4522 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
4523 Electronic Mail") by including:
4525 Auto-submitted: auto-generated
4527 in the messages that it generates (bounce messages and others, such as
4528 warnings). In the case of bounce messages for non-SMTP messages, there was
4529 also a typo: it was using "Auto_submitted" (underscore instead of
4530 hyphen). Since every message generated by Exim is necessarily in response
4531 to another message, thes have all been changed to:
4533 Auto-Submitted: auto-replied
4535 in accordance with these statements in the RFC:
4537 The auto-replied keyword:
4539 - SHOULD be used on messages sent in direct response to another
4540 message by an automatic process,
4542 - MUST NOT be used on manually-generated messages,
4544 - MAY be used on Delivery Status Notifications (DSNs) and Message
4545 Disposition Notifications (MDNs),
4547 - MUST NOT be used on messages generated by automatic or periodic
4548 processes, except for messages which are automatic responses to
4551 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
4552 to the default Received: header definition.
4554 PH/12 Added log selector acl_warn_skipped (default on).
4556 PH/13 After a successful wildlsearch lookup, discard the values of numeric
4557 variables because (a) they are in the wrong storage pool and (b) even if
4558 they were copied, it wouldn't work properly because of the caching.
4560 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
4561 checking when decoding. Apparently there are clients that generate
4562 overlong encoded strings. Why am I not surprised?
4564 PH/15 If the first argument of "${if match_address" was not empty, but did not
4565 contain an "@" character, Exim crashed. Now it writes a panic log message
4566 and treats the condition as false.
4568 PH/16 In autoreply, treat an empty string for "once" the same as unset.
4570 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
4571 extension "envelope-auth". The code is finished and in agreement with
4572 other implementations, but there is no documentation so far and in fact,
4573 nobody wrote the draft yet. This extension is currently #undef'ed, thus
4574 not changing the active code.
4576 Print executed "if" and "elsif" statements when debugging is used. This
4577 helps a great deal to understand what a filter does.
4579 Document more things not specified clearly in RFC3028. I had all this
4580 sorted out, when out of a sudden new issues came to my mind. Oops."
4582 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
4585 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
4586 canonical form (as documented). However, after a host name lookup from
4587 the IP address, check_host() was doing a simple string comparison with
4588 addresses acquired from the DNS when checking that the found name did
4589 have the original IP as one of its addresses. Since any found IPv6
4590 addresses are likely to be in abbreviated form, the comparison could
4591 fail. Luckily, there already exists a function for doing the comparison
4592 by converting both addresses to binary, so now that is used instead of
4593 the text comparison.
4595 PH/20 There was another similar case to PH/19, when a complete host name was
4596 given in a host list; looking up its IP address could give an abbreviated
4597 form, whereas the current host's name might or might not be abbreviated.
4598 The same fix has been applied.
4604 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
4605 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
4608 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
4609 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
4611 PH/03 Typo: missing ".o" in src/pcre/Makefile.
4613 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
4614 header line, restrict the check to what is listed in RFCs 2369 and 2929.
4615 Also, for "Auto-Submitted", treat anything other than "no" as
4616 non-personal, in accordance with RFC 3834. (Previously it treated
4617 anything starting "auto-" as non-personal.)
4619 TF/01 The control=submission/name=... option had a problem with syntax
4620 errors if the name included a slash character. The /name= option
4621 now slurps the rest of the string, so it can include any characters
4622 but it must come last in the list of options (after /sender_retain
4625 PH/05 Some modifications to the interface to the fake nameserver for the new
4633 TK/01 Added the "success_on_redirect" address verification option. See
4634 NewStuff for rationale and an example.
4636 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
4638 PH/02 Patch to exigrep to allow it to work on syslog lines.
4640 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
4641 fread() to skip over the body file's header line, because in Cygwin the
4642 header line is locked and is inaccessible.
4644 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
4645 co-exist for some time) to make it clear that it is the Exim ID that is
4646 referenced, not the Message-ID: header line.
4648 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
4649 string_format() function, because snprintf() does not exist on all
4652 PH/06 The use of forbid_filter_existstest now also locks out the use of the
4653 ${stat: expansion item.
4655 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
4656 protocol synchronization error", to keep the pedants happy.
4658 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
4659 well as for IRIX systems, when gcc is being used. See the host.c source
4662 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
4664 PH/10 Named domain lists were not working if used in a queue_smtp_domains
4667 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
4668 transport and to the smtp transport in LMTP mode.
4670 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
4672 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
4673 run a filter in a subprocess. This could lead to confusion in subsequent
4674 lookups in the parent process. There should also be a search_tidyup() at
4675 the end of the subprocess.
4677 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
4678 only if the host matched helo_try_verify_hosts, which caused the
4679 verification to occur when the EHLO/HELO command was issued. The ACL just
4680 tested the remembered result. Now, if a previous verification attempt has
4681 not happened, "verify = helo" does it there and then.
4683 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
4685 TK/03 Fix log output including CR from clamd.
4687 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
4688 exist provoked a memory error which could cause a segfault.
4690 PH/15 Installed PCRE 6.2
4692 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
4694 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
4695 of the problem. Specifically, suggested +O2 rather than +O1 for the
4698 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
4700 PH/20 If a delivery was routed to a non-standard port by means of an SRV
4701 record, the port was not correctly logged when the outgoing_port log
4702 selector was set (it logged the transort's default port).
4704 PH/21 Added support for host-specific ports to manualroute, queryprogram,
4705 fallback_hosts, and "hosts" in the smtp transport.
4707 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
4708 host errors such as "Connection refused".
4710 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
4711 authentication with radiusclient 0.4.9:
4713 - Error returned from rc_read_config was caught wrongly
4714 - Username/password not passed on to radius server due to wrong length.
4716 The presumption is that some radiusclient API changes for 4.51/PH/17
4717 were not taken care of correctly. The code is still untested by me (my
4718 Linux distribution still has 0.3.2 of radiusclient), but it was
4719 contributed by a Radius user.
4721 PH/24 When doing a callout, the value of $domain wasn't set correctly when
4722 expanding the "port" option of the smtp transport.
4724 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
4725 while reading a MIME header. Thanks to Tom Hughes for a patch.
4727 PH/24 Include config.h inside local_scan.h so that configuration settings are
4730 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
4731 that in an ACL for RCPT (for example), you can examine exactly what was
4734 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
4735 commands, but it was not correctly comparing the address with the actual
4736 client host address. Thus, it would show the EHLO address in Received:
4737 header lines when this was not necessary.
4739 PH/27 Added the % operator to ${eval:}.
4741 PH/28 Exim tries to create and chdir to its spool directory when it starts;
4742 it should be ignoring failures (because with -C, for example, it has lost
4743 privilege). It wasn't ignoring creation failures other than "already
4746 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
4749 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
4750 because an input line was too long, either on its own, or by virtue of
4751 too many continuations, the temporary file was not being removed, and the
4752 return code was incorrect.
4754 PH/31 Missing "BOOL" in function definition in filtertest.c.
4756 PH/32 Applied Sieve patches from the maintainer.
4758 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
4760 PH/33 Added "verify = not_blind".
4762 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
4763 Local/Makefile (with some defaults set). These are used in built scripts
4764 such as exicyclog, but they have never been used in the exim_install
4765 script (though there are many overriding facilities there). I have
4766 arranged that the exim_install script now takes note of these two
4769 PH/35 Installed configuration files for Dragonfly.
4771 PH/36 When a locally submitted message by a trusted user did not contain a
4772 From: header, and the sender address was obtained from -f or from an SMTP
4773 MAIL command, and the trusted user did not use -F to supply a sender
4774 name, $originator_name was incorrectly used when constructing a From:
4775 header. Furthermore, $originator_name was used for submission mode
4776 messages from external hosts without From: headers in a similar way,
4777 which is clearly wrong.
4779 PH/37 Added control=suppress_local_fixups.
4781 PH/38 When log_selector = +received_sender was set, and the addition of the
4782 sender made the log line's construction buffer exactly full, or one byte
4783 less than full, an overflow happened when the terminating "\n" was
4786 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
4787 when the result of a list match is failure because a DNS lookup failed.
4789 PH/40 RM_COMMAND is now used in the building process.
4791 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
4792 the "build-* directories that it finds.
4794 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
4795 address in a domain literal was a prefix of an interface address.
4797 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
4798 when verifying a sender address, unless rewrite_headers is false.
4800 PH/44 Wrote a long comment about why errors_to addresses are verified as
4801 recipients, not senders.
4803 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
4804 the ratelimit ACL was added.
4806 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
4808 PH/46 Added extra information about PostgreSQL errors to the error string.
4810 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
4811 avoiding the need to install special zones in a real server. This is
4812 backwards compatible; if it can't find the fake resolver, it drops back.
4813 Thus, both old and new test suites can be run.
4815 TF/02 Added util/ratelimit.pl
4817 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
4818 clock is set back in time.
4820 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
4821 Candler <B.Candler@pobox.com>.
4823 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
4824 OK for addresses that are the result of redirections.
4826 PH/48 A number of further additions for the benefit of the new test suite,
4827 including a fake gethostbyname() that interfaces to the fake DNS resolver
4830 TF/06 The fix for widen_domains has also been applied to qualify_single and
4831 search_parents which are the other dnslookup options that can cause
4834 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
4837 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
4839 TF/07 Exim produced the error message "an SRV record indicated no SMTP
4840 service" if it encountered an MX record with an empty target hostname.
4841 The message is now "an MX or SRV record indicated no SMTP service".
4843 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
4844 if the DNS of the sending site is misconfigured. This is quite a
4845 common situation. This change restores the behaviour of treating a
4846 helo verification defer as a failure.
4848 PH/51 If self=fail was set on a router, the bounce message did not include the
4849 actual error message.
4855 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
4857 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
4858 command itself ended in error, the underlying message about the transport
4859 filter timeout was being overwritten with the pipe command error. Now the
4860 underlying error message should be appended to the second error message.
4862 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4864 PH/02 Reduce the amount of output that "make" produces by default. Full output
4865 can still be requested.
4867 PH/03 The warning log line about a condition test deferring for a "warn" verb
4868 was being output only once per connection, rather than after each
4869 occurrence (because it was using the same function as for successful
4870 "warn" verbs). This seems wrong, so I have changed it.
4872 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4873 it should not have, which might have caused a crash in the right
4874 circumstances, but probably never did.
4876 PH/04 Installed a modified version of Tony Finch's patch to make submission
4877 mode fix the return path as well as the Sender: header line, and to
4878 add a /name= option so that you can make the user's friendly name appear
4881 TF/03 Added the control = fakedefer ACL modifier.
4883 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4884 Mark Lowes for thorough testing.
4886 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4888 TK/03 Merged latest SRS patch from Miles Wilton.
4890 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4891 with the definition in sysexits.h (which is #included earlier).
4892 Fortunately, Exim does not actually use EX_OK. The code used to try to
4893 preserve the sysexits.h value, by assuming that macro definitions were
4894 scanned for macro replacements. I have been disabused of this notion,
4895 so now the code just undefines EX_OK before #including unistd.h.
4897 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4898 in the smtp transport. When a block could not be written in a single
4899 write() function, the timeout was being re-applied to each part-write.
4900 This seems wrong - if the receiver was accepting one byte at a time it
4901 would take for ever. The timeout is now adjusted when this happens. It
4902 doesn't have to be particularly precise.
4904 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4905 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4907 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4908 without a domain if the check to <postmaster@domain> fails.
4910 SC/01 Eximstats: added -xls and the ability to specify output files
4911 (patch written by Frank Heydlauf).
4913 SC/02 Eximstats: use FileHandles for outputting results.
4915 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4917 SC/04 Eximstats: fixed display of large numbers with -nvr option
4919 SC/05 Eximstats: fixed merging of reports with empty tables.
4921 SC/06 Eximstats: added the -include_original_destination flag
4923 SC/07 Eximstats: removed tabs and trailing whitespace.
4925 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4927 TK/06 MBOX spool code: Add real "From " MBOX separator line
4928 so the .eml file is really in mbox format (even though
4929 most programs do not really care). Patch from Alex Miller.
4931 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4932 The latter is generated from $received_to and is only set if the
4933 message has one envelope recipient. SA can use these headers,
4934 obviously out-of-the-box. Patch from Alex Miller.
4936 PH/08 The ${def test on a variable was returning false if the variable's
4937 value was "0", contrary to what the specification has always said!
4938 The result should be true unless the variable is empty.
4940 PH/09 The syntax error of a character other than { following "${if
4941 def:variable_name" (after optional whitespace) was not being diagnosed.
4942 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4943 accidental colon was present, for example, could give incorrect results.
4945 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4946 result is used (not including appendfile, where other changes are about
4949 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4950 This involved changing a lot of size variables from int to off_t. It
4951 should work with maildirs and everything.
4953 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4954 spamd dying while we are connected to it.
4956 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4959 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4960 function for BDB 4.3.
4962 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4964 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4965 the sources for PCRE have been re-arranged and more files are now
4968 PH/15 The code I had for printing potentially long long variables in PH/11
4969 above was not the best (it lost precision). The length of off_t variables
4970 is now inspected at build time, and an appropriate printing format (%ld
4971 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4972 to be "long long int" or "long int". This is needed for the internal
4973 formatting function string_vformat().
4975 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4976 the configuration file to be ":syslog", then the script "guesses" where
4977 the logs files are, rather than using the compiled in default. In our
4978 case the guess is not the same as the compiled default, so the script
4979 suddenly stopped working when I started to use syslog. The patch checks
4980 to see if log_file_path is "". If so, it attempts to read it from exim
4981 with no configuration file to get the compiled in version, before it
4982 falls back to the previous guessing code."
4984 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4985 implementing BATV in an Exim configuration. See NewStuff for the gory
4988 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4989 Makefile that are specific to HP-UX.
4991 PH/18 If the "use_postmaster" option was set for a recipient callout together
4992 with the "random" option, the postmaster address was used as the MAIL
4993 FROM address for the random test, but not for the subsequent recipient
4994 test. It is now used for both.
4996 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4997 patch removes a few documentation additions to RFC 3028, because the
4998 latest draft now contains them. It adds the new en;ascii-case comparator
4999 and a new error check for 8bit text in MIME parts. Comparator and
5000 require names are now matched exactly. I enabled the subaddress
5001 extension, but it is not well tested yet (read: it works for me)."
5003 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
5004 rework some of the code of TK/09 above to avoid the hardwired use of
5005 "%lld" and "long long". Replaced the call to snprintf() with a call to
5008 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
5009 records point to non-existent hosts", "retry timeout exceeded", and
5010 "retry time not reached for any host after a long failure period".
5012 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
5013 experimental DomainKeys support:
5015 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
5016 (2) On an error such as an illegally used "control", the wrong name for
5017 the control was given.
5019 These problems did NOT occur unless DomainKeys support was compiled.
5021 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
5023 PH/24 Added ${if match_ip condition.
5025 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
5026 databases so that it will be absolutely obvious if a crash occurs in the
5027 DB library. This is a regular occurrence (often caused by mis-matched
5030 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
5031 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
5032 picked up on a user's system that detects such things. There doesn't seem
5033 to be a gcc warning option for this - only an attribute that has to be
5034 put on the function's prototype. It seems that in Fedora Core 4 they have
5035 set this on a number of new functions. No doubt there will be more in due
5038 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
5039 specify a transport. However, if an address that was verified by such a
5040 router was the subject of a callout, Exim crashed because it tried to
5041 read the rcpt_include_affixes from the non-existent transport. Now it
5042 just assumes that the setting of that option is false. This bug was
5043 introduced by 4.51/PH/31.
5045 PH/28 Changed -d+all to exclude +memory, because that information is very
5046 rarely of interest, but it makes the output a lot bigger. People tend to
5047 do -d+all out of habit.
5049 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
5050 code in os-type was giving problems when libc.so lives in lib64, like on
5053 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
5054 aren't the modern standard, and it seems that some systems' include files
5055 don't always have them. Exim was already checking for some of the newer
5056 ones like T_AAAA, and defining it itself. I've added checks for all the
5057 record types that Exim uses.
5059 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
5060 not automatically generating a new one, as it is supposed to. This
5061 prevented TLS from working. If the file did exist, but contained invalid
5062 data, a new version was generated, as expected. It was only the case of a
5063 non-existent file that was broken.
5065 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
5066 with a change in libdomainkeys > 0.64.
5068 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
5069 from DNS. If the selector record carries the flag, it now has
5070 precedence over the domain-wide flag.
5072 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
5074 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
5075 the use of a port name that isn't defined in /etc/services) occurred, the
5076 message was deferred as in a normal delivery, and thus remained on the
5077 spool, instead of being failed because of the mua_wrapper setting. This
5078 is now fixed, and I tidied up some of the mua_wrapper messages at the
5081 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
5082 the messages in a hash of arrays rather than using individual hashes.
5083 This is a bit cleaner and results in dramatic memory savings, albeit
5084 at a slight CPU cost.
5086 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
5087 as requested by Marc Sherman.
5089 SC/10 Eximstats: added histograms for user specified patterns as requested
5092 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
5094 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
5095 fopen() in the content-scanning modules that did not already have it.
5101 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
5102 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
5104 TK/02 Fix ACL "control" statement not being available in MIME ACL.
5106 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
5108 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
5109 to test Sieve filters that use "vacation".
5111 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
5112 that changes the way the GnuTLS parameters are stored in the cache file.
5113 The new format can be generated externally. For backward compatibility,
5114 if the data in the cache doesn't make sense, Exim assumes it has read an
5115 old-format file, and it generates new data and writes a new file. This
5116 means that you can't go back to an older release without removing the
5119 PH/03 A redirect router that has both "unseen" and "one_time" set does not
5120 work if there are any delivery delays because "one_time" forces the
5121 parent to be marked "delivered", so its unseen clone is never tried
5122 again. For this reason, Exim now forbids the simultaneous setting of
5125 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
5126 redirected to themselves ("homonym" addresses). Read the long ChangeLog
5127 entry if you want to know the details. The fix, however, neglected to
5128 consider the case when local delivery batching is involved. The test for
5129 "previously delivered" was not happening when checking to see if an
5130 address could be batched with a previous (undelivered) one; under
5131 certain circumstances this could lead to multiple deliveries to the same
5134 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
5135 in its include files, and this causes problems building Exim.
5137 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
5138 header_syntax) but Exim was just ignoring anything given after a slash.
5139 In particular, this caused confusion with an attempt to use "verify =
5140 reverse_host_lookup/defer_ok". An error is now given when options are
5141 supplied for verify items that do not have them. (Maybe reverse_host_
5142 lookup should have a defer_ok option, but that's a different point.)
5144 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
5145 defined by RFC 821) to 2048, because there were problems with some AUTH
5146 commands, and RFC 1869 says the size should be increased for extended
5147 SMTP commands that take arguments.
5149 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
5152 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
5153 "unknown" error; now it says that the functionality isn't in the binary.
5155 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
5156 an address' error message when a string expansion fails (syntax or
5157 whatever). Otherwise the password may appear in the log. Following change
5158 PH/42 below, there is no longer a chance of it appearing in a bounce
5161 PH/11 Installed exipick version 20050225.0 from John Jetmore.
5163 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
5164 of its addresses was ever tried. (Bugzilla bug #2.)
5166 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
5167 the result incorrectly in the debug output. (It correctly added a newline
5168 to what was transported.)
5170 TF/01 Added $received_time.
5172 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
5173 commented out examples of how to interface to a virus scanner and to
5174 SpamAssassin. Also added commented examples of av_scanner and
5175 spamd_address settings.
5177 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
5178 and controls are allowed in which ACLs. There were a couple of minor
5179 errors. Some of the entries in the conditions table (which is a table of
5180 where they are NOT allowed) were getting very unwieldy; rewrote them as a
5181 negation of where the condition IS allowed.
5183 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
5185 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
5186 header file does not have a version number, so I've had to invent a new
5187 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
5188 API. The code is untested by me (my Linux distribution still has 0.3.2 of
5189 radiusclient), but it was contributed by a Radius user.
5191 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
5192 files or directories, for OpenSSL.
5194 PH/19 When an Exim process that is running as root has to create an Exim log
5195 file, it does so in a subprocess that runs as exim:exim so as to get the
5196 ownership right at creation (otherwise, other Exim processes might see
5197 the file with the wrong ownership). There was no test for failure of this
5198 fork() call, which would lead to the process getting stuck as it waited
5199 for a non-existent subprocess. Forks do occasionally fail when resources
5200 run out. I reviewed all the other calls to fork(); they all seem to check
5203 PH/20 When checking for unexpected SMTP input at connect time (before writing
5204 the banner), Exim was not dealing correctly with a non-positive return
5205 from the read() function. If the client had disconnected by this time,
5206 the result was a log entry for a synchronization error with an empty
5207 string after "input=" when read() returned zero. If read() returned -1
5208 (an event I could not check), uninitialized data bytes were printed.
5209 There were reports of junk text (parts of files, etc) appearing after
5212 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
5214 PH/22 Added support for macro redefinition, and (re)definition in between
5215 driver and ACL definitions.
5217 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
5218 forgetting to use the resulting value; it was using the unexpanded value.
5220 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
5221 hadn't been configured. The fix is from Juergen Kreileder, who
5222 understands it better than I do:
5224 "Here's what I see happening with three configured cyrus_sasl
5225 authenticators configured (plain, login, cram-md5):
5227 On startup auth_cyrus_sasl_init() gets called for each of these.
5228 This means three calls to sasl_listmech() without a specified mech_list.
5229 => SASL tests which mechs of all available mechs actually work
5230 => three warnings about OTP not working
5231 => the returned list contains: plain, login, cram-md5, digest-md5, ...
5233 With the patch, sasl_listmech() also gets called three times. But now
5234 SASL's mech_list option is set to the server_mech specified in the the
5235 authenticator. Or in other words, the answer from sasl_listmech()
5236 gets limited to just the mech you're testing for (which is different
5238 => the return list contains just 'plain' or 'login', 'cram-md5' or
5239 nothing depending on the value of ob->server_mech.
5241 I've just tested the patch: Authentication still works fine,
5242 unavailable mechs specified in the exim configuration are still
5243 caught, and the auth.log warnings about OTP are gone."
5245 PH/25 When debugging is enabled, the contents of the command line are added
5246 to the debugging output, even when log_selector=+arguments is not
5249 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
5250 answer is "GNU", and only if the return is "GNU/something" is the answer
5253 PH/27 $acl_verify_message is now set immediately after the failure of a
5254 verification in an ACL, and so is available in subsequent modifiers. In
5255 particular, the message can be preserved by coding like this:
5257 warn !verify = sender
5258 set acl_m0 = $acl_verify_message
5260 Previously, $acl_verify_message was set only while expanding "message"
5261 and "log_message" when a very denied access.
5263 PH/28 Modified OS/os.c-Linux with
5265 -#ifndef OS_LOAD_AVERAGE
5266 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
5268 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
5269 nomenclature these days.)
5271 PH/29 Installed patch from the Sieve maintainer that adds the options
5272 sieve_useraddress and sieve_subaddress to the redirect router.
5274 PH/30 In these circumstances:
5275 . Two addresses routed to the same list of hosts;
5276 . First host does not offer TLS;
5277 . First host accepts first address;
5278 . First host gives temporary error to second address;
5279 . Second host offers TLS and a TLS session is established;
5280 . Second host accepts second address.
5281 Exim incorrectly logged both deliveries with the TLS parameters (cipher
5282 and peerdn, if requested) that were in fact used only for the second
5285 PH/31 When doing a callout as part of verifying an address, Exim was not paying
5286 attention to any local part prefix or suffix that was matched by the
5287 router that accepted the address. It now behaves in the same way as it
5288 does for delivery: the affixes are removed from the local part unless
5289 rcpt_include_affixes is set on the transport.
5291 PH/32 Add the sender address, as F=<...>, to the log line when logging a
5292 timeout during the DATA phase of an incoming message.
5294 PH/33 Sieve envelope tests were broken for match types other than :is. I have
5295 applied a patch sanctioned by the Sieve maintainer.
5297 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
5298 the uid or gid is negative. A case of a negative gid caused this to be
5299 noticed. The fix allows for either to be negative.
5301 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
5302 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
5305 PH/36 The change PH/12 above was broken. Fixed it.
5307 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
5308 the grounds that routing the same address twice would always produce the
5309 same answer. This might have been true once, but it is certainly no
5310 longer true now. Routing a child address may depend on the previous
5311 routing that produced that child. Some complicated redirection strategies
5312 went wrong when messages had multiple recipients, and made Exim's
5313 behaviour dependent on the order in which the addresses were given.
5315 I have moved the duplicate checking until after the routing is complete.
5316 Exim scans the addresses that are assigned to local and remote
5317 transports, and removes any duplicates. This means that more work will be
5318 done, as duplicates will always all be routed, but duplicates are
5319 presumably rare, so I don't expect this is of any significance.
5321 For deliveries to pipes, files, and autoreplies, the duplicate checking
5322 still happens during the routing process, since they are not going to be
5325 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
5326 It corrects a timeout issue with spamd. This is Ian's comment: "The
5327 background is that sometimes spamd either never reads data from a
5328 connection it has accepted, or it never writes response data. The exiscan
5329 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
5330 blindly assumes that writes won't block so it may never time out."
5332 PH/39 Allow G after quota size as well as K and M.
5334 PH/40 The value set for $authenticated_id in an authenticator may not contain
5335 binary zeroes or newlines because the value is written to log lines and
5336 to spool files. There was no check on this. Now the value is run through
5337 the string_printing() function so that such characters are converted to
5338 printable escape sequences.
5340 PH/41 $message_linecount is a new variable that contains the total number of
5341 lines in the message. Compare $body_linecount, which is the count for the
5344 PH/42 Exim no longer gives details of delivery errors for specific addresses in
5345 bounce and delay warning messages, except in certain special cases, which
5348 (a) An SMTP error message from a remote host;
5349 (b) A message specified in a :fail: redirection;
5350 (c) A message specified in a "fail" command in a system filter;
5351 (d) A message specified in a FAIL return from the queryprogram router;
5352 (e) A message specified by the cannot_route_message router option.
5354 In these cases only, Exim does include the error details in bounce and
5355 warning messages. There are also a few cases where bland messages such
5356 as "unrouteable address" or "local delivery error" are given.
5358 PH/43 $value is now also set for the "else" part of a ${run expansion.
5360 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
5361 being worked on, but at least Exim now implements the latest version to
5364 PH/45 In a pipe transport, although a timeout while waiting for the pipe
5365 process to complete was treated as a delivery failure, a timeout while
5366 writing the message to the pipe was logged, but erroneously treated as a
5367 successful delivery. Such timeouts include transport filter timeouts. For
5368 consistency with the overall process timeout, these timeouts are now
5369 treated as errors, giving rise to delivery failures by default. However,
5370 there is now a new Boolean option for the pipe transport called
5371 timeout_defer, which, if set TRUE, converts the failures into defers for
5372 both kinds of timeout. A transport filter timeout is now identified in
5375 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
5376 systems where "make" and "gmake" are different, calling "gmake" at top
5377 level broke things. I've arranged for the value of $(MAKE) to be passed
5378 from the Makefile to this script so that it can call the same version of
5382 A note about Exim versions 4.44 and 4.50
5383 ----------------------------------------
5385 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
5386 changes of various kinds. As a consequence, a big documentation update was
5387 needed. This delayed the release for rather longer than seemed good, especially
5388 in the light of a couple of (minor) security issues. Therefore, the changes
5389 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
5390 release. So 4.44 and 4.50 are in effect two different branches that both start
5393 I have left the 4.50 change log unchanged; it contains all the changes since
5394 4.43. The change log for 4.44 is below; many of its items are identical to
5395 those for 4.50. This seems to be the most sensible way to preserve the
5396 historical information.
5402 1. Minor wording change to the doc/README.SIEVE file.
5404 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
5405 computation of the current number of files was incorrect.
5407 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5408 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5411 4. Give more explanation in the error message when the command for a transport
5412 filter fails to execute.
5414 5. There are several places where Exim runs a non-Exim command in a
5415 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5416 was being done only for the command run by the queryprogram router. It is
5417 now done for all such subprocesses. The other cases are: ${run, transport
5418 filters, and the commands run by the lmtp and pipe transports.
5420 6. Added CONFIGURE_GROUP build-time option.
5422 7. Some older OS have a limit of 256 on the maximum number of file
5423 descriptors. Exim was using setrlimit() to set 1000 as a large value
5424 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5425 systems. I've change it so that if it can't get 1000, it tries for 256.
5427 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5428 was an oversight, and furthermore, ever since the addition of extra
5429 controls (e.g. 4.43/32), the checks on when to allow different forms of
5430 "control" were broken. There should now be diagnostics for all cases when a
5431 control that does not make sense is encountered.
5433 9. Added the /retain_sender option to "control=submission".
5435 10. $recipients is now available in the predata ACL (oversight).
5437 11. Tidy the search cache before the fork to do a delivery from a message
5438 received from the command line. Otherwise the child will trigger a lookup
5439 failure and thereby defer the delivery if it tries to use (for example) a
5440 cached ldap connection that the parent has called unbind on.
5442 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5443 of $address_data from the recipient verification was clobbered by the
5444 sender verification.
5446 13. The value of address_data from a sender verification is now available in
5447 $sender_address_data in subsequent conditions in the ACL statement.
5449 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
5451 15. Added a new option "connect=<time>" to callout options, to set a different
5454 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5455 was its contents. (It was OK if the option was not defined at all.)
5457 17. A "Completed" log line is now written for messages that are removed from
5458 the spool by the -Mrm option.
5460 18. New variables $sender_verify_failure and $recipient_verify_failure contain
5461 information about exactly what failed.
5463 19. Added -dd to debug only the daemon process.
5465 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
5466 handles timeouts, both on the server side and network timeouts. Renamed the
5467 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
5469 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
5470 It is now set to "smtps".
5472 22. $host_address is now set to the target address during the checking of
5473 ignore_target_hosts.
5475 23. When checking ignore_target_hosts for an ipliteral router, no host name was
5476 being passed; this would have caused $sender_host_name to have been used if
5477 matching the list had actually called for a host name (not very likely,
5478 since this list is usually IP addresses). A host name is now passed as
5481 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5482 code that specifies a non-restarting handler (typically sigaction() in
5483 modern systems) in an attempt to fix a rare and obscure crash bug.
5485 25. Narrowed the window for a race in the daemon that could cause it to ignore
5486 SIGCHLD signals. This is not a major problem, because they are used only to
5487 wake it up if nothing else does.
5489 26. A malformed maildirsize file could cause Exim to calculate negative values
5490 for the mailbox size or file count. Odd effects could occur as a result.
5491 The maildirsize information is now recalculated if the size or filecount
5494 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5495 support for a long time. Removed HAVE_SYS_VFS_H.
5497 28. Installed the latest version of exipick from John Jetmore.
5499 29. In an address list, if the pattern was not a regular expression, an empty
5500 subject address (from a bounce message) matched only if the pattern was an
5501 empty string. Non-empty patterns were not even tested. This was the wrong
5502 because it is perfectly reasonable to use an empty address as part of a
5503 database query. An empty address is now tested by patterns that are
5504 lookups. However, all the other forms of pattern expect the subject to
5505 contain a local part and a domain, and therefore, for them, an empty
5506 address still always fails if the pattern is not itself empty.
5508 30. Exim went into a mad DNS loop when attempting to do a callout where the
5509 host was specified on an smtp transport, and looking it up yielded more
5510 than one IP address.
5512 31. Re-factored the code for checking spool and log partition space into a
5513 function that finds that data and another that does the check. The former
5514 is then used to implement four new variables: $spool_space, $log_space,
5515 $spool_inodes, and $log_inodes.
5517 32. The RFC2047 encoding function was originally intended for short strings
5518 such as real names; it was not keeping to the 75-character limit for
5519 encoded words that the RFC imposes. It now respects the limit, and
5520 generates multiple encoded words if necessary. To be on the safe side, I
5521 have increased the buffer size for the ${rfc2047: expansion operator from
5524 33. It is now permitted to omit both strings after an "if" condition; if the
5525 condition is true, the result is "true". As before, when the second string
5526 is omitted, a false condition yields an empty string. This makes it less
5527 cumbersome to write custom ACL and router conditions.
5529 34. Failure to deliver a bounce message always caused it to be frozen, even if
5530 there was an errors_to setting on the router. The errors_to setting is now
5533 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5534 canonical form (fully expanded) before being placed in
5535 $sender_host_address.
5537 36. The table in the code that translates DNS record types into text (T_A to
5538 "A" for instance) was missing entries for NS and CNAME. It is just possible
5539 that this could have caused confusion if both these types were looked up
5540 for the same domain, because the text type is used as part of Exim's
5541 per-process caching. But the chance of anyone hitting this buglet seems
5544 37. The dnsdb lookup has been extended in a number of ways.
5546 (1) There is a new type, "zns", which walks up the domain tree until it
5547 finds some nameserver records. It should be used with care.
5549 (2) There is a new type, "mxh", which is like "mx" except that it returns
5550 just the host names, not the priorities.
5552 (3) It is now possible to give a list of domains (or IP addresses) to be
5553 looked up. The behaviour when one of the lookups defers can be
5554 controlled by a keyword.
5556 (4) It is now possible to specify the separator character for use when
5557 multiple records are returned.
5559 38. The dnslists ACL condition has been extended: it is now possible to supply
5560 a list of IP addresses and/or domains to be looked up in a particular DNS
5563 39. Added log_selector=+queue_time_overall.
5565 40. When running the queue in the test harness, wait just a tad after forking a
5566 delivery process, to get repeatability of debugging output.
5568 41. Include certificate and key file names in error message when GnuTLS fails
5569 to set them up, because the GnuTLS error message doesn't include the name
5570 of the failing file when there is a problem reading it.
5572 42. Allow both -bf and -bF in the same test run.
5574 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
5576 44. The "Exiscan patch" is now merged into the mainline Exim source.
5578 45. Sometimes the final signoff response after QUIT could fail to get
5579 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5580 < 0 before doing a fflush(). This bug looks as though it goes back to the
5581 introduction of TLS in release 3.20, but "sometimes" must have been rare
5582 because the tests only now provoked it.
5584 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
5585 (this can affect the format of dates).
5587 47. exim_tidydb, when checking for the continued existence of a message for
5588 which it has found a message-specific retry record, was not finding
5589 messages that were in split spool directories. Consequently, it was
5590 deleting retry records that should have stayed in existence.
5592 48. Steve fixed some bugs in eximstats.
5594 49. The SPA authentication driver was not abandoning authentication and moving
5595 on to the next authenticator when an expansion was forced to fail,
5596 contradicting the general specification for all authenticators. Instead it
5597 was generating a temporary error. It now behaves as specified.
5599 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
5600 (the order specifies the preference for clients). The order is now AES256,
5601 AES128, 3DES, ARCFOUR128.
5603 51. Small patch to Sieve code - explicitly set From: when generating an
5606 52. Exim crashed if a remote delivery caused a very long error message to be
5607 recorded - for instance if somebody sent an entire SpamAssassin report back
5608 as a large number of 550 error lines. This bug was coincidentally fixed by
5609 increasing the size of one of Exim's internal buffers (big_buffer) that
5610 happened as part of the Exiscan merge. However, to be on the safe side, I
5611 have made the code more robust (and fixed the comments that describe what
5614 53. Now that there can be additional text after "Completed" in log lines (if
5615 the queue_time_overall log selector is set), a one-byte patch to exigrep
5616 was needed to allow it to recognize "Completed" as not the last thing in
5619 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
5620 patch that reportedly fixes this has been added. I am not expert enough to
5621 create a test for it. This is what the patch creator wrote:
5623 "I found a little strange behaviour of ldap code when working with
5624 Windows 2003 AD Domain, where users was placed in more than one
5625 Organization Units. When I tried to give exim partial DN, the exit code
5626 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
5627 But simultaneously result of request was absolutely normal ldap result,
5628 so I produce this patch..."
5630 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
5631 REFERENCE, so I have modified the code to exclude the patch when that macro
5634 55. Some experimental protocols are using DNS PTR records for new purposes. The
5635 keys for these records are domain names, not reversed IP addresses. The
5636 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5637 leaves it alone. Component reversal etc. now happens only for IP addresses.
5640 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
5642 57. Double the size of the debug message buffer (to 2048) so that more of very
5643 long debug lines gets shown.
5645 58. The exicyclog utility now does better if the number of log files to keep
5646 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
5648 59. Two changes related to the smtp_active_hostname option:
5650 (1) $smtp_active_hostname is now available as a variable.
5651 (2) The default for smtp_banner uses $smtp_active_hostname instead
5652 of $primary_hostname.
5654 60. The host_aton() function is supposed to be passed a string that is known
5655 to be a valid IP address. However, in the case of IPv6 addresses, it was
5656 not checking this. This is a hostage to fortune. Exim now panics and dies
5657 if the condition is not met. A case was found where this could be provoked
5658 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5659 components; fortuitously, this particular loophole had already been fixed
5660 by change 4.50/55 above.
5662 If there are any other similar loopholes, the new check in host_aton()
5663 itself should stop them being exploited. The report I received stated that
5664 data on the command line could provoke the exploit when Exim was running as
5665 exim, but did not say which command line option was involved. All I could
5666 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5667 running as the user.
5670 61. There was a buffer overflow vulnerability in the SPA authentication code
5671 (which came originally from the Samba project). I have added a test to the
5672 spa_base64_to_bits() function which I hope fixes it.
5675 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
5676 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
5678 63. The daemon start-up calls getloadavg() while still root for those OS that
5679 need the first call to be done as root, but it missed one case: when
5680 deliver_queue_load_max is set with deliver_drop_privilege. This is
5681 necessary for the benefit of the queue runner, because there is no re-exec
5682 when deliver_drop_privilege is set.
5684 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
5685 This has been fixed.
5687 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5688 was in use, was not putting the data itself into the right store pool;
5689 consequently, it could be overwritten for a subsequent message in the same
5690 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5693 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
5695 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
5696 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
5697 the function were treating the return as a boolean value, which happened to
5698 work because 0=false and not-0=true, but is not correct code.
5700 68. The host_aton() function was not handling scoped IPv6 addresses (those
5701 with, for example, "%eth0" on the end) correctly.
5703 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
5704 negated items (that is, ~something) in unsigned ints. Some compilers
5705 apparently mutter when there is no cast.
5707 70. If an address verification called from an ACL failed, and did not produce a
5708 user-specific message (i.e. there was only a "system" message), nothing was
5709 put in $acl_verify_message. In this situation, it now puts the system
5712 71. Change 4.23/11 added synchronization checking at the start of an SMTP
5713 session; change 4.31/43 added the unwanted input to the log line - except
5714 that it did not do this in the start of session case. It now does.
5716 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
5717 This is wrong and can cause the other end to generate a synchronization
5718 error if it is another Exim or anything else that does the synchronization
5719 check. A QUIT command is no longer sent after a timeout.
5721 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
5722 during host lookups.
5724 74. The defer_ok option of callout verification was not working if it was used
5725 when verifying addresses in header lines, that is, for this case:
5727 verify = header_sender/callout=defer_ok
5729 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
5730 those file descriptors could be used for SMTP connections. If anything
5731 wrote to stderr (the example that came up was "warn" in embedded Perl), it
5732 could be sent to the SMTP client, causing chaos. The daemon now opens
5733 stdin, stdout, and stderr to /dev/null when it puts itself into the
5736 76. Arrange for output from Perl's "warn" command to be written to Exim's main
5737 log by default. The user can override this with suitable Perl magic.
5739 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
5740 the log message when discard triggers, was not working for the DATA ACL or
5741 for the non-SMTP ACL.
5743 78. Error message wording change in sieve.c.
5745 79. If smtp_accept_max_per_host was set, the number of connections could be
5746 restricted to fewer than expected, because the daemon was trying to set up
5747 a new connection before checking whether the processes handling previous
5748 connections had finished. The check for completed processes is now done
5749 earlier. On busy systems, this bug wouldn't be noticed because something
5750 else would have woken the daemon, and it would have reaped the completed
5753 80. If a message was submitted locally by a user whose login name contained one
5754 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
5755 It caused a spool format error. I have fixed the spool reading code. A
5756 related problem was that the "from" clause in the Received: line became
5757 illegal because of the space(s). It is now covered by ${quote_local_part.
5759 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
5762 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
5764 83. Patch from Sieve maintainer to fix unterminated string problem in
5765 "vacation" handling.
5767 84. Some minor changes to the Linux configuration files to help with other
5768 OS variants using glibc.
5770 85. One more patch for Sieve to update vacation handling to latest spec.
5773 ----------------------------------------------------
5774 See the note above about the 4.44 and 4.50 releases.
5775 ----------------------------------------------------
5781 1. Change 4.43/35 introduced a bug that caused file counts to be
5782 incorrectly computed when quota_filecount was set in an appendfile
5785 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5786 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5789 3. Give more explanation in the error message when the command for a transport
5790 filter fails to execute.
5792 4. There are several places where Exim runs a non-Exim command in a
5793 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5794 was being done only for the command run by the queryprogram router. It is
5795 now done for all such subprocesses. The other cases are: ${run, transport
5796 filters, and the commands run by the lmtp and pipe transports.
5798 5. Some older OS have a limit of 256 on the maximum number of file
5799 descriptors. Exim was using setrlimit() to set 1000 as a large value
5800 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5801 systems. I've change it so that if it can't get 1000, it tries for 256.
5803 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5804 was an oversight, and furthermore, ever since the addition of extra
5805 controls (e.g. 4.43/32), the checks on when to allow different forms of
5806 "control" were broken. There should now be diagnostics for all cases when a
5807 control that does not make sense is encountered.
5809 7. $recipients is now available in the predata ACL (oversight).
5811 8. Tidy the search cache before the fork to do a delivery from a message
5812 received from the command line. Otherwise the child will trigger a lookup
5813 failure and thereby defer the delivery if it tries to use (for example) a
5814 cached ldap connection that the parent has called unbind on.
5816 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5817 of $address_data from the recipient verification was clobbered by the
5818 sender verification.
5820 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5821 was its contents. (It was OK if the option was not defined at all.)
5823 11. A "Completed" log line is now written for messages that are removed from
5824 the spool by the -Mrm option.
5826 12. $host_address is now set to the target address during the checking of
5827 ignore_target_hosts.
5829 13. When checking ignore_target_hosts for an ipliteral router, no host name was
5830 being passed; this would have caused $sender_host_name to have been used if
5831 matching the list had actually called for a host name (not very likely,
5832 since this list is usually IP addresses). A host name is now passed as
5835 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5836 code that specifies a non-restarting handler (typically sigaction() in
5837 modern systems) in an attempt to fix a rare and obscure crash bug.
5839 15. Narrowed the window for a race in the daemon that could cause it to ignore
5840 SIGCHLD signals. This is not a major problem, because they are used only to
5841 wake it up if nothing else does.
5843 16. A malformed maildirsize file could cause Exim to calculate negative values
5844 for the mailbox size or file count. Odd effects could occur as a result.
5845 The maildirsize information is now recalculated if the size or filecount
5848 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5849 support for a long time. Removed HAVE_SYS_VFS_H.
5851 18. Updated exipick to current release from John Jetmore.
5853 19. Allow an empty sender to be matched against a lookup in an address list.
5854 Previously the only cases considered were a regular expression, or an
5857 20. Exim went into a mad DNS lookup loop when doing a callout where the
5858 host was specified on the transport, if the DNS lookup yielded more than
5861 21. The RFC2047 encoding function was originally intended for short strings
5862 such as real names; it was not keeping to the 75-character limit for
5863 encoded words that the RFC imposes. It now respects the limit, and
5864 generates multiple encoded words if necessary. To be on the safe side, I
5865 have increased the buffer size for the ${rfc2047: expansion operator from
5868 22. Failure to deliver a bounce message always caused it to be frozen, even if
5869 there was an errors_to setting on the router. The errors_to setting is now
5872 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5873 canonical form (fully expanded) before being placed in
5874 $sender_host_address.
5876 24. Updated eximstats to version 1.33
5878 25. Include certificate and key file names in error message when GnuTLS fails
5879 to set them up, because the GnuTLS error message doesn't include the name
5880 of the failing file when there is a problem reading it.
5882 26. Expand error message when OpenSSL has problems setting up cert/key files.
5885 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5886 (this can affect the format of dates).
5888 28. exim_tidydb, when checking for the continued existence of a message for
5889 which it has found a message-specific retry record, was not finding
5890 messages that were in split spool directories. Consequently, it was
5891 deleting retry records that should have stayed in existence.
5893 29. eximstats updated to version 1.35
5894 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5895 1.35 - bugfix such that pie charts by volume are generated correctly
5897 30. The SPA authentication driver was not abandoning authentication and moving
5898 on to the next authenticator when an expansion was forced to fail,
5899 contradicting the general specification for all authenticators. Instead it
5900 was generating a temporary error. It now behaves as specified.
5902 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5903 (the order specifies the preference for clients). The order is now AES256,
5904 AES128, 3DES, ARCFOUR128.
5906 31. Small patch to Sieve code - explicitly set From: when generating an
5909 32. Exim crashed if a remote delivery caused a very long error message to be
5910 recorded - for instance if somebody sent an entire SpamAssassin report back
5911 as a large number of 550 error lines. This bug was coincidentally fixed by
5912 increasing the size of one of Exim's internal buffers (big_buffer) that
5913 happened as part of the Exiscan merge. However, to be on the safe side, I
5914 have made the code more robust (and fixed the comments that describe what
5917 33. Some experimental protocols are using DNS PTR records for new purposes. The
5918 keys for these records are domain names, not reversed IP addresses. The
5919 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5920 leaves it alone. Component reversal etc. now happens only for IP addresses.
5923 34. The host_aton() function is supposed to be passed a string that is known
5924 to be a valid IP address. However, in the case of IPv6 addresses, it was
5925 not checking this. This is a hostage to fortune. Exim now panics and dies
5926 if the condition is not met. A case was found where this could be provoked
5927 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5928 components; fortuitously, this particular loophole had already been fixed
5929 by change 4.50/55 or 4.44/33 above.
5931 If there are any other similar loopholes, the new check in host_aton()
5932 itself should stop them being exploited. The report I received stated that
5933 data on the command line could provoke the exploit when Exim was running as
5934 exim, but did not say which command line option was involved. All I could
5935 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5936 running as the user.
5939 35. There was a buffer overflow vulnerability in the SPA authentication code
5940 (which came originally from the Samba project). I have added a test to the
5941 spa_base64_to_bits() function which I hope fixes it.
5944 36. The daemon start-up calls getloadavg() while still root for those OS that
5945 need the first call to be done as root, but it missed one case: when
5946 deliver_queue_load_max is set with deliver_drop_privilege. This is
5947 necessary for the benefit of the queue runner, because there is no re-exec
5948 when deliver_drop_privilege is set.
5950 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5951 was in use, was not putting the data itself into the right store pool;
5952 consequently, it could be overwritten for a subsequent message in the same
5953 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5956 38. Sometimes the final signoff response after QUIT could fail to get
5957 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5958 < 0 before doing a fflush(). This bug looks as though it goes back to the
5959 introduction of TLS in release 3.20, but "sometimes" must have been rare
5960 because the tests only now provoked it.
5966 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5967 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5968 Now it returns an integer. A number of calls were still expecting a T/F
5969 return. Fortuitously, in all cases, the tests worked in OK situations,
5970 which is the norm. However, things would have gone wrong on any write
5971 failures on the smtp file descriptor. This function is used when sending
5972 messages over SMTP and also when doing verify callouts.
5974 2. When Exim is called to do synchronous delivery of a locally submitted
5975 message (the -odf or -odi options), it no longer closes stderr before doing
5978 3. Implemented the mua_wrapper option.
5980 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5982 5. Implemented the functions header_remove(), header_testname(),
5983 header_add_at_position(), and receive_remove_recipient(), and exported them
5986 6. If an ACL "warn" statement specified the addition of headers, Exim already
5987 inserted X-ACL-Warn: at the start if there was no header name. However, it
5988 was not making this test for the second and subsequent header lines if
5989 there were newlines in the string. This meant that an invalid header could
5990 be inserted if Exim was badly configured.
5992 7. Allow an ACL "warn" statement to add header lines at the start or after all
5993 the Received: headers, as well as at the end.
5995 8. Added the rcpt_4xx retry error code.
5997 9. Added postmaster_mailfrom=xxx to callout verification option.
5999 10. Added mailfrom=xxxx to the callout verification option, for verify=
6002 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
6003 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
6005 12. Inserted some casts to stop certain compilers warning when using pointer
6006 differences as field lengths or precisions in printf-type calls (mostly
6007 affecting debugging statements).
6009 13. Added optional readline() support for -be (dynamically loaded).
6011 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
6012 same clock tick as a message's arrival, so that its received time was the
6013 same as the "first fail" time on the retry record, and that message
6014 remained on the queue past the ultimate address timeout, every queue runner
6015 would try a delivery (because it was past the ultimate address timeout) but
6016 after another failure, the ultimate address timeout, which should have then
6017 bounced the address, did not kick in. This was a "< instead of <=" error;
6018 in most cases the first failure would have been in the next clock tick
6019 after the received time, and all would be well.
6021 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
6022 being recognized when the domain list was tested by the match_domain
6023 condition in an expansion string.
6025 16. Added the ${str2b64: operator.
6027 17. Exim was always calling setrlimit() to set a large limit for the number of
6028 processes, without checking whether the existing limit was already
6029 adequate. (It did check for the limit on file descriptors.) Furthermore,
6030 errors from getrlimit() and setrlimit() were being ignored. Now they are
6031 logged to the main and panic logs, but Exim does carry on, to try to do its
6032 job under whatever limits there are.
6034 18. Imported PCRE 5.0.
6036 19. Trivial typo in log message " temporarily refused connection" (the leading
6039 20. If the log selector return_path_on_delivery was set and an address was
6040 redirected to /dev/null, the delivery process crashed because it assumed
6041 that a return path would always be set for a "successful" delivery. In this
6042 case, the whole delivery is bypassed as an optimization, and therefore no
6045 21. Internal re-arrangement: the function for sending a challenge and reading
6046 a response while authentication was assuming a zero-terminated challenge
6047 string. It's now changed to take a pointer and a length, to allow for
6048 binary data in such strings.
6050 22. Added the cyrus_sasl authenticator (code supplied by MBM).
6052 23. Exim was not respecting finduser_retries when seeking the login of the
6053 uid under which it was called; it was always trying 10 times. (The default
6054 setting of finduser_retries is zero.) Also, it was sleeping after the final
6055 failure, which is pointless.
6057 24. Implemented tls_on_connect_ports.
6059 25. Implemented acl_smtp_predata.
6061 26. If the domain in control=submission is set empty, Exim assumes that the
6062 authenticated id is a complete email address when it generates From: or
6063 Sender: header lines.
6065 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
6066 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
6067 chown and chgrp in /bin and hostname in /usr/bin.
6069 28. Exim was keeping the "process log" file open after each use, just as it
6070 does for the main log. This opens the possibility of it remaining open for
6071 long periods when the USR1 signal hits a daemon. Occasional processlog
6072 errors were reported, that could have been caused by this. Anyway, it seems
6073 much more sensible not to leave this file open at all, so that is what now
6076 29. The long-running daemon process does not normally write to the log once it
6077 has entered its main loop, and it closes the log before doing so. This is
6078 so that log files can straightforwardly be renamed and moved. However,
6079 there are a couple of unusual error situations where the daemon does write
6080 log entries, and I had neglected to close the log afterwards.
6082 30. The text of an SMTP error response that was received during a remote
6083 delivery was being truncated at 512 bytes. This is too short for some of
6084 the long messages that one sometimes sees. I've increased the limit to
6087 31. It is now possible to make retry rules that apply only when a message has a
6088 specific sender, in particular, an empty sender.
6090 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
6091 it possible to be selective about when SMTP synchronization is enforced.
6093 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
6095 32. Implemented hosts_connection_nolog.
6097 33. Added an ACL for QUIT.
6099 34. Setting "delay_warning=" to disable warnings was not working; it gave a
6102 35. Added mailbox_size and mailbox_filecount to appendfile.
6104 36. Added control = no_multiline_responses to ACLs.
6106 37. There was a bug in the logic of the code that waits for the clock to tick
6107 in the case where the clock went backwards by a substantial amount such
6108 that the microsecond fraction of "now" was more than the microsecond
6109 fraction of "then" (but the whole seconds number was less).
6111 38. Added support for the libradius Radius client library this is found on
6112 FreeBSD (previously only the radiusclient library was supported).
6118 1. When certain lookups returned multiple values in the form name=value, the
6119 quoting of the values was not always being done properly. Specifically:
6120 (a) If the value started with a double quote, but contained no whitespace,
6122 (b) If the value contained whitespace other than a space character (i.e.
6123 tabs or newlines or carriage returns) it was not quoted.
6124 This fix has been applied to the mysql and pgsql lookups by writing a
6125 separate quoting function and calling it from the lookup code. The fix
6126 should probably also be applied to nisplus, ibase and oracle lookups, but
6127 since I cannot test any of those, I have not disturbed their existing code.
6129 2. A hit in the callout cache for a specific address caused a log line with no
6130 reason for rejecting RCPT. Now it says "Previous (cached) callout
6131 verification failure".
6133 3. There was an off-by-one bug in the queryprogram router. An over-long
6134 return line was truncated at 256 instead of 255 characters, thereby
6135 overflowing its buffer with the terminating zero. As well as fixing this, I
6136 have increased the buffer size to 1024 (and made a note to document this).
6138 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
6139 when Exim is waiting for an SMTP response from a remote server, Exim
6140 restarts its select() call on the socket, thereby resetting its timeout.
6141 This is not a problem when such interrupts are rare. Somebody set up a cron
6142 job to run exiwhat every 2 minutes, which is less than the normal select()
6143 timeout (5 or 10 minutes). This meant that the select() timeout never
6144 kicked in because it was always reset. I have fixed this by comparing the
6145 time when an interrupt arrives with the time at the start of the first call
6146 to select(). If more time than the timeout has elapsed, the interrupt is
6147 treated as a timeout.
6149 5. Some internal re-factoring in preparation for the addition of Sieve
6150 extensions (by MH). In particular, the "personal" test is moved to a
6151 separate function, and given an option for scanning Cc: and Bcc: (which is
6152 not set for Exim filters).
6154 6. When Exim created an email address using the login of the caller as the
6155 local part (e.g. when creating a From: or Sender: header line), it was not
6156 quoting the local part when it contained special characters such as @.
6158 7. Installed new OpenBSD configuration files.
6160 8. Reworded some messages for syntax errors in "and" and "or" conditions to
6161 try to make them clearer.
6163 9. Callout options, other than the timeout value, were being ignored when
6164 verifying sender addresses in header lines. For example, when using
6166 verify = header_sender/callout=no_cache
6168 the cache was (incorrectly) being used.
6170 10. Added a missing instance of ${EXE} to the exim_install script; this affects
6171 only the Cygwin environment.
6173 11. When return_path_on_delivery was set as a log selector, if different remote
6174 addresses in the same message used different return paths and parallel
6175 remote delivery occurred, the wrong values would sometimes be logged.
6176 (Whenever a remote delivery process finished, the return path value from
6177 the most recently started remote delivery process was logged.)
6179 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
6180 lines when AUTH and/or TLS are in use. This is the "received protocol"
6181 field. Exim used to use "asmtp" for authenticated SMTP, without any
6182 indication (in the protocol name) for TLS use. Now it follows the RFC and
6183 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
6184 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
6185 names appear in log lines as well as in Received: header lines.
6187 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
6188 extensions, and comparison tests, and to fix some bugs.
6190 14. Changes to the "personal" filter test:
6192 (1) The test was buggy in that it was just doing the equivalent of
6193 "contains" tests on header lines. For example, if a user's address was
6194 anne@some.where, the "personal" test would incorrectly be true for
6196 To: susanne@some.where
6198 This test is now done by extracting each address from the header in turn,
6199 and checking the entire address. Other tests that are part of "personal"
6200 are now done using regular expressions (for example, to check local parts
6201 of addresses in From: header lines).
6203 (2) The list of non-personal local parts in From: addresses has been
6204 extended to include "listserv", "majordomo", "*-request", and "owner-*",
6205 taken from the Sieve specification recommendations.
6207 (3) If the message contains any header line starting with "List-" it is
6208 treated as non-personal.
6210 (4) The test for "circular" in the Subject: header line has been removed
6211 because it now seems ill-conceived.
6213 15. Minor typos in src/EDITME comments corrected.
6215 16. Installed latest exipick from John Jetmore.
6217 17. If headers_add on a router specified a text string that was too long for
6218 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
6219 of string_sprintf() is now avoided.
6221 18. $message_body_size was not set (it was always zero) when running the DATA
6222 ACL and the local_scan() function.
6224 19. For the "mail" command in an Exim filter, no default was being set for
6225 the once_repeat time, causing a random time value to be used if "once" was
6226 specified. (If the value happened to be <= 0, no repeat happened.) The
6227 default is now 0s, meaning "never repeat". The "vacation" command was OK
6228 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
6229 (I found it when inspecting the code).
6231 20. There is now an overall timeout for performing a callout verification. It
6232 defaults to 4 times the callout timeout, which applies to individual SMTP
6233 commands during the callout. The overall timeout applies when there is more
6234 than one host that can be tried. The timeout is checked before trying the
6235 next host. This prevents very long delays if there are a large number of
6236 hosts and all are timing out (e.g. when the network connections are timing
6237 out). The value of the overall timeout can be changed by specifying an
6238 additional sub-option for "callout", called "maxwait". For example:
6240 verify = sender/callout=5s,maxwait=20s
6242 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
6243 to the end before writing, but this should make it even safer).
6245 22. Exim was forgetting that it had advertised PIPELINING for the second and
6246 subsequent messages on an SMTP connection. It was also not resetting its
6247 memory on STARTTLS and an internal HELO.
6249 23. When Exim logs an SMTP synchronization error within a session, it now
6250 records whether PIPELINING has been advertised or not.
6252 24. Added 3 instances of "(long int)" casts to time_t variables that were being
6253 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
6254 rather than long int.
6256 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
6258 26. Added the never_mail option to autoreply.
6264 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
6265 crash if the getsockname() call failed; this can happen if a connection is
6266 closed very soon after it is established. The problem was simply in the
6267 order in which certain operations were done, causing Exim to try to write
6268 to the SMTP stream before it had set up the file descriptor. The bug has
6269 been fixed by making things happen in the correct order.
6275 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
6276 before the connection was closed, thus losing the rejection response.
6278 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
6279 some early Solaris releases, but causes trouble in current releases where
6280 socklen_t is defined.
6282 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
6285 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
6288 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
6289 that releasing the top of it at the end releases what was used for sub-
6290 expansions (unless the block got too big). However, discard this block if
6291 the first thing is a variable or header, so that we can use its block when
6292 it is dynamic (useful for very large $message_headers, for example).
6294 6. Lookups now cache *every* query, not just the most recent. A new, separate
6295 store pool is used for this. It can be recovered when all lookup caches are
6296 flushed. Lookups now release memory at the end of their result strings.
6297 This has involved some general refactoring of the lookup sources.
6299 7. Some code has been added to the store_xxx() functions to reduce the amount
6300 of flapping under certain conditions.
6302 8. log_incoming_interface used to affect only the <= reception log lines. Now
6303 it causes the local interface and port to be added to several more SMTP log
6304 lines, for example "SMTP connection from", and rejection lines.
6306 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
6308 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
6310 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
6311 could be overwritten at the end of the current message (or the start of a
6312 new message if it was set in a HELO ACL). The value is now preserved for
6313 the duration of the SMTP connection.
6315 12. If a transport had a headers_rewrite setting, and a matching header line
6316 contained an unqualified address, that address was qualified, even if it
6317 did not match any rewriting rules. The underlying bug was that the values
6318 of the flags that permit the existence of unqualified sender and recipient
6319 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
6320 non-local messages, and by -bnq for local messages) were not being
6321 preserved with the message after it was received.
6323 13. When Exim was logging an SMTP synchronization error, it could sometimes log
6324 "next input=" as part of the text comprising the host identity instead of
6325 the correct text. The code was using the same buffer for two different
6326 strings. However, depending on which order the printing function evaluated
6327 its arguments, the bug did not always show up. Under Linux, for example, my
6328 test suite worked just fine.
6330 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
6331 doesn't work with some older versions of Perl. It has been changed to "my",
6332 which in any case is probably the better facility to use.
6334 15. A really picky compiler found some instances of statements for creating
6335 error messages that either had too many or two few arguments for the format
6338 16. The size of the buffer for calls to the DNS resolver has been increased
6339 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
6340 for addresses that have a lot of PTR records. This alleviates a problem; it
6341 does not fully solve it.
6343 17. A dnsdb lookup for PTR records that receives more data than will fit in the
6344 buffer now truncates the list and logs the incident, which is the same
6345 action as happens when Exim is looking up a host name and its aliases.
6346 Previously in this situation something unpredictable would happen;
6347 sometimes it was "internal error: store_reset failed".
6349 18. If a server dropped the connection unexpectedly when an Exim client was
6350 using GnuTLS and trying to read a response, the client delivery process
6351 crashed while trying to generate an error log message.
6353 19. If a "warn" verb in an ACL added multiple headers to a message in a single
6354 string, for example:
6356 warn message = H1: something\nH2: something
6358 the text was added as a single header line from Exim's point of view
6359 though it ended up OK in the delivered message. However, searching for the
6360 second and subsequent header lines using $h_h2: did not work. This has been
6361 fixed. Similarly, if a system filter added multiple headers in this way,
6362 the routers could not see them.
6364 20. Expanded the error message when iplsearch is called with an invalid key to
6365 suggest using net-iplsearch in a host list.
6367 21. When running tests using -bh, any delays imposed by "delay" modifiers in
6368 ACLs are no longer actually imposed (and a message to that effect is
6371 22. If a "gecos" field in a passwd entry contained escaped characters, in
6372 particular, if it contained a \" sequence, Exim got it wrong when building
6373 a From: or a Sender: header from that name. A second bug also caused
6374 incorrect handling when an unquoted " was present following a character
6375 that needed quoting.
6377 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
6378 was not being matched caselessly.
6380 24. Arranged for all hyphens in the exim.8 source to be escaped with
6383 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
6384 properly. Recipient callout cache records were still being keyed to include
6385 the sender, even when use_sender was set false. This led to far more
6386 callouts that were necessary. The sender is no longer included in the key
6387 when use_sender is false.
6389 26. Added "control = submission" modifier to ACLs.
6391 27. Added the ${base62d: operator to decode base 62 numbers.
6393 28. dnsdb lookups can now access SRV records.
6395 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
6396 the configuration file.
6398 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
6399 (-v) mode. This makes the output for a verbose queue run more intelligible.
6401 31. Added a use_postmaster feature to recipient callouts.
6403 32. Added the $body_zerocount variable, containing the number of binary zero
6404 bytes in the message body.
6406 33. The time of last modification of the "new" subdirectory is now used as the
6407 "mailbox time last read" when there is a quota error for a maildir
6410 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
6412 35. Added +ignore_unknown as a special item in host lists.
6414 36. Code for decoding IPv6 addresses in host lists is now included, even if
6415 IPv6 support is not being compiled. This fixes a bug in which an IPv6
6416 address was recognized as an IP address, but was then not correctly decoded
6417 into binary, causing unexpected and incorrect effects when compared with
6424 1. Very minor rewording of debugging text in manualroute to say "list of
6425 hosts" instead of "hostlist".
6427 2. If verify=header_syntax was set, and a header line with an unqualified
6428 address (no domain) and a large number of spaces between the end of the
6429 name and the colon was received, the reception process suffered a buffer
6430 overflow, and (when I tested it) crashed. This was caused by some obsolete
6431 code that should have been removed. The fix is to remove it!
6433 3. When running in the test harness, delay a bit after writing a bounce
6434 message to get a bit more predictability in the log output.
6436 4. Added a call to search_tidyup() just before forking a reception process. In
6437 theory, someone could use a lookup in the expansion of smtp_accept_max_
6438 per_host which, without the tidyup, could leave open a database connection.
6440 5. Added the variables $recipient_data and $sender_data which get set from a
6441 lookup success in an ACL "recipients" or "senders" condition, or a router
6442 "senders" option, similar to $domain_data and $local_part_data.
6444 6. Moved the writing of debug_print from before to after the "senders" test
6447 7. Change 4.31/66 (moving the time when the Received: is generated) caused
6448 problems for message scanning, either using a data ACL, or using
6449 local_scan() because the Received: header was not generated till after they
6450 were called (in order to set the time as the time of reception completion).
6451 I have revised the way this works. The header is now generated after the
6452 body is received, but before the ACL or local_scan() are called. After they
6453 are run, the timestamp in the header is updated.
6459 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
6460 before starting a queue runner without re-exec. This happened only when
6461 deliver_drop_privilege was set or when the Exim user was set to root. The
6462 effect of the bug was that timeouts during subsequent deliveries caused
6463 crashes instead of being properly handled. The handler is now left at its
6464 default (and expected) setting.
6466 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
6467 message, again when deliver_drop_privilege is set or Exim is run as root.
6468 The bug described in (1) was not present in this case, but the tidying up
6469 of the other signals was missing. I have made the two cases consistent.
6471 3. The ignore_target_hosts setting on a manualroute router was being ignored
6472 for hosts that were looked up using the /MX notation.
6474 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
6477 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
6478 operated on the sender address. After changing the $sender_address to <>
6479 for the sender address verify, Exim was re-instated it as the original
6480 (before rewriting) address, but remembering that it had rewritten it, so it
6481 wasn't rewriting it again. This bug also had the effect of breaking the
6482 sender address verification caching when the sender address was rewritten.
6484 6. The ignore_target_hosts option was being ignored by the ipliteral router.
6485 This has been changed so that if the ip literal address matches
6486 ignore_target_hosts, the router declines.
6488 7. Added expansion conditions match_domain, match_address, and match_local_
6489 part (NOT match_host).
6491 8. The placeholder for the Received: header didn't have a length field set.
6493 9. Added code to Exim itself and to exim_lock to test for a specific race
6494 condition that could lead to file corruption when using MBX delivery. The
6495 issue is with the lockfile that is created in /tmp. If this file is removed
6496 after a process has opened it but before that process has acquired a lock,
6497 there is the potential for a second process to recreate the file and also
6498 acquire a lock. This could lead to two Exim processes writing to the file
6499 at the same time. The added code performs the same test as UW imapd; it
6500 checks after acquiring the lock that its file descriptor still refers to
6501 the same named file.
6503 10. The buffer for building added header lines was of fixed size, 8192 bytes.
6504 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
6507 11. Added the smtp_active_hostname option. If used, this will typically be made
6508 to depend on the incoming interface address. Because $interface_address is
6509 not set up until the daemon has forked a reception process, error responses
6510 that can happen earlier (such as "too many connections") no longer contain
6513 12. If an expansion in a condition on a "warn" statement fails because a lookup
6514 defers, the "warn" statement is abandoned, and the next ACL statement is
6515 processed. Previously this caused the whole ACL to be aborted.
6517 13. Added the iplsearch lookup type.
6519 14. Added ident_timeout as a log selector.
6521 15. Added tls_certificate_verified as a log selector.
6523 16. Added a global option tls_require_ciphers (compare the smtp transport
6524 option of the same name). This controls incoming TLS connections.
6526 17. I finally figured out how to make tls_require_ciphers do a similar thing
6527 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
6528 before starting the TLS session.
6530 18. Tabs are now shown as \t in -bP output.
6532 19. If the log selector return_path_on_delivery was set, Exim crashed when
6533 bouncing a message because it had too many Received: header lines.
6535 20. If two routers both had headers_remove settings, and the first one included
6536 a superfluous trailing colon, the final name in the first list and the
6537 first name in the second list were incorrectly joined into one item (with a
6538 colon in the middle).
6544 1. Added -C and -D options to the exinext utility, mainly to make it easier
6545 to include in the automated testing, but these could be helpful when
6546 multiple configurations are in use.
6548 2. The exinext utility was not formatting the output nicely when there was
6549 an alternate port involved in the retry record key, nor when there was a
6550 message id as well (for retries that were specific to a specific message
6551 and a specific host). It was also confused by IPv6 addresses, because of
6552 the additional colons they contain. I have fixed the IPv4 problem, and
6553 patched it up to do a reasonable job for IPv6.
6555 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
6556 delivery, the log line now contains "pipelined" if PIPELINING was used.
6558 4. An SMTP transport process used to panic and die if the bind() call to set
6559 an explicit outgoing interface failed. This has been changed; it is now
6560 treated in the same way as a connect() failure.
6562 5. A reference to $sender_host_name in the part of a conditional expansion
6563 that was being skipped was still causing a DNS lookup. This no longer
6566 6. The def: expansion condition was not recognizing references to header lines
6567 that used bh_ and bheader_.
6569 7. Added the _cache feature to named lists.
6571 8. The code for checking quota_filecount in the appendfile transport was
6572 allowing one more file than it should have been.
6574 9. For compatibility with Sendmail, the command line option
6582 and sets the incoming protocol and host name (for trusted callers). The
6583 host name and its colon can be omitted when only the protocol is to be set.
6584 Note the Exim already has two private options, -pd and -ps, that refer to
6585 embedded Perl. It is therefore impossible to set a protocol value of "d" or
6586 "s", but I don't think that's a major issue.
6588 10. A number of refactoring changes to the code, none of which should affect
6591 (a) The number of logging options was getting close to filling up the
6592 32-bit word that was used as a bit map. I have split them into two classes:
6593 those that are passed in the argument to log_write(), and those that are
6594 only ever tested independently outside of that function. These are now in
6595 separate 32-bit words, so there is plenty of room for expansion again.
6596 There is no change in the user interface or the logging behaviour.
6598 (b) When building, for example, log lines, the code previously used a
6599 macro that called string_cat() twice, in order to add two strings. This is
6600 not really sufficiently general. Furthermore, there was one instance where
6601 it was actually wrong because one of the argument was used twice, and in
6602 one call a function was used. (As it happened, calling the function twice
6603 did not affect the overall behaviour.) The macro has been replaced by a
6604 function that can join an arbitrary number of extra strings onto a growing
6607 (c) The code for expansion conditions now uses a table and a binary chop
6608 instead of a serial search (which was left over from when there were very
6609 few conditions). Also, it now recognizes conditions like "pam" even when
6610 the relevant support is not compiled in: a suitably worded error message is
6611 given if an attempt is made to use such a condition.
6613 11. Added ${time_interval:xxxxx}.
6615 12. A bug was causing one of the ddress fields not to be passed back correctly
6616 from remote delivery subprocesses. The field in question was not being
6617 subsequently used, so this caused to problems in practice.
6619 13. Added new log selectors queue_time and deliver_time.
6621 14. Might have fixed a bug in maildirsizefile handling that threw up
6622 "unexpected character" debug warnings, and recalculated the data
6623 unnecessarily. In any case, I expanded the warning message to give more
6626 15. Added the message "Restricted characters in address" to the statements in
6627 the default ACL that block characters like @ and % in local parts.
6629 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
6630 Three changes have been made:
6632 (a) There was a serious bug; a negative response to MAIL caused the whole
6633 recipient domain to be cached as invalid, thereby blocking all messages
6634 to all local parts at the same domain, from all senders. This bug has
6635 been fixed. The domain is no longer cached after a negative response to
6636 MAIL if the sender used is not empty.
6638 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
6641 (c) A new callout option, "use_sender" has been added for people who want
6642 the modified behaviour.
6648 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
6651 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
6652 indeed breaks things for older releases.
6654 3. Added additional logging to the case where there is a problem reading data
6655 from a filter that is running in a subprocess using a pipe, in order to
6656 try to track down a specific problem.
6658 4. Testing facility fudge: when running in the test harness and attempting
6659 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
6660 getting "No route to host". Convert this to a timeout.
6662 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
6665 6. Some OS don't have socklen_t but use size_t instead. This affects the
6666 fifth argument of getsockopt() amongst other things. This is now
6667 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
6668 can be set for individual OS. I have set it for SunOS5, OSF1, and
6669 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
6670 some earlier ones do not.
6672 7. Change 4.30/15 was not doing the test caselessly.
6674 8. The standard form for an IPv6 address literal was being rejected by address
6675 parsing in, for example, MAIL and RCPT commands. An example of this kind of
6676 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6677 this, as well as the form without the "IPv6" on the front (but only when
6678 address literals are enabled, of course).
6680 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
6682 10. Exim crashed if a message with an empty sender address specified by -f
6683 encountered a router with an errors_to setting. This could be provoked only
6684 by a command such as
6688 where an empty string was supplied; "<>" did not hit this bug.
6690 11. Installed PCRE release 4.5.
6692 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
6693 remained set. It is now erased.
6695 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
6696 times from message ids (which are base 36 rather than the normal 62).
6698 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
6699 were being counted as actual protocol errors, and logged if the log
6700 selector +smtp_protocol_error was set. One cannot be perfect in this test,
6701 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
6702 and DATA following a set of rejected RCPTs do not count as protocol errors.
6703 In other words, Exim assumes they were pipelined, though this may not
6704 actually be the case. Of course, in all cases the client gets an
6705 appropriate error code.
6707 15. If a lookup fails in an ACL condition, a message about the failure may
6708 be available; it is used if testing the ACL cannot continue, because most
6709 such messages specify what the cause of the deferral is. However, some
6710 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
6711 that caused an old message to be retained and used if a later statement
6712 caused a defer, replacing the real cause of the deferral.
6714 16. If an IP address had so many PTR records that the DNS lookup buffer
6715 was not large enough to hold them, Exim could crash while trying to process
6716 the truncated data. It now detects and logs this case.
6718 17. Further to 4.21/58, another change has been made: if (and only if) the
6719 first line of a message (the first header line) ends with CRLF, a bare LF
6720 in a subsequent header line has a space inserted after it, so as not to
6721 terminate the header.
6723 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
6724 unnecessarily, used atoi() instead of strtol(), and didn't check the
6725 termination when getting file sizes from file names by regex.
6727 19. Completely re-implemented the support for maildirsize files, in the light
6728 of a number of problems with the previous contributed implementation
6729 (4.30/29). In particular:
6731 . If the quota is zero, the maildirsize file is maintained, but no quota is
6734 . If the maildir directory does not exist, it is created before any attempt
6735 to write a maildirsize file.
6737 . The quota value in the file is just a cache; if the quota is changed in
6738 the transport, the new value overrides.
6740 . A regular expression is available for excluding directories from the
6743 20. The autoreply transport checks the characters in options that define the
6744 message's headers; it allows continued headers, but it was checking with
6745 isspace() after an embedded newline instead of explicitly looking for a
6748 21. If all the "regular" hosts to which an address was routed had passed their
6749 expiry times, and had not reached their retry times, the address was
6750 bounced, even if fallback hosts were defined. Now Exim should go on to try
6753 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
6754 equivalent code in the SMTP transport. Some hosts send humungous responses
6755 to HELO/EHLO, more than 1024 it seems.
6757 23. Refactoring: code in filter.c used (void *) for "any old type" but this
6758 gives compiler warnings in some environments. I've now done it "properly",
6761 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
6762 (because of problems with the built-in one) was declared to return uschar *
6763 instead of char *, causing compiler failure.
6765 25. Fixed a file descriptor leak when processing alias/forward files.
6767 26. Fixed a minor format string issue in dbfn.c.
6769 27. Typo in exim.c: ("dmbnz" for "dbmnz").
6771 28. If a filter file refered to $h_xxx or $message_headers, and the headers
6772 contained RFC 2047 "words", Exim's memory could, under certain conditions,
6775 29. When a sender address is verified, it is cached, to save repeating the test
6776 when there is more than one recipient in a message. However, when the
6777 verification involves a callout, it is possible for different callout
6778 options to be set for different recipients. It is too complicated to keep
6779 track of this in the cache, so now Exim always runs a verification when a
6780 callout is required, relying on the callout cache for the optimization.
6781 The overhead is duplication of the address routing, but this should not be
6784 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
6785 to be verified with callout=postmaster, and the main callout worked but the
6786 postmaster check failed, the verification correctly failed. However, if a
6787 subsequent RCPT command asked for sender verification *without* the
6788 postmaster check, incorrect caching caused this verification also to fail,
6791 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
6792 it was not caching the DNS options (qualify_single, search_parents) that
6793 were used when the lookup failed. A subsequent lookup with different
6794 options therefore always gave the same answer, though there were cases
6795 where it should not have. (Example: a "domains = !$mx_any" option on a
6796 dnslookup router: the "domains" option is always processed without any
6797 widening, but the router might have qualify_single set.) Now Exim uses the
6798 cached value only when the same options are set.
6800 32. Added John Jetmore's "exipick" utility to the distribution.
6802 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
6803 than a timeout (e.g. a certificate is required, and is not provided), an
6804 Exim server now closes the connection immediately. Previously it waited for
6805 the client to close - but if the client is SSL, it seems that they each
6806 wait for each other, leading to a delay before one of them times out.
6808 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
6809 maintained 0.8.x compatibility because I don't think many are using it, and
6810 it is clearly obsolete.
6812 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
6815 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
6816 client certificate was expired. A simple patch fixes this, though I don't
6817 understand the full logic of why the verify callback is called multiple
6820 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
6821 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
6822 which causes problems with some clients (such as the Certicom SSL Plus
6823 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
6824 disables the coutermeasure allowing Eudora to connect."
6826 38. Exim was not checking that a write() to a log file succeeded. This could
6827 lead to Bad Things if a log got too big, in particular if it hit a file
6828 size limit. Exim now panics and dies if it cannot write to a log file, just
6829 as it does if it cannot open a log file.
6831 39. Modified OS/Makefile-Linux so that it now contains
6833 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
6835 The two -D definitions ensure that Exim is compiled with large file
6836 support, which makes it possible to handle log files that are bigger than
6839 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
6840 instance) a domain was checked against a named list that involved a lookup,
6841 causing $domain_data to be set, then another domain was checked against the
6842 same list, then the first domain was re-checked, the value of $domain_data
6843 after the final check could be wrong. In particular, if the second check
6844 failed, it could be set empty. This bug probably also applied to
6847 41. The strip_trailing_dot option was not being applied to the address given
6848 with the -f command-line option.
6850 42. The code for reading a message's header from the spool was incrementing
6851 $received_count, but never initializing it. This meant that the value was
6852 incorrect (doubled) while delivering a message in the same process in which
6853 it was received. In the most common configuration of Exim, this never
6854 happens - a fresh exec is done - but it can happen when
6855 deliver_drop_privilege is set.
6857 43. When Exim logs an SMTP synchronization error - client data sent too soon -
6858 it now includes up to 150 characters of the unexpected data in the log
6861 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6862 and building data strings. The size of both of these buffers was 10 000
6863 bytes - far larger than anybody would *ever* want, thought I. Needless to
6864 say, somebody hit the limit. I have increased the maximum line length to
6865 20 000 and the maximum data length of concatenated lines to 100 000. I have
6866 also fixed two bugs, because there was no checking on these buffers. Tsk,
6867 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6868 buffer is too small.
6870 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6871 lsearch lookups. Now it does.
6873 46. When parsing a route_list item in a manualroute router, a fixed-length
6874 buffer was used for the list of hosts. I made this 1024 bytes long,
6875 thinking that nobody would ever have a list of hosts that long. Wrong.
6876 Somebody had a whole pile of complicated expansion conditions, and the
6877 string was silently truncated, leading to an expansion error. It turns out
6878 that it is easier to change to an unlimited length (owing to other changes
6879 that have happened since this code was originally written) than to build
6880 structure for giving a limitation error. The length of the item that
6881 expands into the list of hosts is now unlimited.
6883 47. The lsearch lookup could not handle data where the length of text line was
6884 more than 4095 characters. Such lines were truncated, leading to shortened
6885 data being returned. It should now handle lines of any length.
6887 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6888 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6891 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6892 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6893 no longer happens in comment lines. A list of replacements is now placed
6894 at the head of all of the source files, except those whose only change is
6895 to replace PERL_COMMAND in the very first #! line.
6897 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6898 messages on the queue, with a bottom-up merge sort, using code contributed
6899 by Michael Haardt. This should make operations like -bp somewhat faster on
6900 large queues. It won't affect queue runners, except when queue_run_in_order
6903 51. Installed eximstats 1.31 in the distribution.
6905 52. Added support for SRV lookups to the dnslookup router.
6907 53. If an ACL referred to $message_body or $message_body_end, the value was not
6908 reset for any messages that followed in the same SMTP session.
6910 54. The store-handling optimization for building very long strings was not
6911 differentiating between the different store pools. I don't think this
6912 actually made any difference in practice, but I've tidied it.
6914 55. While running the routers to verify a sender address, $sender_address
6915 was still set to the sender address. This is wrong, because when routing to
6916 send a bounce to the sender, it would be empty. Therefore, I have changed
6917 it so that, while verifying a sender address, $sender_address is set to <>.
6918 (There is no change to what happens when verifying a recipient address.)
6920 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6921 target A or AAAA records (if not already returned) without resetting the
6922 qualify_single or search_parents options of the DNS resolver. These are
6923 inappropriate in this case because the targets of MX and SRV records must
6924 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6925 target that, when qualified, matched something in the local domain. These
6926 two options are now turned off when doing these lookups.
6928 57. It seems that at least some releases of Reiserfs (which does not have the
6929 concept of a fixed number of inodes) returns zero and not -1 for the
6930 number of available inodes. This interacted badly with check_spool_inodes,
6931 which assumed that -1 was the "no such thing" setting. What I have done is
6932 to check that the total number of inodes is greater than zero before doing
6933 the test of how many are available.
6935 58. When a "warn" ACL statement has a log_message modifier, the message is
6936 remembered, and not repeated. This is to avoid a lot of repetition when a
6937 message has many recipients that cause the same warning to be written.
6938 However, Exim was preserving the list of already written lines for an
6939 entire SMTP session, which doesn't seem right. The memory is now reset if a
6940 new message is started.
6942 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6943 debugging output unless log_rewrite was also set.
6945 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6946 of (int)(handle) when we know that handle contains (void *)(-1).
6948 61. The Exim daemon panic-logs an error return when it closes the incoming
6949 connection. However "connection reset by peer" seems to be common, and
6950 isn't really an error worthy of noting specially, so that particular error
6953 62. When Exim is trying to find all the local interfaces, it used to panic and
6954 die if the ioctl to get the interface flags failed. However, it seems that
6955 on at least one OS (Solaris 9) it is possible to have an interface that is
6956 included in the list of interfaces, but for which you get a failure error
6957 for this call. This happens when the interface is not "plumbed" into a
6958 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6959 failure of the "get flags" call assumes that the interface is down.
6961 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6962 makes life easier for people who are doing arithmetic on fields extracted
6963 from dates, where you often get leading zeros that should not be
6964 interpreted as octal.
6966 64. Added qualify_domain to the redirect router, to override the global
6969 65. If a pathologically long header line contained very many addresses (the
6970 report of this problem mentioned 10 000) and each of them was rewritten,
6971 Exim could use up a very large amount of memory. (It kept on making new
6972 copies of the header line as it rewrote, and never released the old ones.)
6973 At the expense of a bit more processing, the header rewriting function has
6974 been changed so that it no longer eats memory in this way.
6976 66. The generation of the Received: header has been moved from the time that a
6977 message starts to be received, to the time that it finishes. The timestamp
6978 in the Received: header should now be very close to that of the <= log
6979 line. There are two side-effects of this change:
6981 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6982 logged header lines no longer include the local Received: line, because
6983 it has not yet been created. The same applies to a copy of the message
6984 that is returned to a non-SMTP sender when a message is rejected.
6986 (b) When a filter file is tested using -bf, no additional Received: header
6987 is added to the test message. After some thought, I decided that this
6990 This change does not affect the value of $received_for. It is still set
6991 after address rewriting, but before local_scan() is called.
6993 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6995 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6996 gave an unhelpful panic error message, and a defer error. I have managed to
6997 change this behaviour so that it now rejects any supplied certificate,
6998 which seems right, as the list of acceptable certificates is empty.
7000 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
7001 gave an unhelpful defer error. I have not managed to make this reject any
7002 supplied certificates, but the error message it gives is "no certificate
7003 supplied", which is not helpful.
7005 70. exigrep's output now also includes lines that are not associated with any
7006 message, but which match the given pattern. Implemented by a patch from
7007 Martin Sluka, which also tidied up the Perl a bit.
7009 71. Recipient callout verification, like sender verification, was using <> in
7010 the MAIL FROM command. This isn't really the right thing, since the actual
7011 sender may affect whether the remote host accepts the recipient or not. I
7012 have changed it to use the actual sender in the callout; this means that
7013 the cache record is now keyed on a recipient/sender pair, not just the
7014 recipient address. There doesn't seem to be a real danger of callout loops,
7015 since a callout by the remote host to check the sender would use <>.
7016 [SEE ABOVE: changed after hitting problems.]
7018 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
7019 temporary errors. However, in the case of such a code being given after
7020 the end of a data transmission (i.e. after ".") Exim was failing to write
7021 a retry record for the message. (Yes, there was some broken host that was
7022 actually sending 8xx at this point.)
7024 73. An unknown lookup type in a host list could cause Exim to panic-die when
7025 the list was checked. (An example that provoked this was putting <; in the
7026 middle of a list instead of at the start.) If this happened during a DATA
7027 ACL check, a -D file could be left lying around. This kind of configuration
7028 error no longer causes Exim to die; instead it causes a defer error. The
7029 incident is still logged to the main and panic logs.
7031 74. Buglet left over from Exim 3 conversion. The message "too many messages
7032 in one connection" was written to the rejectlog but not the mainlog, except
7033 when address rewriting (yes!) was being logged.
7035 75. Added write_rejectlog option.
7037 76. When a system filter was run not as root (that is, when system_filter_user
7038 was set), the values of the $n variables were not being returned to the
7039 main process; thus, they were not subsequently available in the $sn
7042 77. Added +return_path_on_delivery log selector.
7044 78. A connection timeout was being treated differently from recipients deferred
7045 when testing hosts_max_try with a message that was older than the host's
7046 retry timeout. (The host should not be counted, thus allowing all hosts to
7047 be tried at least once before bouncing.) This may have been the cause of an
7048 occasionally reported bug whereby a message would remain on the queue
7049 longer than the retry timeout, but would be bounced if a delivery was
7050 forced. I say "may" because I never totally pinned down the problem;
7051 setting up timeout/retry tests is difficult. See also the next item.
7053 79. The ultimate address timeout was not being applied to errors that involved
7054 a combination of host plus message (for example, a timeout on a MAIL
7055 command). When an address resolved to a number of possible hosts, and they
7056 were not all tried for each delivery (e.g. because of hosts_max_try), a
7057 message could remain on the queue longer than the retry timeout.
7059 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
7062 81. Fixed an obscure SMTP outgoing bug which required at least the following
7063 conditions: (a) there was another message waiting for the same server;
7064 (b) the server returned 5xx to all RCPT commands in the first message so
7065 that the message was not completed; (c) the server dropped the connection
7066 or gave a negative response to the RSET that Exim sends to abort the
7067 transaction. The observed case was a dropped connection after DATA that had
7068 been sent in pipelining mode. That is, the server had advertised PIPELINING
7069 but was not implementing it correctly. The effect of the bug was incorrect
7070 behaviour, such as trying another host, and this could lead to a crash.
7076 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
7077 and daemon.c were passed as pointers to ints; they should have been
7078 pointers to socklen_t variables (which are typically unsigned ints).
7080 2. Some signed/unsigned type warnings in the os.c file for Linux have been
7083 3. Fixed a really odd bug that affected only the testing scheme; patching a
7084 certain fixed string in the binary changed the value of another string that
7085 happened to be identical to the end of the original first string.
7087 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
7088 name", it returns that address as the IP address. On some operating
7089 systems (e.g. Solaris), it also passes back the IP address string as the
7090 "host name". However, on others (e.g. Linux), it passes back an empty
7091 string. Exim wasn't checking for this, and was changing the host name to an
7092 empty string, assuming it had been canonicalized.
7094 5. Although rare, it is permitted to have more than one PTR record for a given
7095 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
7096 all the names associated with an address, because they do in Solaris.
7097 However, it seems that they do not in Linux for data that comes from the
7098 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
7099 I found this out when I moved to a new Linux workstation and tried to run
7100 the Exim test suite.
7102 To get round this problem I have changed the code so that it now does its
7103 own call to the DNS to look up PTR records when searching for a host name.
7104 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
7105 addresses that are only in /etc/hosts are still found.
7107 This behaviour is, however, controlled by an option called host_lookup_
7108 order, which defaults to "bydns:byaddr". If people want to use the other
7109 order, or indeed, just use one or the other means of lookup, they can
7110 specify it in this variable.
7112 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
7113 some operating systems, this comes back from gethostbyaddr() as an empty
7114 string, and this is what Exim used to test for. However, it seems that in
7115 other systems, "." is yielded. Exim now tests for this case too.
7117 7. The values of check_spool_space and check_log_space are now held internally
7118 as a number of kilobytes instead of an absolute number of bytes. If a
7119 numbers is specified without 'K' or 'M', it is rounded up to the nearest
7120 kilobyte. This means that much larger values can be stored.
7122 8. Exim monitor: an attempt to get the action menu when not actually pointing
7123 at a message produces an empty menu entitled "No message selected". This
7124 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
7125 no entries in it ("Shell widget menu has zero width and/or height"). So I
7126 have added a single, blank menu entry in this case.
7128 9. Added ${quote_local_part.
7130 10. MIME decoding is now applied to the contents of Subject: header lines when
7133 11. Now that a reference to $sender_host_address automatically causes a reverse
7134 lookup to occur if necessary (4.13/18), there is no need to arrange for a
7135 host lookup before query-style lookups in lists that might use this
7136 variable. This has therefore been abolished, and the "net-" prefix is no
7137 longer necessary for query-style lookups.
7139 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
7140 have been a typo for LFLAGS, so it has been changed.
7142 13. The install script calls Exim with "-C /dev/null" in order to find the
7143 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
7144 to be output. However, since Exim outputs its version number before the
7145 error, it didn't break the script. It just looked ugly. I fixed this by
7146 always allowing "-C /dev/null" if the caller is root.
7148 14. Ignore overlarge ACL variable number when reading spool file - insurance
7149 against a later release with more variables having written the file.
7151 15. The standard form for an IPv6 address literal was being rejected by EHLO.
7152 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
7153 this, as well as the form without the "IPv6" on the front.
7155 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
7156 OS/Makefile-Darwin file.
7158 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
7159 only to LDAP libraries that do not have LDAP_OPT_DEREF.
7161 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
7162 variable. However, some OS use ints for inodes. I've added cast to long int
7163 to get rid of the compiler warning.
7165 19. I had forgotten to lock out "/../" in configuration file names when
7166 ALT_CONFIG_PREFIX was set.
7168 20. Routers used for verification do not need to specify transports. However,
7169 if such a router generated a host list, and callout was configured, Exim
7170 crashed, because it could not find a port number from the (non-existent)
7171 transport. It now assumes port 25 in this circumstance.
7173 21. Added the -t option to exigrep.
7175 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
7176 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
7177 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
7178 them is set, LOOKUP_LSEARCH is forced.
7180 23. "exim -bV" now outputs a list of lookups that are included in the binary.
7182 24. Added sender and host information to the "rejected by local_scan()" log
7183 line; previously there was no indication of these.
7185 25. Added .include_if_exists.
7187 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
7188 Linux. It turns out that not all file systems support this. Apparently some
7189 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
7190 do it.) To cope with this, the error EINVAL, which means that sync-ing is
7191 not supported on the file descriptor, is now ignored when Exim is trying to
7192 sync a directory. This applies only to Linux.
7194 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
7196 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
7197 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
7200 29. Added support for maildirsize files from supplied patch (modified a bit).
7202 30. The use of :fail: followed by an empty string could lead Exim to respond to
7203 sender verification failures with (e.g.):
7205 550 Verification failed for <xxx>
7206 550 Sender verify failed
7208 where the first response line was missing the '-' that indicates it is not
7209 the final line of the response.
7211 31. The loop for finding the name of the user that called Exim had a hardwired
7212 limit of 10; it now uses the value of finduser_retries, which is used for
7213 all other user lookups.
7215 32. Added $received_count variable, available in data and not_smtp ACLs, and at
7218 33. Exim was neglecting to zero errno before one call of strtol() when
7219 expanding a string and expecting an integer value. On some systems this
7220 resulted in spurious "integer overflow" errors. Also, it was casting the
7221 result into an int without checking.
7223 34. Testing for a connection timeout using "timeout_connect" in the retry rules
7224 did not work. The code looks as if it has *never* worked, though it appears
7225 to have been documented since at least release 1.62. I have made it work.
7227 35. The "timeout_DNS" error in retry rules, also documented since at least
7228 1.62, also never worked. As it isn't clear exactly what this means, and
7229 clearly it isn't a major issue, I have abolished the feature by treating it
7230 as "timeout", and writing a warning to the main and panic logs.
7232 36. The display of retry rules for -brt wasn't always showing the error code
7235 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
7236 timeout_connect_A, timeout_connect_MX.
7238 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
7239 to the empty sender.
7241 39. The daemon was not analysing the content of -oX till after it had closed
7242 stderr and disconnected from the controlling terminal. This meant that any
7243 syntax errors were only noted on the panic log, and the return code from
7244 the command was 0. By re-arranging the code a little, I've made the
7245 decoding happen first, so such errors now appear on stderr, and the return
7246 code is 1. However, the actual setting up of the sockets still happens in
7247 the disconnected process, so errors there are still only recorded on the
7250 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
7251 connections (as happens on some IP stacks) was logged at start up time as
7252 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
7253 it from "IPv6 and IPv4", which means that two separate sockets are being
7256 41. The debug output for gethostbyname2() or getipnodebyname() failures now
7257 says whether AF_INET or AF_INET6 was passed as an argument.
7259 42. Exiwhat output was messed up when time zones were included in log
7262 43. Exiwhat now gives more information about the daemon's listening ports,
7263 and whether -tls-on-connect was used.
7265 44. The "port" option of the smtp transport is now expanded.
7267 45. A "message" modifier in a "warn" statement in a non-message ACL was being
7268 silently ignored. Now an error message is written to the main and panic
7271 46. There's a new ACL modifier called "logwrite" which writes to a log file
7272 as soon as it is encountered.
7274 47. Added $local_user_uid and $local_user_gid at routing time.
7276 48. Exim crashed when trying to verify a sender address that was being
7279 49. Exim was recognizing only a space character after ".include". It now also
7280 recognizes a tab character.
7282 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
7283 extracting the relevant information from the specification. The man page no
7284 longer contains scrambled data for the -d option, and I've added a section
7285 at the front about calling Exim under different names.
7287 51. Added "extra_headers" argument to the "mail" command in filter files.
7289 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
7292 53. Installed eximstats 1.29.
7294 54. Added transport_filter_timeout as a generic transport option.
7296 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
7297 Cc: header lines. This was required by RFC 822, but it not required by RFC
7300 56. Exim used to add From:, Date:, and Message-Id: header lines to any
7301 incoming messages that did not have them. Now it does so only if the
7302 message originates locally, that is, if there is no associated remote host
7303 address. When Resent- header lines are present, this applies to the Resent-
7304 lines rather than the non-Resent- lines.
7306 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
7307 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
7309 58. Messages for configuration errors now include the name of the main
7310 configuration file - useful now that there may be more than one file in a
7311 list (.included file names were always shown).
7313 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
7314 for those rare installations that do not start the daemon as root or run it
7315 setuid root. I've cut out the call to initgroups() if the daemon is not
7318 60. The Exim user and group can now be bound into the binary as text strings
7319 that are looked up at the start of Exim's processing.
7321 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
7323 62. Added $mailstore_basename variable.
7325 63. Installed patch to sieve.c from Michael Haardt.
7327 64. When Exim failed to open the panic log after failing to open the main log,
7328 the original message it was trying to log was written to stderr and debug
7329 output, but if they were not available (the usual case in production), it
7330 was lost. Now it is written to syslog before the two lines that record the
7331 failures to open the logs.
7333 65. Users' Exim filters run in subprocesses under the user's uid. It is
7334 possible for a "deliver" command or an alias in a "personal" command to
7335 provoke an address rewrite. If logging of address rewriting is configured,
7336 this fails because the process is not running as root or exim. There may be
7337 a better way of dealing with this, but for the moment (because 4.30 needs
7338 to be released), I have disabled address rewrite logging when running a
7339 filter in a non-root, non-exim process.
7345 1. The buildconfig auxiliary program wasn't quoting the value set for
7346 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
7347 not defined. This bug was masked in 4.22 by the effect that was fixed in
7350 2. Some messages that were rejected after a message id was allocated were
7351 shown as "incomplete" by exigrep. It no longer does this for messages that
7352 are rejected by local_scan() or the DATA or non-SMTP ACLs.
7354 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
7355 have allow_domain_literals set, the ID did not get logged in the <= line.
7356 Domain literals are now always recognized in Message-ID: header lines.
7358 4. The first argument for a ${extract expansion item is the key name or field
7359 number. Leading and trailing spaces in this item were not being ignored,
7360 causing some misleading effects.
7362 5. When deliver_drop_privilege was set, single queue runner processes started
7363 manually (i.e. by the command "exim -q") or by the daemon (which uses the
7364 same command in the process it spins off) were not dropping privilege.
7366 6. When the daemon running as "exim" started a queue runner, it always
7367 re-executed Exim in the spun-off process. This is a waste of effort when
7368 deliver_drop_privilege is set. The new process now just calls the
7369 queue-runner function directly.
7375 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
7378 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
7379 ignored. Though the use of -oP was forcing the writing of a pid file, it
7380 was always written to the default place.
7382 3. If the message "no IP address found for host xxxx" is generated during
7383 incoming verification, it is now followed by identification of the incoming
7384 connection (so you can more easily find what provoked it).
7386 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
7388 5. Added some features to "harden" Exim a bit more against certain attacks:
7390 (a) There is now a build-time option called FIXED_NEVER_USERS that can
7391 be put in Local/Makefile. This is like the never_users runtime option,
7392 but it cannot be overridden. The default setting is "root".
7394 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
7395 prefix string with which any file named in a -C command line option
7398 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
7399 is retained for -C and -D only if the caller of Exim is root. Without
7400 it, the exim user may also use -C and -D and retain privilege.
7402 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
7403 command line option is disabled.
7405 6. Macro names set by the -D option must start with an upper case letter, just
7406 like macro names defined in the configuration file.
7408 7. Added "dereference=" facility to LDAP.
7410 8. Two instances of the typo "uknown" in the source files are fixed.
7412 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
7413 the Configure-Makefile script screwed up while processing it.
7415 10. Incorporated PCRE 4.4.
7417 11. The SMTP synchronization check was not operating right at the start of an
7418 SMTP session. For example, it could not catch a HELO sent before the client
7419 waited for the greeting. There is now a check for outstanding input at the
7420 point when the greeting is written. Because of the duplex, asynchronous
7421 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
7422 way, but not yet received, when the check is performed.
7424 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
7425 on TCP/IP sockets, because this apparently causes some broken clients to
7428 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
7429 unchanged) from the Cygwin maintainer.
7431 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
7432 was supported instead of testing for mbx. Effectively a typo.
7434 15. The spa authenticator server code was not checking that the input it
7435 received was valid base64.
7437 16. The debug output line for the "set" modifier in ACLs was not showing the
7438 name of the variable that was being set.
7440 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
7442 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
7443 lookup on its own. Something else was needed to trigger the lookup. For
7444 example, a match in host_lookup or the need for a host name in a host list.
7445 Now, if $sender_host_name is referenced and the host name has not yet been
7446 looked up, a lookup is performed. If the lookup fails, the variable remains
7447 empty, and $host_lookup_failed is set to "1".
7449 19. Added "eqi" as a case-independent comparison operator.
7451 20. The saslauthd authentication condition could segfault if neither service
7452 nor realm was specified.
7454 21. If an overflowing value such as "2048M" was set for message_size_limit, the
7455 error message that was logged was misleading, and incoming SMTP
7456 connections were dropped. The message is now more accurate, and temporary
7457 errors are given to SMTP connections.
7459 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
7460 (except RSET) with a 421 error, until QUIT is received. However, it was
7461 failing to send a response to QUIT.
7463 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
7464 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
7465 ACL is now run after the helo_try_verify_hosts code.
7467 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
7468 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
7469 case-independent, so other case variants are also recognized). Apparently
7470 some systems use these upper case variants.
7472 25. If more than two messages were waiting for the same host, and a transport
7473 filter was specified for the transport, Exim sent two messages over the
7474 same TCP/IP connection, and then failed with "socket operation on non-
7475 socket" when it tried to send the third.
7477 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
7479 27. The extern definition of crypt16() in expand.c was not being excluded when
7480 the OS had its own crypt16() function.
7482 28. Added bounce_return_body as a new option, and bounce_return_size_limit
7483 as a preferred synonym for return_size_limit, both as an option and as an
7486 29. Added LIBS=-liconv to OS/Makefile-OSF1.
7488 30. Changed the default configuration ACL to relax the local part checking rule
7489 for addresses that are not in any local domains. For these addresses,
7490 slashes and pipe symbols are allowed within local parts, but the sequence
7491 /../ is explicitly forbidden.
7493 31. SPA server authentication was not clearing the challenge buffer before
7496 32. log_message in a "warn" ACL statement was writing to the reject log as
7497 well as to the main log, which contradicts the documentation and doesn't
7498 seem right (because no rejection is happening). So I have stopped it.
7500 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
7501 However, I am unable to do any testing of this.
7503 34. Fixed an infelicity in the appendfile transport. When checking directories
7504 for a mailbox, to see if any needed to be created, it was accidentally
7505 using path names with one or more superfluous leading slashes; tracing
7506 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
7508 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
7509 contents are added to the log line that is written for every discarded
7510 recipient. (Previously a log_message setting was ignored.)
7512 36. The ${quote: operator now quotes the string if it is empty.
7514 37. The install script runs exim in order to find its version number. If for
7515 some reason other than non-existence or emptiness, which it checks, it
7516 could not run './exim', it was installing it with an empty version number,
7517 i.e. as "exim-". This error state is now caught, and the installation is
7520 38. An argument was missing from the function that creates an error message
7521 when Exim fails to connect to the socket for saslauthd authentication.
7522 This could cause Exim to crash, or give a corrupted message.
7524 39. Added isip, isip4, and isip6 to ${if conditions.
7526 40. The ACL variables $acl_xx are now saved with the message, and can be
7527 accessed later in routers, transports, and filters.
7529 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
7530 strings in the file are not string-expanded.
7532 42. If a MAIL command specified a SIZE value that was too large to fit into an
7533 int variable, the check against message_size_limit failed. Such values are
7534 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
7535 day this will have to be increased, but I don't think I want to be around
7536 when emails are that large.
7543 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
7544 iconv() is not standard in FreeBSD.
7546 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
7547 IPv6 enabled. The observed symptom was a segmentation fault on return from
7548 the function os_common_find_running_interfaces() in src/os.c.
7550 3. In the check_special_case() function in daemon.c I had used "errno" as an
7551 argument name, which causes warnings on some systems. This was basically a
7552 typo, since it was named "eno" in the comments!
7554 4. The code that waits for the clock to tick (at a resolution of some fraction
7555 of a second) so as to ensure message-id uniqueness was always waiting for
7556 at least one whole tick, when it could have waited for less. [This is
7557 almost certainly not relevant at current processor speeds, where it is
7558 unlikely to ever wait at all. But we try to future-proof.]
7560 5. The function that sleeps for a time interval that includes fractions of a
7561 second contained a race. It did not block SIGALRM between setting the
7562 timer, and suspending (a couple of lines later). If the interval was short
7563 and the sigsuspend() was delayed until after it had expired, the suspension
7564 never ended. On busy systems this could lead to processes getting stuck for
7567 6. Some uncommon configurations may cause a lookup to happen in a queue runner
7568 process, before it forks any delivery processes. The open lookup caching
7569 mechanism meant that the open file or database connection was passed into
7570 the delivery process. The problem was that delivery processes always tidy
7571 up cached lookup data. This could cause a problem for the next delivery
7572 process started by the queue runner, because the external queue runner
7573 process does not know about the closure. So the next delivery process
7574 still has data in the lookup cache. In the case of a file lookup, there was
7575 no problem because closing a file descriptor in a subprocess doesn't affect
7576 the parent. However, if the lookup was caching a connection to a database,
7577 the connection was closed, and the second delivery process was likely to
7578 see errors such as "PGSQL: query failed: server closed the connection
7579 unexpectedly". The problem has been fixed by closing all cached lookups
7580 in a queue runner before running a delivery process.
7582 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
7583 seem to have the "const" qualifier which it has on other OS. I've
7586 8. Change 4.21/2 was too strict. It is only if there are two authenticators
7587 *of the same type* (client or server) with the same public name that an
7588 error should be diagnosed.
7590 9. When Exim looked up a host name for an IP address, but failed to find the
7591 original IP address when looking up the host name (a safety check), it
7592 output the message "<ip address> does not match any IP for NULL", which was
7593 confusing, to say the least. The bug was that the host name should have
7594 appeared instead of "NULL".
7596 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
7597 user that is built into the binary, and the -C or -D options is used, root
7598 privilege is dropped before the configuration file is read. In addition,
7599 logging is switched to stderr instead of the normal log files. If the
7600 configuration then re-defines the Exim user, the unprivileged environment
7601 is probably not what is expected, so Exim logs a panic warning message (but
7604 However, if deliver_drop_privilege is set, the unprivileged state may well
7605 be exactly what is intended, so the warning has been cut out in that case,
7606 and Exim is allowed to try to write to its normal log files.
7612 1. smtp_return_error_details was not giving details for temporary sender
7613 or receiver verification errors.
7615 2. Diagnose a configuration error if two authenticators have the same public
7618 3. Exim used not to create the message log file for a message until the first
7619 delivery attempt. This could be confusing when incoming messages were held
7620 for policy or load reasons. The message log file is now created at the time
7621 the message is received, and an initial "Received" line is written to it.
7623 4. The automatically generated man page for command line options had a minor
7624 bug that caused no ill effects; however, a more serious problem was that
7625 the procedure for building the man page automatically didn't always
7626 operate. Consequently, release 4.20 contains an out-of-date version. This
7627 shouldn't happen again.
7629 5. When building Exim with embedded Perl support, the script that builds the
7630 Makefile was calling 'perl' to find its compile-time parameters, ignoring
7631 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
7633 6. The freeze_tell option was not being used for messages that were frozen on
7634 arrival, either by an ACL or by local_scan().
7636 7. Added the smtp_incomplete_transaction log selector.
7638 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
7639 was accepting AUTH without a new EHLO.
7641 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
7642 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
7645 10. Logging of TCP/IP connections (when configured) now happens in the main
7646 daemon process instead of the child process, so that the TCP/IP connection
7647 count is more accurate (but it can never be perfect).
7649 11. The use of "drop" in a nested ACL was not being handled correctly in the
7650 outer ACL. Now, if condition failure induced by the nested "drop" causes
7651 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
7652 or "require"), the connection is dropped.
7654 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
7655 that yield "discard" can now be used with an "accept" or a "discard" verb,
7656 but an error is generated for any others (because I can't see a useful way
7657 to define what should happen).
7659 13. When an ACL is read dynamically from a file (or anywhere else), the lines
7660 are now processed in the same way as lines in the Exim configuration file.
7661 In particular, continuation lines are supported.
7663 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
7665 15. Added -ti meaning -t -i.
7667 16. Check for letters, digits, hyphens, and dots in the names of dnslist
7668 domains, and warn by logging if others are found.
7670 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
7671 returned from GIFCONF when Exim is trying to find the list of interfaces on
7672 a host. The code in os.c has been modified to copy each ifreq to an aligned
7673 structure in all cases.
7675 Also, in some cases, the returned ifreq's were being copied to a 'struct
7676 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
7677 means the last couple of bytes of an IPv6 address could be chopped if the
7678 ifreq contained only a normal sockaddr (14 bytes storage).
7680 18. Named domain lists were not supported in the hosts_treat_as_local option.
7681 An entry such as +xxxx was not recognized, and was treated as a literal
7684 19. Ensure that header lines added by a DATA ACL are included in the reject log
7685 if the ACL subsequently rejects the message.
7687 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
7688 MD5 (which is deprecated).
7690 21. When testing a filter file using -bf, Exim was writing a message when it
7691 took the sender from a "From " line in the message, but it was not doing so
7692 when it took $return_path from a Return-Path: header line. It now does.
7694 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
7695 with a valid header line field name (a series of printing characters
7696 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
7698 23. Changed "disc" in the source to "disk" to conform to the documentation and
7699 the book and for uniformity.
7701 24. Ignore Sendmail's -Ooption=value command line item.
7703 25. When execve() failed while trying to run a command in a pipe transport,
7704 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
7705 could be confused with a return value of 69 from the command itself. This
7706 has been changed to 127, the value the shell returns if it is asked to run
7707 a non-existent command. The wording for the related log line suggests a
7708 non-existent command as the problem.
7710 26. If received_header_text expands to an empty string, do not add a Received:
7711 header line to the message. (Well, it adds a token one on the spool, but
7712 marks it "old" so that it doesn't get used or transmitted.)
7714 27. Installed eximstats 1.28 (addition of -nt option).
7716 28. There was no check for failure on the call to getsockname() in the daemon
7717 code. This can fail if there is a shortage of resources on the system, with
7718 ENOMEM, for example. A temporary error is now given on failure.
7720 29. Contrary to the C standard, it seems that in some environments, the
7721 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
7722 program. Exim now does this explicitly; it affects the formatting of
7723 timestamps using strftime().
7725 30. If exiqsumm was given junk data, it threw up some uninitialized variable
7726 complaints. I've now initialized all the variables, to avoid this.
7728 32. Header lines added by a system filter were not being "seen" during
7729 transport-time rewrites.
7731 33. The info_callback() function passed to OpenSSL is set up with type void
7732 (*)(SSL *, int, int), as described somewhere. However, when calling the
7733 function (actually a macro) that sets it up, the type void(*)() is
7734 expected. I've put in a cast to prevent warnings from picky compilers.
7736 34. If a DNS black list lookup found a CNAME record, but there were no A
7737 records associated with the domain it pointed at, Exim crashed.
7739 35. If a DNS black list lookup returned more than one A record, Exim ignored
7740 all but the first. It now scans all returned addresses if a particular IP
7741 value is being sought. In this situation, the contents of the
7742 $dnslist_value variable are a list of all the addresses, separated by a
7745 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
7746 to accept a host name and all its aliases if the forward lookup for any of
7747 them yielded the IP address of the incoming connection. Now it accepts only
7748 those names whose forward lookup yields the correct IP address. Any other
7749 names are discarded. This closes a loophole whereby a rogue DNS
7750 administrator could create reverse DNS records to break through a
7751 wildcarded host restriction in an ACL.
7753 37. If a user filter or a system filter that ran in a subprocess used any of
7754 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
7755 the wrong values were passed to the pipe command ($thisaddress had the
7756 value of $0, $0 had the value of $1, etc). This bug was introduced by
7757 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
7759 38. Improved the line breaking for long SMTP error messages from ACLs.
7760 Previously, if there was no break point between 40 and 75 characters, Exim
7761 left the rest of the message alone. Two changes have been made: (a) I've
7762 reduced the minimum length to 35 characters; (b) if it can't find a break
7763 point between 35 and 75 characters, it looks ahead and uses the first one
7764 that it finds. This may give the occasional overlong line, but at least the
7765 remaining text gets split now.
7767 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
7768 file descriptors might be low, and that setting 1000 would always raise it.
7769 It turns out that in some environments, the limit is already over 1000 and
7770 that lowering it causes trouble. So now Exim takes care not to decrease it.
7772 40. When delivering a message, the value of $return_path is set to $sender_
7773 address at the start of routing (routers may change the value). By an
7774 oversight, this default was not being set up when an address was tested by
7775 -bt or -bv, which affected the outcome if any router or filter referred to
7778 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
7779 log only when "message" or "log_message" are set. However, if one of the
7780 conditions was an address verification, or a call to a nested ACL, the
7781 messages generated by the underlying test were being passed through. This
7782 no longer happens. The underlying message is available in $acl_verify_
7783 message for both "message" and "log_message" expansions, so it can be
7784 passed through if needed.
7786 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
7787 new expansion $bh_ to give the encoded byte string without charset
7788 translation. Translation happens only if iconv() is available; HAVE_ICONV
7789 indicates this at build time. HEADERS_CHARSET gives the charset to
7790 translate to; headers_charset can change it in the configuration, and
7791 "headers charset" can change it in an individual filter file.
7793 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
7794 that creates RFC 2047 encoded "words" labels them as that charset instead
7795 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
7796 expansion operator; (ii) when Exim creates a From: line for a local
7797 message; (iii) when a header line is rewritten to include a "phrase" part.
7799 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
7800 buggy, causing it to skip the first lines of messages whose message ID
7801 ended in 'D'. This would not have bitten before Exim release 4.14, because
7802 message IDs were unlikely to end in 'D' before then. The effect was to have
7803 incorrect size information for certain domains.
7805 45. #include "config.h" was missing at the start of the crypt16.c module. This
7806 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
7809 46. If there was a timeout during a "random" callout check, Exim treated it as
7810 a failure of the random address, and carried on sending RSET and the real
7811 address. If the delay was just some slowness somewhere, the response to the
7812 original RCPT would be taken as a response to RSET and so on, causing
7813 mayhem of various kinds.
7815 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
7816 when I implemented it. It didn't allow for the fact that some option values
7817 may legitimately be negative (e.g. size_addition), and it didn't even do
7818 the right test for positive values.
7820 48. Domain names in DNS records are case-independent. Exim always looks them up
7821 in lower case. Some resolvers return domain names in exactly the case they
7822 appear in the zone file, that is, they may contain uppercase letters. Not
7823 all resolvers do this - some return always lower case. Exim was treating a
7824 change of case by a resolver as a change of domain, similar to a widening
7825 of a domain abbreviation. This triggered its re-routing code and so it was
7826 trying to route what was effectively the same domain again. This normally
7827 caused routing to fail (because the router wouldn't handle the domain
7828 twice). Now Exim checks for this case specially, and just changes the
7829 casing of the domain that it ultimately uses when it transmits the message
7832 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
7835 50. If a filter generated a file delivery with a non-absolute name (possible if
7836 no home directory exists for the router), the forbid_file option was not
7839 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
7840 the existing equality matching.
7842 52. Exim was using ints instead of ino_t variables in some places where it was
7843 dealing with inode numbers.
7845 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
7846 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
7847 called TMPDIR, and if it finds it is different, it changes its value.
7849 54. The smtp_printf() function is now made available to local_scan() so
7850 additional output lines can be written before returning. There is also an
7851 smtp_fflush() function to enable the detection of a dropped connection.
7852 The variables smtp_input and smtp_batched_input are exported to
7855 55. Changed the default runtime configuration: the message "Unknown user"
7856 has been removed from the ACL, and instead placed on the localuser router,
7857 using the cannot_route_message feature. This means that any verification
7858 failures that generate their own messages won't get overridden. Similarly,
7859 the "Unrouteable address" message that was in the ACL for unverifiable
7860 relay addresses has also been removed.
7862 56. Added hosts_avoid_esmtp to the smtp transport.
7864 57. The exicyclog script was not checking for the esoteric option
7865 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7866 will work only if exicyclog is run under the appropriate euid.
7868 58. Following a discussion on the list, the rules by which Exim recognises line
7869 endings on incoming messages have been changed. The -dropcr and drop_cr
7870 options are now no-ops, retained only for backwards compatibility. The
7871 following line terminators are recognized: LF CRLF CR. However, special
7872 processing applies to CR:
7874 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7875 nor a local message in the state where . is a terminator.
7877 (ii) If a bare CR is encountered in a header line, an extra space is added
7878 after the line terminator so as not to end the header. The reasoning
7879 behind this is that bare CRs in header lines are most likely either
7880 to be mistakes, or people trying to play silly games.
7882 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7883 was being incorrectly given as 18 bytes larger than it should have been.
7884 This is a VOB (very old bug).
7886 60. This may never have affected anything current, but just in case it has:
7887 When the local host is found other than at the start of a list of hosts,
7888 the local host, those with the same MX, and any that follow, are discarded.
7889 When the list in question was part of a longer list of hosts, the following
7890 hosts (not currently being processed) were also being discarded. This no
7891 longer happens. I'm not sure if this situation could ever has previously
7894 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7897 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7898 header. This is something that is recommended in a new Internet Draft, and
7899 is something that is documented as being done by Sendmail. There are two
7900 possible values. For messages generated by the autoreply transport, Exim
7903 Auto-Submitted: auto-replied
7905 whereas for all other generated messages (e.g. bounces) it adds
7907 Auto-Submitted: auto-generated
7909 63. The "personal" condition in filters now includes a test for the
7910 Auto-Submitted: header. If it contains the string "auto-" the message it
7911 not considered personal.
7913 64. Added rcpt_include_affixes as a generic transport option.
7915 65. Added queue_only_override (default true).
7917 66. Added the syslog_duplication option.
7919 67. If what should have been the first header line of a message consisted of
7920 a space followed by a colon, Exim was mis-interpreting it as a header line.
7921 It isn't of course - it is syntactically invalid and should therefore be
7922 treated as the start of the message body. The misbehaviour could have
7923 caused a number of strange effects, including loss of data in subsequent
7924 header lines, and spool format errors.
7926 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7927 client host had authenticated. This control can now be exercised by an ACL
7928 for more flexibility.
7930 69. By default, callouts do not happen when testing with -bh. There is now a
7931 variant, -bhc, which does actually run the callout code, including
7932 consulting and updating the callout cache.
7934 70. Added support for saslauthd authentication, courtesy of Alexander
7937 71. If statvfs() failed on the spool or log directories while checking their
7938 size for availability, Exim confusingly gave the error "space shortage".
7939 Furthermore, in debugging mode it crashed with a floating point exception.
7940 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7941 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7942 serious problem, Exim now writes to the main and panic logs when this
7943 happens, with details of the failure. It then refuses to accept the
7944 incoming message, giving the message "spool directory problem" or "log
7945 directory problem" with a 421 code for SMTP messages.
7947 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7948 0, 1, and 2 exist, because some OS complain for execs without them (see
7949 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7950 descriptors. However, the code omitted to check that the open succeeded,
7951 causing mysterious errors if for some reason the permissions on /dev/null
7952 got screwed. Now Exim writes a message to the main and panic logs, and
7953 bombs out if it can't open /dev/null.
7955 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7956 interact so that it is all more flexible. It is supposed to remain
7957 backwards compatible. Also added extra_local_interfaces.
7959 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7960 to bomb out with an assertion failure - to the client this appears as a
7961 connection drop. This problem occurs in the part of the code that was taken
7962 from the Samba project. Fortunately, the assertion is in a very simple
7963 function, so I have fixed this by reproducing the function inline in the
7964 one place where it is called, and arranging for authentication to fail
7965 instead of killing the process with assert().
7967 75. The SPA client code was not working when the server requested OEM rather
7968 than Unicode encoding.
7970 76. Added code to make require_files with a specific uid setting more usable in
7971 the case where statting the file as root fails - usually a non-root-mounted
7972 NFS file system. When this happens and the failure is EACCES, Exim now
7973 forks a subprocess and does the per-uid checking as the relevant uid.
7975 77. Added process_log_path.
7977 78. If log_file_path was not explicitly set, a setting of check_log_space or
7978 check_log_inodes was ignored.
7980 79. If a space check for the spool or log partitions fails, the incident is now
7981 logged. Of course, in the latter case the data may get lost...
7983 80. Added the %p formatting code to string_format() so that it can be used to
7984 print addresses in debug_print(). Adjusted all the address printing in the
7985 debugging in store.c to use %p rather than %d.
7987 81. There was a concern that a line of code in smtp_in.c could overflow a
7988 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7989 initially expressed, the concern was not well-founded, because trailing
7990 spaces are removed early. However, if the trailing spaces were followed by
7991 a NULL, they did not get removed, so the overflow was possible. Two fixes
7994 (a) I re-wrote the offending code in a cleaner fashion.
7995 (b) If an incoming SMTP command contains a NULL character, it is rejected
7998 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7999 runs initgroups(), so that if the Exim user is in any additional groups,
8000 they will be used during message reception.
8006 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob