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. Found and fixed by
164 JH/01 Avoid costly startup code when not strictly needed. This reduces time
165 for some exim process initialisations. It does mean that the logging
166 of TLS configuration problems is only done for the daemon startup.
168 JH/02 Early-pipelining support code is now included unless disabled in Makefile.
170 JH/03 DKIM verification defaults no long accept sha1 hashes, to conform to
171 RFC 8301. They can still be enabled, using the dkim_verify_hashes main
174 JH/04 Support CHUNKING from an smtp transport using a transport_filter, when
175 DKIM signing is being done. Previously a transport_filter would always
176 disable CHUNKING, falling back to traditional DATA.
178 JH/05 Regard command-line receipients as tainted.
180 JH/06 Bug 340: Remove the daemon pid file on exit, whe due to SIGTERM.
182 JH/07 Bug 2489: Fix crash in the "pam" expansion condition. It seems that the
183 PAM library frees one of the arguments given to it, despite the
184 documentation. Therefore a plain malloc must be used.
186 JH/08 Bug 2491: Use tainted buffers for the transport smtp context. Previously
187 on-stack buffers were used, resulting in a taint trap when DSN information
188 copied from a received message was written into the buffer.
190 JH/09 Bug 2493: Harden ARC verify against Outlook, whick has been seen to mix
191 the ordering of its ARC headers. This caused a crash.
193 JH/10 Bug 2492: Use tainted memory for retry record when needed. Previously when
194 a new record was being constructed with information from the peer, a trap
197 JH/11 Bug 2494: Unset the default for dmarc_tld_file. Previously a naiive
198 installation would get error messages from DMARC verify, when it hit the
199 nonexistent file indicated by the default. Distros wanting DMARC enabled
200 should both provide the file and set the option.
201 Also enforce no DMARC verification for command-line sourced messages.
203 JH/12 Fix an uninitialised flag in early-pipelining. Previously connections
204 could, depending on the platform, hang at the STARTTLS response.
206 JH/13 Bug 2498: Reset a counter used for ARC verify before handling another
207 message on a connection. Previously if one message had ARC headers and
208 the following one did not, a crash could result when adding an
209 Authentication-Results: header.
211 JH/14 Bug 2500: Rewind some of the common-coding in string handling between the
212 Exim main code and Exim-related utities. The introduction of taint
213 tracking also did many adjustments to string handling. Since then, eximon
214 frequently terminated with an assert failure.
216 JH/15 When PIPELINING, synch after every hundred or so RCPT commands sent and
217 check for 452 responses. This slightly helps the inefficieny of doing
218 a large alias-expansion into a recipient-limited target. The max_rcpt
219 transport option still applies (and at the current default, will override
220 the new feature). The check is done for either cause of synch, and forces
221 a fast-retry of all 452'd recipients using a new MAIL FROM on the same
222 connection. The new facility is not tunable at this time.
224 JH/16 Fix the variables set by the gsasl authenticator. Previously a pointer to
225 library live data was being used, so the results became garbage. Make
226 copies while it is still usable.
228 JH/17 Logging: when the deliver_time selector ise set, include the DT= field
229 on delivery deferred (==) and failed (**) lines (if a delivery was
230 attemtped). Previously it was only on completion (=>) lines.
232 JH/18 Authentication: the gsasl driver not provides the $authN variables in time
233 for the expansion of the server_scram_iter and server_scram_salt options.
235 WB/01 SPF: DNS lookups for the obsolete SPF RR type done by the libspf2 library
236 are now specifically given a NO_DATA response without hitting the system
237 resolver. The library goes on to do the now-standard TXT lookup.
238 Use of dnsdb lookups is not affected.
240 JH/19 Bug 2507: Modules: on handling a dynamic-module (lookups) open failure,
241 only retrieve the errormessage once. Previously two calls to dlerror()
242 were used, and the second one (for mainlog/paniclog) retrieved null
245 JH/20 Taint checking: disallow use of tainted data for
246 - the appendfile transport file and directory options
247 - the pipe transport command
248 - the autoreply transport file, log and once options
249 - file names used by the redirect router (including filter files)
251 - paths used by single-key lookups
252 Previously this was permitted.
254 JH/21 Bug 2501: Fix init call in the heimdal authenticator. Previously it
255 adjusted the size of a major service buffer; this failed because the
256 buffer was in use at the time. Change to a compile-time increase in the
257 buffer size, when this authenticator is compiled into exim.
259 JH/22 Taint-checking: move to safe-mode taint checking on all platforms. The
260 previous fast-mode was untenable in the face of glibs using mmap to
261 support larger malloc requests.
263 PP/01 Update the openssl_options possible values through OpenSSL 1.1.1c.
264 New values supported, if defined on system where compiled:
265 allow_no_dhe_kex, cryptopro_tlsext_bug, enable_middlebox_compat,
266 no_anti_replay, no_encrypt_then_mac, prioritize_chacha, tlsext_padding
268 JH/23 Performance improvement in the initial phase of a two-pass queue run. By
269 running a limited number of proceses in parallel, a benefit is gained. The
270 amount varies with the platform hardware and load. The use of the option
271 queue_run_in_order means we cannot do this, as ordering becomes
274 JH/24 Bug 2524: fix the cyrus_sasl auth driver gssapi usage. A previous fix
275 had introduced a string-copy (for ensuring NUL-termination) which was not
276 appropriate for that case, which can include embedded NUL bytes in the
277 block of data. Investigation showed the copy to actually be needless, the
278 data being length-specified.
280 JH/25 Fix use of concurrent TLS connections under GnuTLS. When a callout was
281 done during a receiving connection, and both used TLS, global info was
282 used rather than per-connection info for tracking the state of data
283 queued for transmission. This could result in a connection hang.
285 JH/26 Fix use of the SIZE parameter on MAIL commands, on continued connections.
286 Previously, when delivering serveral messages down a single connection
287 only the first would provide a SIZE. This was due to the size information
288 not being properly tracked.
290 JH/27 Bug 2530: When operating in a timezone with sub-minute offset, such as
291 TAI (at 37 seconds currently), pretend to be in UTC for time-related
292 expansion and logging. Previously, spurious values such as a future
293 minute could be seen.
295 JH/28 Bug 2533: Fix expansion of ${tr } item. When called in some situations
296 it could crash from a null-deref. This could also affect the
297 ${addresses: } operator and ${readsock } item.
299 JH/29 Bug 2537: Fix $mime_part_count. When a single connection had a non-mime
300 message following a mime one, the variable was not reset.
302 JH/30 When an pipelined-connect fails at the first response, assume incorrect
303 cached capability (perhaps the peer reneged?) and immediately retry in
306 JH/31 Fix spurious detection of timeout while writing to transport filter.
308 JH/32 Bug 2541: Fix segfault on bad cmdline -f (sender) argument. Previously
309 an attempt to copy the string was made before checking it.
311 JH/33 Fix the dsearch lookup to return an untainted result. Previously the
312 taint of the lookup key was maintained; we now regard the presence in the
313 filesystem as sufficient validation.
315 JH/34 Fix the readsocket expansion to not segfault when an empty "options"
316 argument is supplied.
318 JH/35 The dsearch lookup now requires that the directory is an absolute path.
319 Previously this was not checked, and nonempty relative paths made an
320 access under Exim's current working directory.
322 JH/36 Bug 2554: Fix msg:defer event for the hosts_max_try_hardlimit case.
323 Previously no event was raised.
325 JH/37 Bug 2552: Fix the check on spool space during reception to use the SIZE
326 parameter supplied by the sender MAIL FROM command. Previously it was
327 ignored, and only the check_spool_space option value for the required
330 JH/38 Fix $dkim_key_length. This should, after a DKIM verification, present
331 the size of the signing public-key. Previously it was instead giving
332 the size of the signature hash.
334 JH/39 DKIM verification: the RFC 8301 restriction on sizes of RSA keys is now
335 the default. See the (new) dkim_verify_min_keysizes option.
337 JH/40 Fix a memory-handling bug: when a connection carried multiple messages
338 and an ACL use a lookup for checking either the local_part or domain,
339 stale data could be accessed. Ensure that variable references are
340 dropped between messages.
342 JH/41 Bug 2571: Fix SPA authenticator. Running as a server, an offset supplied
343 by the client was not checked as pointing within response data before
344 being used. A malicious client could thus cause an out-of-bounds read and
345 possibly gain authentication. Fix by adding the check.
347 JH/42 Internationalisation: change the default for downconversion in the smtp
348 transport to be "if needed". Previously it was "as previously set" for
349 the message, which usually meant "if needed" for message-submission but
350 "no" for everything else. However, MTAs have been seen using SMTPUTF8
351 even when the envelope addresses did not need it, resulting in forwarding
352 failures to non-supporting MTAs. A downconvert in such cases will be
353 a no-op on the addresses, merely dropping the use of SMTPUTF8 by the
354 transport. The change does mean that addresses needing conversion will
355 be converted when previously a delivery failure would occur.
357 JH/43 Fix possible long line in DSN. Previously when a very long SMTP error
358 response was received it would be used unchecked in a fail-DSN, violating
359 standards on line-length limits. Truncate if needed.
361 HS/01 Remove parameters of the link to www.open-spf.org. The linked form
362 doesn't work. (Additionally add a new main config option to configure the
369 JH/01 OpenSSL: With debug enabled output keying information sufficient, server
370 side, to decode a TLS 1.3 packet capture.
372 JH/02 OpenSSL: Suppress the sending of (stateful) TLS1.3 session tickets.
373 Previously the default library behaviour applied, sending two, each in
376 JH/03 Debug output for ACL now gives the config file name and line number for
379 JH/04 The default received_header_text now uses the RFC 8314 tls cipher clause.
381 JH/05 DKIM: ensure that dkim_domain elements are lowercased before use.
383 JH/06 Fix buggy handling of autoreply bounce_return_size_limit, and a possible
384 buffer overrun for (non-chunking) other transports.
386 JH/07 GnuTLS: Our use of late (post-handshake) certificate verification, under
387 TLS1.3, means that a server rejecting a client certificate is not visible
388 to the client until the first read of encrypted data (typically the
389 response to EHLO). Add detection for that case and treat it as a failed
390 TLS connection attempt, so that the normal retry-in-clear can work (if
391 suitably configured).
393 JB/01 Bug 2375: fix expansions of 822 addresses having comments in local-part
394 and/or domain. Found and fixed by Jason Betts.
396 JH/08 Add hardening against SRV & TLSA lookups the hit CNAMEs (a nonvalid
397 configuration). If a CNAME target was not a wellformed name pattern, a
400 JH/09 Logging: Fix initial listening-on line for multiple ports for an IP when
401 the OS reports them interleaved with other addresses.
403 JH/10 OpenSSL: Fix aggregation of messages. Previously, when PIPELINING was
404 used both for input and for a verify callout, both encrypted, SMTP
405 responses being sent by the server could be lost. This resulted in
406 dropped connections and sometimes bounces generated by a peer sending
409 JH/11 Harden plaintext authenticator against a badly misconfigured client-send
410 string. Previously it was possible to cause undefined behaviour in a
411 library routine (usually a crash). Found by "zerons".
413 JH/12 Bug 2384: fix "-bP smtp_receive_timeout". Previously it returned no
416 JH/13 Bug 2386: Fix builds with Dane under LibreSSL 2.9.0 onward. Some old
417 API was removed, so update to use the newer ones.
419 JH/14 Bug 1891: Close the log file if receiving a non-smtp message, without
420 any timeout set, is taking a long time. Previously we would hang on to a
421 rotated logfile "forever" if the input was arriving with long gaps
422 (a previous attempt to fix addressed lack, for a long time, of initial
425 HS/01 Bug 2390: Use message_id for tempfile creation to avoid races in a
426 shared (NFS) environment. The length of the tempfile name is now
427 4 + 16 ("hdr.$message_exim_id") which might break on file
428 systems which restrict the file name length to lower values.
431 HS/02 Bug 2390: Use message_id for tempfile creation to avoid races in a
432 shared (NFS) environment.
434 HS/03 Bug 2392: exigrep does case sensitive *option* processing (as it
435 did for all versions <4.90). Notably -M, -m, --invert, -I may be
438 JH/15 Use unsigned when creating bitmasks in macros, to avoid build errors
439 on some platforms for bit 31.
441 JH/16 GnuTLS: rework ciphersuite strings under recent library versions. Thanks
442 to changes apparently associated with TLS1.3 handling some of the APIs
443 previously used were either nonfunctional or inappropriate. Strings
444 like TLS1.3:ECDHE_SECP256R1__RSA_PSS_RSAE_SHA256__AES_256_GCM__AEAD:256
445 and TLS1.2:ECDHE_SECP256R1__RSA_SHA256__AES_128_CBC__SHA256:128 replace
446 the previous TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256 .
447 This affects log line X= elements, the $tls_{in,out}_cipher variables,
448 and the use of specific cipher names in the encrypted= ACL condition.
450 JH/17 OpenSSL: the default openssl_options now disables ssl_v3.
452 JH/18 GnuTLS: fix $tls_out_ocsp under hosts_request_ocsp. Previously the
453 verification result was not updated unless hosts_require_ocsp applied.
455 JH/19 Bug 2398: fix listing of a named-queue. Previously, even with the option
456 queue_list_requires_admin set to false, non-admin users were denied the
459 JH/20 Bug 2389: fix server advertising of usable certificates, under GnuTLS in
460 directory-of-certs mode. Previously they were advertised despite the
463 JH/21 The smtp transport option "hosts_noproxy_tls" is now unset by default.
464 A single TCP connection by a client will now hold a TLS connection open
465 for multiple message deliveries, by default. Previously the default was to
468 JH/22 The smtp transport option "hosts_try_dane" now enables all hosts by
469 default. If built with the facility, DANE will be used. The facility
470 SUPPORT_DANE is now enabled in the prototype build Makefile "EDITME".
472 JH/23 The build default is now for TLS to be included; the SUPPORT_TLS define
473 is replaced with DISABLE_TLS. Either USE_GNUTLS or (the new) USE_OPENSSL
474 must be defined and you must still, unless you define DISABLE_TLS, manage
475 the the include-dir and library-file requirements that go with that
476 choice. Non-TLS builds are still supported.
478 JH/24 Fix duplicated logging of peer name/address, on a transport connection-
481 JH/25 The smtp transport option "hosts_try_fastopen" now enables all hosts by
482 default. If the platform supports and has the facility enabled, it will
483 be requested on all coneections.
485 JH/26 The PIPE_CONNECT facility is promoted from experimental status and is now
486 controlled by the build-time option SUPPORT_PIPE_CONNECT.
488 PP/01 Unbreak heimdal_gssapi, broken in 4.92.
490 JH/27 Bug 2404: Use the main-section configuration option "dsn_from" for
491 success-DSN messages. Previously the From: header was always the default
492 one for these; the option was ignored.
494 JH/28 Fix the timeout on smtp response to apply to the whole response.
495 Previously it was reset for every read, so a teergrubing peer sending
496 single bytes within the time limit could extend the connection for a
497 long time. Credit to Qualsys Security Advisory Team for the discovery.
499 JH/29 Fix DSN Final-Recipient: field. Previously it was the post-routing
500 delivery address, which leaked information of the results of local
501 forwarding. Change to the original envelope recipient address, per
504 JH/30 Bug 2411: Fix DSN generation when RFC 3461 failure notification is
505 requested. Previously not bounce was generated and a log entry of
506 error ignored was made.
508 JH/31 Avoid re-expansion in ${sort } expansion. (CVE-2019-13917)
510 JH/32 Introduce a general tainting mechanism for values read from the input
511 channel, and values derived from them. Refuse to expand any tainted
512 values, to catch one form of exploit.
514 JH/33 Bug 2413: Fix dkim_strict option. Previously the expansion result
515 was unused and the unexpanded text used for the test. Found and
516 fixed by Ruben Jenster.
518 JH/34 Fix crash after TLS shutdown. When the TCP/SMTP channel was left open,
519 an attempt to use a TLS library read routine dereffed a nul pointer,
522 JH/35 Bug 2409: filter out-of-spec chars from callout response before using
523 them in our smtp response.
525 JH/36 Have the general router option retry_use_local_part default to true when
526 any of the restrictive preconditions are set (to anything). Previously it
527 was only for check_local user. The change removes one item of manual
528 configuration which is required for proper retries when a remote router
529 handles a subset of addresses for a domain.
531 JH/37 Appendfile: when evaluating quota use (non-quota_size_regex) take the file
532 link count into consideration.
534 HS/04 Fix handling of very log lines in -H files. If a -<key> <value> line
535 caused the extension of big_buffer, the following lines were ignored.
537 JH/38 Bug 1395: Teach the DNS negative-cache about TTL value from the SOA in
538 accordance with RFC 2308. Previously there was no expiry, so a longlived
539 receive process (eg. due to ACL delays) versus a short SOA value could
542 HS/05 Handle trailing backslash gracefully. (CVE-2019-15846)
544 JH/39 Promote DMARC support to mainline.
546 JH/40 Bug 2452: Add a References: header to DSNs.
548 JH/41 With GnuTLS 3.6.0 (and later) do not attempt to manage Diffie-Hellman
549 parameters. The relevant library call is documented as "Deprecated: This
550 function is unnecessary and discouraged on GnuTLS 3.6.0 or later. Since
551 3.6.0, DH parameters are negotiated following RFC7919."
553 HS/06 Change the default of dnssec_request_domains to "*"
555 JH/42 Bug 2545: Fix CHUNKING for all RCPT commands rejected. Previously we
556 carried on and emitted a BDAT command, even when PIPELINING was not
559 JH/43 Bug 2465: Fix taint-handling in dsearch lookup. Previously a nontainted
560 buffer was used for the filename, resulting in a trap when tainted
561 arguments (eg. $domain) were used.
563 JH/44 With OpenSSL 1.1.1 (onwards) disable renegotiation for TLS1.2 and below;
564 recommended to avoid a possible server-load attack. The feature can be
565 re-enabled via the openssl_options main cofiguration option.
567 JH/45 local_scan API: documented the current smtp_printf() call. This changed
568 for version 4.90 - adding a "more data" boolean to the arguments.
569 Bumped the ABI version number also, this having been missed previously;
570 release versions 4.90 to 4.92.3 inclusive were effectively broken in
571 respect of usage of smtp_printf() by either local_scan code or libraries
572 accessed via the ${dlfunc } expansion item. Both will need coding
573 adjustment for any calls to smtp_printf() to match the new function
574 signature; a FALSE value for the new argument is always safe.
576 JH/46 FreeBSD: fix use of the sendfile() syscall. The shim was not updating
577 the file-offset (which the Linux syscall does, and exim expects); this
578 resulted in an indefinite loop.
580 JH/47 ARC: fix crash in signing, triggered when a configuration error failed
581 to do ARC verification. The Authentication-Results: header line added
582 by the configuration then had no ARC item.
588 JH/01 Remove code calling the customisable local_scan function, unless a new
589 definition "HAVE_LOCAL_SCAN=yes" is present in the Local/Makefile.
591 JH/02 Bug 1007: Avoid doing logging from signal-handlers, as that can result in
592 non-signal-safe functions being used.
594 JH/03 Bug 2269: When presented with a received message having a stupidly large
595 number of DKIM-Signature headers, disable DKIM verification to avoid
596 a resource-consumption attack. The limit is set at twenty.
598 JH/04 Add variables $arc_domains, $arc_oldest_pass for ARC verify. Fix the
599 report of oldest_pass in ${authres } in consequence, and separate out
600 some descriptions of reasons for verification fail.
602 JH/05 Bug 2273: Cutthrough delivery left a window where the received messsage
603 files in the spool were present and unlocked. A queue-runner could spot
604 them, resulting in a duplicate delivery. Fix that by doing the unlock
605 after the unlink. Investigation by Tim Stewart. Take the opportunity to
606 add more error-checking on spoolfile handling while that code is being
609 PP/01 Refuse to open a spool data file (*-D) if it's a symlink.
610 No known attacks, no CVE, this is defensive hardening.
612 JH/06 Bug 2275: The MIME ACL unlocked the received message files early, and
613 a queue-runner could start a delivery while other operations were ongoing.
614 Cutthrough delivery was a common victim, resulting in duplicate delivery.
615 Found and investigated by Tim Stewart. Fix by using the open message data
616 file handle rather than opening another, and not locally closing it (which
617 releases a lock) for that case, while creating the temporary .eml format
618 file for the MIME ACL. Also applies to "regex" and "spam" ACL conditions.
620 JH/07 Bug 177: Make a random-recipient callout success visible in ACL, by setting
621 $sender_verify_failure/$recipient_verify_failure to "random".
623 JH/08 When generating a selfsigned cert, use serial number 1 since zero is not
626 JH/09 Bug 2274: Fix logging of cmdline args when starting in an unlinked cwd.
627 Previously this would segfault.
629 JH/10 Fix ARC signing for case when DKIM signing failed. Previously this would
632 JH/11 Bug 2264: Exim now only follows CNAME chains one step by default. We'd
633 like zero, since the resolver should be doing this for us, But we need one
634 as a CNAME but no MX presence gets the CNAME returned; we need to check
635 that doesn't point to an MX to declare it "no MX returned" rather than
636 "error, loop". A new main option is added so the older capability of
637 following some limited number of chain links is maintained.
639 JH/12 Add client-ip info to non-pass iprev ${authres } lines.
641 JH/13 For receent Openssl versions (1.1 onward) use modern generic protocol
642 methods. These should support TLS 1.3; they arrived with TLS 1.3 and the
643 now-deprecated earlier definitions used only specified the range up to TLS
644 1.2 (in the older-version library docs).
646 JH/14 Bug 2284: Fix DKIM signing for body lines starting with a pair of dots.
648 JH/15 Rework TLS client-side context management. Stop using a global, and
649 explicitly pass a context around. This enables future use of TLS for
650 connections to service-daemons (eg. malware scanning) while a client smtp
651 connection is using TLS; with cutthrough connections this is quite likely.
653 JH/16 Fix ARC verification to do AS checks in reverse order.
655 JH/17 Support a "tls" option on the ${readsocket } expansion item.
657 JH/18 Bug 2287: Fix the protocol name (eg utf8esmtp) for multiple messages
658 using the SMTPUTF8 option on their MAIL FROM commands, in one connection.
659 Previously the "utf8" would be re-prepended for every additional message.
661 JH/19 Reject MAIL FROM commands with SMTPUTF8 when the facility was not advertised.
662 Previously thery were accepted, resulting in issues when attempting to
663 forward messages to a non-supporting MTA.
665 PP/02 Let -n work with printing macros too, not just options.
667 JH/20 Bug 2296: Fix cutthrough for >1 address redirection. Previously only
668 one parent address was copied, and bogus data was used at delivery-logging
669 time. Either a crash (after delivery) or bogus log data could result.
670 Discovery and analysis by Tim Stewart.
672 PP/03 Make ${utf8clean:} expansion operator detect incomplete final character.
673 Previously if the string ended mid-character, we did not insert the
674 promised '?' replacement.
676 PP/04 Documentation: current string operators work on bytes, not codepoints.
678 JH/21 Change as many as possible of the global flags into one-bit bitfields; these
679 should pack well giving a smaller memory footprint so better caching and
680 therefore performance. Group the declarations where this can't be done so
681 that the byte-sized flag variables are not interspersed among pointer
682 variables, giving a better chance of good packing by the compiler.
684 JH/22 Bug 1896: Fix the envelope from for DMARC forensic reports to be possibly
685 non-null, to avoid issues with sites running BATV. Previously reports were
686 sent with an empty envelope sender so looked like bounces.
688 JH/23 Bug 2318: Fix the noerror command within filters. It wasn't working.
689 The ignore_error flag wasn't being returned from the filter subprocess so
690 was not set for later routers. Investigation and fix by Matthias Kurz.
692 JH/24 Bug 2310: Raise a msg:fail:internal event for each undelivered recipient,
693 and a msg:complete for the whole, when a message is manually removed using
694 -Mrm. Developement by Matthias Kurz, hacked on by JH.
696 JH/25 Avoid fixed-size buffers for pathnames in DB access. This required using
697 a "Gnu special" function, asprintf() in the DB utility binary builds; I
698 hope that is portable enough.
700 JH/26 Bug 2311: Fix DANE-TA verification under GnuTLS. Previously it was also
701 requiring a known-CA anchor certificate; make it now rely entirely on the
702 TLSA as an anchor. Checking the name on the leaf cert against the name
703 on the A-record for the host is still done for TA (but not for EE mode).
705 JH/27 Fix logging of proxy address. Previously, a pointless "PRX=[]:0" would be
706 included in delivery lines for non-proxied connections, when compiled with
707 SUPPORT_SOCKS and running with proxy logging enabled.
709 JH/28 Bug 2314: Fire msg:fail:delivery event even when error is being ignored.
710 Developement by Matthias Kurz, tweaked by JH. While in that bit of code,
711 move the existing event to fire before the normal logging of message
712 failure so that custom logging is bracketed by normal logging.
714 JH/29 Bug 2322: A "fail" command in a non-system filter (file) now fires the
715 msg:fail:internal event. Developement by Matthias Kurz.
717 JH/30 Bug 2329: Increase buffer size used for dns lookup from 2k, which was
718 far too small for todays use of crypto signatures stored there. Go all
719 the way to the max DNS message size of 64kB, even though this might be
720 overmuch for IOT constrained device use.
722 JH/31 Fix a bad use of a copy function, which could be used to pointlessly
723 copy a string over itself. The library routine is documented as not
724 supporting overlapping copies, and on MacOS it actually raised a SIGABRT.
726 JH/32 For main options check_spool_space and check_inode_space, where the
727 platform supports 64b integers, support more than the previous 2^31 kB
728 (i.e. more than 2 TB). Accept E, P and T multipliers in addition to
729 the previous G, M, k.
731 JH/33 Bug 2338: Fix the cyrus-sasl authenticator to fill in the
732 $authenticated_fail_id variable on authentication failure. Previously
735 JH/34 Increase RSA keysize of autogen selfsign cert from 1024 to 2048. RHEL 8.0
736 OpenSSL didn't want to use such a weak key. Do for GnuTLS also, and for
737 more-modern GnuTLS move from GNUTLS_SEC_PARAM_LOW to
738 GNUTLS_SEC_PARAM_MEDIUM.
740 JH/35 OpenSSL: fail the handshake when SNI processing hits a problem, server
741 side. Previously we would continue as if no SNI had been received.
743 JH/36 Harden the handling of string-lists. When a list consisted of a sole
744 "<" character, which should be a list-separator specification, we walked
745 off past the nul-terimation.
747 JH/37 Bug 2341: Send "message delayed" warning MDNs (restricted to external
748 causes) even when the retry time is not yet met. Previously they were
749 not, meaning that when (say) an account was over-quota and temp-rejecting,
750 and multiple senders' messages were queued, only one sender would get
751 notified on each configured delay_warning cycle.
753 JH/38 Bug 2351: Log failures to extract envelope addresses from message headers.
755 JH/39 OpenSSL: clear the error stack after an SSL_accept(). With anon-auth
756 cipher-suites, an error can be left on the stack even for a succeeding
757 accept; this results in impossible error messages when a later operation
760 AM/01 Bug 2359: GnuTLS: repeat lowlevel read and write operations while they
761 return error codes indicating retry. Under TLS1.3 this becomes required.
763 JH/40 Fix the feature-cache refresh for EXPERIMENTAL_PIPE_CONNECT. Previously
764 it only wrote the new authenticators, resulting in a lack of tracking of
765 peer changes of ESMTP extensions until the next cache flush.
767 JH/41 Fix the loop reading a message header line to check for integer overflow,
768 and more-often against header_maxsize. Previously a crafted message could
769 induce a crash of the recive process; now the message is cleanly rejected.
771 JH/42 Bug 2366: Fix the behaviour of the dkim_verify_signers option. It had
772 been totally disabled for all of 4.91. Discovery and fix by "Mad Alex".
778 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
779 When redis_servers is set to a list of > 1 element, and the Redis servers
780 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
781 case of MOVED into a DEFER case instead, thus moving the query onto the
782 next server in the list. For a cluster of N elements, all N servers must
783 be defined in redis_servers.
785 GF/02 Catch and remove uninitialized value warning in exiqsumm
786 Check for existence of @ARGV before looking at $ARGV[0]
788 JH/01 Replace the store_release() internal interface with store_newblock(),
789 which internalises the check required to safely use the old one, plus
790 the allocate and data copy operations duplicated in both (!) of the
791 extant use locations.
793 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
794 modifier. This matches the restriction on the commandline.
796 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
797 Previously only the last row was returned.
799 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
800 we assumed that tags in the header were well-formed, and parsed the
801 element content after inspecting only the first char of the tag.
802 Assumptions at that stage could crash the receive process on malformed
805 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
806 While running the DKIM ACL we operate on the Permanent memory pool so that
807 variables created with "set" persist to the DATA ACL. Also (at any time)
808 DNS lookups that fail create cache records using the Permanent pool. But
809 expansions release any allocations made on the current pool - so a dnsdb
810 lookup expansion done in the DKIM ACL releases the memory used for the
811 DNS negative-cache, and bad things result. Solution is to switch to the
812 Main pool for expansions.
813 While we're in that code, add checks on the DNS cache during store_reset,
814 active in the testsuite.
815 Problem spotted, and debugging aided, by Wolfgang Breyha.
817 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
818 When none of the hosts presented to a transport match an already-open
819 connection, close it and proceed with the list. Previously we would
820 queue the message. Spotted by Lena with Yahoo, probably involving
823 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
824 Previously a spurious "250 OK id=" response was appended to the proper
827 JH/08 The "support for" informational output now, which built with Content
828 Scanning support, has a line for the malware scanner interfaces compiled
829 in. Interface can be individually included or not at build time.
831 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
832 by the template makefile "src/EDITME". The "STREAM" support for an older
833 ClamAV interface method is removed.
835 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
836 rows affected is given instead).
838 JH/11 The runtime Berkeley DB library version is now additionally output by
839 "exim -d -bV". Previously only the compile-time version was shown.
841 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
842 SMTP connection. Previously, when one had more receipients than the
843 first, an abortive onward connection was made. Move to full support for
844 multiple onward connections in sequence, handling cutthrough connection
845 for all multi-message initiating connections.
847 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
848 routers. Previously, a multi-recipient message would fail to match the
849 onward-connection opened for the first recipient, and cause its closure.
851 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
852 a timeout on read on a GnuTLS initiating connection, resulting in the
853 initiating connection being dropped. This mattered most when the callout
854 was marked defer_ok. Fix to keep the two timeout-detection methods
857 JH/15 Relax results from ACL control request to enable cutthrough, in
858 unsupported situations, from error to silently (except under debug)
859 ignoring. This covers use with PRDR, frozen messages, queue-only and
862 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
864 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
865 metadata, resulting in a crash in free().
867 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
868 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
869 Broken also in d185889f4, with init system revamp.
871 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
872 Previously we abruptly closed the connection after reading a malware-
873 found indication; now we go on to read the "scan ok" response line,
876 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
877 ACL. Previously, a crash would result.
879 JH/19 Speed up macro lookups during configuration file read, by skipping non-
880 macro text after a replacement (previously it was only once per line) and
881 by skipping builtin macros when searching for an uppercase lead character.
883 JH/20 DANE support moved from Experimental to mainline. The Makefile control
884 for the build is renamed.
886 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
887 was allocated for every new TLS startup, meaning one per message. Fix
888 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
890 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
891 reported the original. Fix to report (as far as possible) the ACL
892 result replacing the original.
894 JH/23 Fix memory leak during multi-message connections using STARTTLS under
895 OpenSSL. Certificate information is loaded for every new TLS startup,
896 and the resources needed to be freed.
898 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
900 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
903 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
904 DATA response info to the (existing) per-recipient response info for
905 the "C=" log element. It can have useful tracking info from the
906 destination system. Patch from Simon Arlott.
908 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
909 length value. Previously this would segfault.
911 HS/02 Support Avast multiline protoocol, this allows passing flags to
912 newer versions of the scanner.
914 JH/28 Ensure that variables possibly set during message acceptance are marked
915 dead before release of memory in the daemon loop. This stops complaints
916 about them when the debug_store option is enabled. Discovered specifically
917 for sender_rate_period, but applies to a whole set of variables.
918 Do the same for the queue-runner and queue-list loops, for variables set
919 from spool message files. Do the same for the SMTP per-message loop, for
920 certain variables indirectly set in ACL operations.
922 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
923 as a multi-recipient message from a mailinglist manager). The coding had
924 an arbitrary cutoff number of characters while checking for more input;
925 enforced by writing a NUL into the buffer. This corrupted long / fast
926 input. The problem was exposed more widely when more pipelineing of SMTP
927 responses was introduced, and one Exim system was feeding another.
928 The symptom is log complaints of SMTP syntax error (NUL chars) on the
929 receiving system, and refused recipients seen by the sending system
930 (propating to people being dropped from mailing lists).
931 Discovered and pinpointed by David Carter.
933 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
934 replaced by the ${authresults } expansion.
936 JH/31 Bug 2257: Fix pipe transport to not use a socket-only syscall.
938 HS/03 Set a handler for SIGTERM and call exit(3) if running as PID 1. This
939 allows proper process termination in container environments.
941 JH/32 Bug 2258: Fix spool_wireformat in combination with LMTP transport.
942 Previously the "final dot" had a newline after it; ensure it is CR,LF.
944 JH/33 SPF: remove support for the "spf" ACL condition outcome values "err_temp"
945 and "err_perm", deprecated since 4.83 when the RFC-defined words
946 "temperror" and "permerror" were introduced.
948 JH/34 Re-introduce enforcement of no cutthrough delivery on transports having
949 transport-filters or DKIM-signing. The restriction was lost in the
950 consolidation of verify-callout and delivery SMTP handling.
951 Extend the restriction to also cover ARC-signing.
953 JH/35 Cutthrough: for a final-dot response timeout (and nonunderstood responses)
954 in defer=pass mode supply a 450 to the initiator. Previously the message
957 PP/02 DANE: add dane_require_tls_ciphers SMTP Transport option; if unset,
958 tls_require_ciphers is used as before.
960 HS/03 Malware Avast: Better match the Avast multiline protocol. Add
961 "pass_unscanned". Only tmpfails from the scanner are written to
962 the paniclog, as they may require admin intervention (permission
963 denied, license issues). Other scanner errors (like decompression
964 bombs) do not cause a paniclog entry.
966 JH/36 Fix reinitialisation of DKIM logging variable between messages.
967 Previously it was possible to log spurious information in receive log
970 JH/37 Bug 2255: Revert the disable of the OpenSSL session caching. This
971 triggered odd behaviour from Outlook Express clients.
973 PP/03 Add util/renew-opendmarc-tlds.sh script for safe renewal of public
976 JH/38 DKIM: accept Ed25519 pubkeys in SubjectPublicKeyInfo-wrapped form,
977 since the IETF WG has not yet settled on that versus the original
978 "bare" representation.
980 JH/39 Fix syslog logging for syslog_timestamp=no and log_selector +millisec.
981 Previously the millisecond value corrupted the output.
982 Fix also for syslog_pid=no and log_selector +pid, for which the pid
983 corrupted the output.
989 JH/01 Rework error string handling in TLS interface so that the caller in
990 more cases is responsible for logging. This permits library-sourced
991 string to be attached to addresses during delivery, and collapses
992 pairs of long lines into single ones.
994 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
995 during configuration. Wildcards are allowed and expanded.
997 JH/02 Rework error string handling in DKIM to pass more info back to callers.
998 This permits better logging.
1000 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
1001 do not close it down and have the child transport start it up again on
1002 the passed-on TCP connection. Instead, proxy the child (and any
1003 subsequent ones) for TLS via a unix-domain socket channel. Logging is
1004 affected: the continued delivery log lines do not have any DNSSEC, TLS
1005 Certificate or OCSP information. TLS cipher information is still logged.
1007 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
1008 identical IP addresses on different listening ports. Will also affect
1011 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
1012 add noisy ifdef guards to special-case this sillyness.
1013 Patch from Bernd Kuhls.
1015 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
1016 than 255 are no longer allowed.
1018 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
1019 Disable the session-cache too, which might reduce our load. Since we
1020 currrectly use a new context for every connection, both as server and
1021 client, there is no benefit for these.
1022 GnuTLS appears to not support tickets server-side by default (we don't
1023 call gnutls_session_ticket_enable_server()) but client side is enabled
1024 by default on recent versions (3.1.3 +) unless the PFS priority string
1027 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
1028 <https://reproducible-builds.org/specs/source-date-epoch/>.
1030 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
1031 the check for any unsuccessful recipients did not notice the limit, and
1032 erroneously found still-pending ones.
1034 JH/08 Pipeline CHUNKING command and data together, on kernels that support
1035 MSG_MORE. Only in-clear (not on TLS connections).
1037 JH/09 Avoid using a temporary file during transport using dkim. Unless a
1038 transport-filter is involved we can buffer the headers in memory for
1039 creating the signature, and read the spool data file once for the
1040 signature and again for transmission.
1042 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
1043 4.77 as the kernel support then wasn't solid, having issues in 64bit
1044 mode. Now, it's been long enough. Add support for FreeBSD also.
1046 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
1047 case where the routing stage had gathered several addresses to send to
1048 a host before calling the transport for the first, we previously failed
1049 to close down TLS in the old transport process before passing the TCP
1050 connection to the new process. The new one sent a STARTTLS command
1051 which naturally failed, giving a failed delivery and bloating the retry
1052 database. Investigation and fix prototype from Wolfgang Breyha.
1054 JH/12 Fix check on SMTP command input synchronisation. Previously there were
1055 false-negatives in the check that the sender had not preempted a response
1056 or prompt from Exim (running as a server), due to that code's lack of
1057 awareness of the SMTP input buffering.
1059 PP/04 Add commandline_checks_require_admin option.
1060 Exim drops privileges sanely, various checks such as -be aren't a
1061 security problem, as long as you trust local users with access to their
1062 own account. When invoked by services which pass untrusted data to
1063 Exim, this might be an issue. Set this option in main configuration
1064 AND make fixes to the calling application, such as using `--` to stop
1067 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
1068 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
1069 responses to those, into a single TLS record each way (this usually means
1070 a single packet). As a side issue, smtp_enforce_sync now works on TLS
1073 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
1074 affects you only if you're dancing at the edge of the param size limits.
1075 If you are, and this message makes sense to you, then: raise the
1076 configured limit or use OpenSSL 1.1. Nothing we can do for older
1079 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
1080 cmdline element to get the documented default one. Previously it was
1083 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
1084 are unsuable for non-message-specific reasons - eg. connection timeout,
1085 banner-time rejection.
1087 JH/16 Fix logging of delivery remote port, when specified by router, under
1090 PP/06 Repair manualroute's ability to take options in any order, even if one
1091 is the name of a transport.
1094 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
1096 JH/17 Change the list-building routines interface to use the expanding-string
1097 triplet model, for better allocation and copying behaviour.
1099 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
1100 Previously it was constructed the first time a possibly-matching string
1101 was met in the configuration file input during startup; now it is done
1104 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
1105 compatible one, to avoid the (poorly documented) possibility of a config
1106 file in the working directory redirecting the DB files, possibly correpting
1107 some existing file. CVE-2017-10140 assigned for BDB.
1109 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
1110 cache-hot. Previously, although the result was properly cached, the
1111 initial verify call returned a defer.
1113 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
1114 the main verify for receipient in uncached-mode.
1116 JH/22 Retire historical build files to an "unsupported" subdir. These are
1117 defined as "ones for which we have no current evidence of testing".
1119 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
1120 if present. Previously it was ignored.
1122 JH/24 Start using specified-initialisers in C structure init coding. This is
1123 a C99 feature (it's 2017, so now considered safe).
1125 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
1126 if was a fixed-sized field and bitmask ops via macros; it is now more
1129 PP/07 GitHub PR 56: Apply MariaDB build fix.
1130 Patch provided by Jaroslav Škarvada.
1132 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
1133 during Coverity cleanups [4.87 JH/47]
1134 Diagnosis and fix provided by Michael Fischer v. Mollard.
1136 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
1137 the right size to place the terminating semicolon on its own folded
1138 line, the header hash was calculated to an incorrect value thanks to
1139 the (relaxed) space the fold became.
1141 HS/02 Fix Bug 2130: large writes from the transport subprocess were chunked
1142 and confused the parent.
1144 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
1145 which could crash as a result. This could lead to undeliverable messages.
1147 JH/28 Logging: "next input sent too soon" now shows where input was truncated
1150 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
1151 matters on fast-turnover and PID-randomising systems, which were getting
1152 out-of-order delivery.
1154 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
1155 a possibly-overlapping copy. The symptom was that "Remote host closed
1156 connection in response to HELO" was logged instead of the actual 4xx
1159 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
1160 Previously only that bufferd was discarded, resulting in SYMTP command
1163 JH/32 DKIM: when a message has multiple signatures matching an identity given
1164 in dkim_verify_signers, run the dkim acl once for each. Previously only
1165 one run was done. Bug 2189.
1167 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
1168 "panic the current process" to "deliberately defer". The panic log is
1169 still written with the problem list name; the mail and reject logs now
1170 get a temp-reject line for the message that was being handled, saying
1171 something like "domains check lookup or other defer". The SMTP 451
1172 message is still "Temporary local problem".
1174 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
1175 A crafted sequence of BDAT commands could result in in-use memory beeing
1176 freed. CVE-2017-16943.
1178 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
1179 from SMTP input. Previously it was always done; now only done for DATA
1180 and not BDAT commands. CVE-2017-16944.
1182 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
1183 to the message (such as an overlong header line). Previously this was
1184 not done and we did not exit BDAT mode. Followon from the previous item
1185 though a different problem.
1191 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
1192 than -2003 did; needs libidn2 in addition to libidn.
1194 JH/02 The path option on a pipe transport is now expanded before use.
1196 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
1197 Patch provided by "Björn", documentation fix added too.
1199 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
1200 missing a wire-to-host endian conversion.
1202 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
1203 close after a BDAT command line could be taken as a following command,
1204 giving a synch failure. Fix by only checking for synch immediately
1205 before acknowledging the chunk.
1207 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
1208 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
1209 macro. Patches provided by Josh Soref.
1211 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
1212 Previously we did not; the RFC seems ambiguous and VRFY is not listed
1213 by IANA as a service extension. However, John Klensin suggests that we
1216 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
1217 the dkim code may be unix-mode line endings rather than smtp wire-format
1218 CRLF, so prepend a CR to any bare LF.
1220 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
1221 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
1223 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
1224 the first were themselves being wrongly included in the feed into dkim
1225 processing; with most chunk sizes in use this resulted in an incorrect
1226 body hash calculated value.
1228 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
1229 DKIM signature block, for verification. Although advised against by
1230 standards it is specifically not ruled illegal.
1232 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
1234 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
1235 missing a body hash (the bh= tag).
1237 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
1238 It seems that HAProxy sends the Proxy Protocol information in clear and
1239 only then does a TLS startup, so do the same.
1241 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
1242 TCP connections (such as for Spamd) unless the daemon successfully set
1243 Fast Open mode on its listening sockets. This fixes breakage seen on
1244 too-old kernels or those not configured for Fast Open, at the cost of
1245 requiring both directions being enabled for TFO, and TFO never being used
1246 by non-daemon-related Exim processes.
1248 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
1249 endings, at least on the first header line. Try to canonify any that get
1250 past that check, despite the cost.
1252 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
1253 now limited to an arbitrary five deep, while parsing addresses with the
1254 strip_excess_angle_brackets option enabled.
1256 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
1257 instead leave the unprompted TLS handshake in socket buffer for the
1258 TLS library to consume.
1260 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
1262 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
1264 JH/16 Drop variables when they go out of scope. Memory management drops a whole
1265 region in one operation, for speed, and this leaves assigned pointers
1266 dangling. Add checks run only under the testsuite which checks all
1267 variables at a store-reset and panics on a dangling pointer; add code
1268 explicitly nulling out all the variables discovered. Fixes one known
1269 bug: a transport crash, where a dangling pointer for $sending_ip_address
1270 originally assigned in a verify callout, is re-used.
1272 PP/06 Drop '.' from @INC in various Perl scripts.
1274 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
1276 PP/08 Reduce a number of compilation warnings under clang; building with
1277 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
1278 should be warning-free.
1280 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
1282 HS/01 Fix portability problems introduced by PP/08 for platforms where
1283 realloc(NULL) is not equivalent to malloc() [SunOS et al].
1285 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
1286 chunk. This allows us to accept broken chunked messages. We need a more
1287 general solution here.
1289 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
1290 already-broken messages in the queue.
1292 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
1294 JH/19 Fix reference counting bug in routing-generated-address tracking.
1300 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
1301 supports it and a size is available (ie. the sending peer gave us one).
1303 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
1304 years of being deprecated). The replacements are the ACLs
1305 acl_smtp_mime and acl_not_smtp_mime.
1307 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
1308 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
1309 or even an in-clear connection were permitted. Now, if the host lookup
1310 was dnssec and dane was requested then the host is only used if the
1311 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
1312 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
1313 if one fails this test.
1314 This means that a poorly-configured remote DNS will make it incommunicado;
1315 but it protects against a DNS-interception attack on it.
1317 JH/04 Bug 1810: make continued-use of an open smtp transport connection
1318 non-noisy when a race steals the message being considered.
1320 JH/05 If main configuration option tls_certificate is unset, generate a
1321 self-signed certificate for inbound TLS connections.
1323 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
1324 in rewrites and routers.
1326 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
1327 and logged a warning sing 4.83; now they are a configuration file error.
1329 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
1330 (lacking @domain). Apply the same qualification processing as RCPT.
1332 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
1334 JH/10 Support ${sha256:} applied to a string (as well as the previous
1337 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
1338 a cutthrough deliver is pending, as we always want to make a connection.
1339 This also avoids re-routing the message when later placing the cutthrough
1340 connection after a verify cache hit.
1341 Do not update it with the verify result either.
1343 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
1344 when routing results in more than one destination address.
1346 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
1347 signing (which inhibits the cutthrough capability). Previously only
1348 the presence of an option was tested; now an expansion evaluating as
1349 empty is permissible (obviously it should depend only on data available
1350 when the cutthrough connection is made).
1352 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
1353 the relevant preceding SMTP command did not note the pipelining mode.
1355 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
1356 Previously they were not counted.
1358 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
1359 as one having no matching records. Previously we deferred the message
1360 that needed the lookup.
1362 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
1363 distinguished as "(=".
1365 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
1366 for missing MX records. Previously it only worked for missing A records.
1368 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
1370 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
1371 after the data-go-ahead and data-ack. Patch from Jason Betts.
1373 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
1374 even for a "none" policy. Patch from Tony Meyer.
1376 JH/22 Fix continued use of a connection for further deliveries. If a port was
1377 specified by a router, it must also match for the delivery to be
1380 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
1381 When one of the recipients of a message was unsuitable for the connection
1382 (has no matching addresses), we lost track of needing to mark it
1383 deferred. As a result mail would be lost.
1385 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
1387 JH/25 Decoding ACL controls is now done using a binary search; the source code
1388 takes up less space and should be simpler to maintain. Merge the ACL
1389 condition decode tables also, with similar effect.
1391 JH/26 Fix problem with one_time used on a redirect router which returned the
1392 parent address unchanged. A retry would see the parent address marked as
1393 delivered, so not attempt the (identical) child. As a result mail would
1396 JH/27 Fix a possible security hole, wherein a process operating with the Exim
1397 UID can gain a root shell. Credit to http://www.halfdog.net/ for
1398 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
1401 JH/28 Enable {spool,log} filesystem space and inode checks as default.
1402 Main config options check_{log,spool}_{inodes,space} are now
1403 100 inodes, 10MB unless set otherwise in the configuration.
1405 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
1406 Previously it only applied to the main-section connection policy
1409 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
1411 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
1412 by me. Added RFC7919 DH primes as an alternative.
1414 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
1415 are not in the system include path.
1417 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
1418 GnuTLS, when a session startup failed (eg because the client disconnected)
1419 Exim did stdio operations after fclose. This was exposed by a recent
1420 change which nulled out the file handle after the fclose.
1422 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
1423 signed directly by the cert-signing cert, rather than an intermediate
1424 OCSP-signing cert. This is the model used by LetsEncrypt.
1426 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
1428 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
1429 an incoming connection.
1431 HS/02 Bug 1802: Do not half-close the connection after sending a request
1434 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
1435 fallback to "prime256v1".
1437 JH/34 SECURITY: Use proper copy of DATA command in error message.
1438 Could leak key material. Remotely exploitable. CVE-2016-9963.
1444 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
1445 and 3.4.4 - once the server is enabled to respond to an OCSP request
1446 it does even when not requested, resulting in a stapling non-aware
1447 client dropping the TLS connection.
1449 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
1450 support variable-length bit vectors. No functional change.
1452 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
1453 The I= interface field on outgoing lines is now after the H= remote
1454 host field, same as incoming lines. There is a separate
1455 outgoing_interface log selector which allows you to disable the
1458 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
1459 If not running log_selector +smtp_connection the mainlog would be held
1460 open indefinitely after a "too many connections" event, including to a
1461 deleted file after a log rotate. Leave the per net connection logging
1462 leaving it open for efficiency as that will be quickly detected by the
1463 check on the next write.
1465 HS/01 Bug 1671: Fix post transport crash.
1466 Processing the wait-<transport> messages could crash the delivery
1467 process if the message IDs didn't exist for some reason. When
1468 using 'split_spool_directory=yes' the construction of the spool
1469 file name failed already, exposing the same netto behaviour.
1471 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
1472 mime_regex ACL conditions.
1474 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
1475 to DSN fail messages (bounces): remote IP, remote greeting, remote response
1476 to HELO, local diagnostic string.
1478 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
1479 log line to debug. Even when configured with a tls authenticator many
1480 client connections are expected to not authenticate in this way, so
1481 an authenticate fail is not an error.
1483 HS/02 Add the Exim version string to the process info. This way exiwhat
1484 gives some more detail about the running daemon.
1486 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
1487 matter for fast-change records such as DNSBLs.
1489 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
1490 retry record, even if constant. There may be multiple transports with
1491 different interface settings and the retry behaviour needs to be kept
1494 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
1496 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
1498 JH/10 Bug 840: fix log_defer_output option of pipe transport
1500 JH/11 Bug 830: use same host for all RCPTS of a message, even under
1501 hosts_randomize. This matters a lot when combined with mua_wrapper.
1503 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
1504 ${quote_pgsql:<string>} operator.
1506 JH/13 Bug 1708: avoid misaligned access in cached lookup.
1508 JH/14 Change header file name for freeradius-client. Relevant if compiling
1509 with Radius support; from the Gentoo tree and checked under Fedora.
1511 JH/15 Bug 1712: Introduce $prdr_requested flag variable
1513 JH/16 Bug 1714: Permit an empty string as expansion result for transport
1514 option transport_filter, meaning no filtering.
1516 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
1518 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
1519 defaults to "*" (all hosts). The variable is now available when not built
1520 with TLS, default unset, mainly to enable keeping the testsuite sane.
1521 If a server certificate is not supplied (via tls_certificate) an error is
1522 logged, and clients will find TLS connections fail on startup. Presumably
1523 they will retry in-clear.
1524 Packagers of Exim are strongly encouraged to create a server certificate
1525 at installation time.
1527 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
1528 with the $config_file variable.
1530 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
1531 in transport context, after the attempt, and per-recipient. The latter type
1532 is per host attempted. The event data is the error message, and the errno
1533 information encodes the lookup type (A vs. MX) used for the (first) host,
1534 and the trailing two digits of the smtp 4xx response.
1536 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
1537 to write to mainlog (or rejectlog, paniclog) in the window between file
1538 creation and permissions/ownership being changed. Particularly affects
1539 installations where exicyclog is run as root, rather than exim user;
1540 result is that the running daemon panics and dies.
1542 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
1544 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
1545 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
1546 "pri" and "weight". Note that the previous implicit priority given by the
1547 list order is no longer honoured.
1549 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
1550 for DKIM processing.
1552 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
1553 by defining SUPPORT_SOCKS.
1555 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
1556 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
1557 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
1558 variables are renamed to proxy_{local,external}_{address,port}.
1560 JH/27 Move Internationalisation support from Experimental to mainline, enabled
1561 for a build by defining SUPPORT_I18N
1563 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
1564 of the query string, and make ${quote_redis:} do that quoting.
1566 JH/29 Move Events support from Experimental to mainline, enabled by default
1567 and removable for a build by defining DISABLE_EVENT.
1569 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
1571 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
1572 cached by the daemon.
1574 JH/32 Move Redis support from Experimental to mainline, enabled for a build
1575 by defining LOOKUP_REDIS. The libhiredis library is required.
1577 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
1578 keys are given for lookup.
1580 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
1581 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
1582 only supported when built with TLS support. The PolarSSL SHA routines
1583 are still used when the TLS library is too old for convenient support.
1585 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
1586 openssl_options), for security. OpenSSL forces this from version 1.1.0
1587 server-side so match that on older versions.
1589 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
1590 allocation for $value could be released as the expansion processing
1591 concluded, but leaving the global pointer active for it.
1593 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
1594 and to use the domains and local_parts ACL conditions.
1596 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
1597 incorrectly not doubled on cutthrough transmission, hence seen as a
1598 body-termination at the receiving system - resulting in truncated mails.
1599 Commonly the sender saw a TCP-level error, and retransmitted the message
1600 via the normal store-and-forward channel. This could result in duplicates
1601 received - but deduplicating mailstores were liable to retain only the
1602 initial truncated version.
1604 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
1606 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
1608 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
1609 we're in there, support oversigning also; bug 1309.
1611 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
1613 HS/04 Add support for keep_environment and add_environment options.
1615 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
1616 either intentional arithmetic overflow during PRNG, or testing config-
1619 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
1620 delivery resulted in actual delivery. Cancel cutthrough before DATA
1623 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
1624 rejects a recipient: pass the reject to the originator.
1626 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
1627 Many were false-positives and ignorable, but it's worth fixing the
1630 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
1631 for the new environment-manipulation done at startup. Move the routines
1632 from being local to tls.c to being global via the os.c file.
1634 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
1635 an extract embedded as result-arg for a map, the first arg for extract
1636 is unavailable so we cannot tell if this is a numbered or keyed
1637 extraction. Accept either.
1643 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
1646 JH/02 The smtp transport option "multi_domain" is now expanded.
1648 JH/03 The smtp transport now requests PRDR by default, if the server offers
1651 JH/04 Certificate name checking on server certificates, when exim is a client,
1652 is now done by default. The transport option tls_verify_cert_hostnames
1653 can be used to disable this per-host. The build option
1654 EXPERIMENTAL_CERTNAMES is withdrawn.
1656 JH/05 The value of the tls_verify_certificates smtp transport and main options
1657 default to the word "system" to access the system default CA bundle.
1658 For GnuTLS, only version 3.0.20 or later.
1660 JH/06 Verification of the server certificate for a TLS connection is now tried
1661 (but not required) by default. The verification status is now logged by
1662 default, for both outbound TLS and client-certificate supplying inbound
1665 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
1668 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
1669 Status Notification (bounce) messages are now MIME format per RFC 3464.
1670 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
1671 under the control of the dsn_advertise_hosts option, and routers may
1672 have a dsn_lasthop option.
1674 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
1675 default, modifiable by a malware= option. The list separator for
1676 the options can now be changed in the usual way. Bug 68.
1678 JH/10 The smtp_receive_timeout main option is now expanded before use.
1680 JH/11 The incoming_interface log option now also enables logging of the
1681 local interface on delivery outgoing connections.
1683 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
1684 if the interface and destination host and port all match.
1686 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
1689 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
1690 Patch from Andrew Lewis.
1692 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
1693 now supports optional time-restrictions, weighting, and priority
1694 modifiers per server. Patch originally by <rommer@active.by>.
1696 JH/16 The spamd_address main option now supports a mixed list of local
1697 and remote servers. Remote servers can be IPv6 addresses, and
1698 specify a port-range.
1700 JH/17 Bug 68: The spamd_address main option now supports an optional
1701 timeout value per server.
1703 JH/18 Bug 1581: Router and transport options headers_add/remove can
1704 now have the list separator specified.
1706 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
1709 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
1712 JH/21 Support for the A6 type of dns record is withdrawn.
1714 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
1715 rather than the verbs used.
1717 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
1718 from 255 to 1024 chars.
1720 JH/24 Verification callouts now attempt to use TLS by default.
1722 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
1723 are generic router options now. The defaults didn't change.
1725 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
1726 Original patch from Alexander Shikoff, worked over by JH.
1728 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
1729 files if ZCAT_COMMAND is not executable.
1731 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
1733 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
1735 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
1736 Normally benign, it bites when the pair was led to by a CNAME;
1737 modern usage is to not canonicalize the domain to a CNAME target
1738 (and we were inconsistent anyway for A-only vs AAAA+A).
1740 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
1742 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
1743 when evaluating $sender_host_dnssec.
1745 JH/31 Check the HELO verification lookup for DNSSEC, adding new
1746 $sender_helo_dnssec variable.
1748 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
1750 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
1752 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
1754 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
1755 documented as working, but never had. Support all but $spam_report.
1757 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
1758 added for tls authenticator.
1760 HS/03 Add perl_taintmode main config option
1766 TL/01 When running the test suite, the README says that variables such as
1767 no_msglog_check are global and can be placed anywhere in a specific
1768 test's script, however it was observed that placement needed to be near
1769 the beginning for it to behave that way. Changed the runtest perl
1770 script to read through the entire script once to detect and set these
1771 variables, reset to the beginning of the script, and then run through
1772 the script parsing/test process like normal.
1774 TL/02 The BSD's have an arc4random API. One of the functions to induce
1775 adding randomness was arc4random_stir(), but it has been removed in
1776 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
1777 function when detected.
1779 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
1780 cause callback expansion.
1782 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
1783 syntax errors in an expansion can be treated as a string instead of
1784 logging or causing an error, due to the internal use of bool_lax
1785 instead of bool when processing it.
1787 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
1788 server certificates when making smtp deliveries.
1790 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
1792 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
1794 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
1796 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
1797 Merged patch from Sebastian Wiedenroth.
1799 JH/05 Fix results-pipe from transport process. Several recipients, combined
1800 with certificate use, exposed issues where response data items split
1801 over buffer boundaries were not parsed properly. This eventually
1802 resulted in duplicates being sent. This issue only became common enough
1803 to notice due to the introduction of connection certificate information,
1804 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1806 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1807 size buffer was used, resulting in syntax errors when an expansion
1810 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1811 version 3.3.6 or later.
1813 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1814 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1815 both become "event_action", the variables become $event_name, $event_data
1816 and $event_defer_errno. There is a new variable $verify_mode, usable in
1817 routers, transports and related events. The tls:cert event is now also
1818 raised for inbound connections, if the main configuration event_action
1821 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1822 early OCSP support, but no stapling (appears to be less than 1.0.0).
1824 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1825 server certificate names available under the smtp transport option
1826 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1829 JH/10 Time-related extraction expansions from certificates now use the main
1830 option "timezone" setting for output formatting, and are consistent
1831 between OpenSSL and GnuTLS compilations. Bug 1541.
1833 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1834 encoded parameter in the incoming message. Bug 1558.
1836 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1837 include certificate info, eximon was claiming there were spoolfile
1840 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1842 JH/14 Log delivery-related information more consistently, using the sequence
1843 "H=<name> [<ip>]" wherever possible.
1845 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1846 are problematic for Debian distribution, omit them from the release
1849 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1851 JH/16 Fix string representation of time values on 64bit time_t architectures.
1854 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1855 output list separator was used.
1860 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1861 checkers that were complaining about end of non-void function with no
1864 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1865 This was a regression introduced in 4.83 by another bugfix.
1867 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1869 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1870 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1876 TF/01 Correctly close the server side of TLS when forking for delivery.
1878 When a message was received over SMTP with TLS, Exim failed to clear up
1879 the incoming connection properly after forking off the child process to
1880 deliver the message. In some situations the subsequent outgoing
1881 delivery connection happened to have the same fd number as the incoming
1882 connection previously had. Exim would try to use TLS and fail, logging
1883 a "Bad file descriptor" error.
1885 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1886 utilities have not been installed.
1888 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1889 temporary space as the ACL may create new global variables.
1891 TL/01 LDAP support uses per connection or global context settings, depending
1892 upon the detected version of the libraries at build time.
1894 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1895 to extract and use the src ip:port in logging and expansions as if it
1896 were a direct connection from the outside internet. PPv2 support was
1897 updated based on HAProxy spec change in May 2014.
1899 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1901 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1902 Properly escape header and check for NULL return.
1904 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1907 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1909 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1910 characters in header names, implemented as a verify condition.
1911 Contributed by Michael Fischer v. Mollard.
1913 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1914 results permerror and temperror. Previous values are deprecated but
1915 still accepted. In a future release, err_perm and err_temp will be
1916 completely removed, which will be a backward incompatibility if the
1917 ACL tests for either of these two old results. Patch contributed by
1918 user bes-internal on the mailing list.
1920 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1922 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1923 selectors, in both main and reject logs.
1925 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1928 JH/07 Add malware type "sock" for talking to simple daemon.
1930 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1932 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1933 routers/transports under cutthrough routing.
1935 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1936 numbers. Touch up "bool" conditional to keep the same definition.
1938 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1940 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1942 JH/12 Expand items in router/transport headers_add or headers_remove lists
1943 individually rather than the list as a whole. Bug 1452.
1945 Required for reasonable handling of multiple headers_ options when
1946 they may be empty; requires that headers_remove items with embedded
1947 colons must have them doubled (or the list-separator changed).
1949 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1950 view the policy declared in the DMARC record. Currently, $dmarc_status
1951 is a combined value of both the record presence and the result of the
1954 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1956 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1957 dnslookup router and the smtp transport (applying to the forward
1960 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1961 of ldap servers used for a specific lookup. Patch provided by Heiko
1964 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1965 New variable $lookup_dnssec_authenticated for observability.
1967 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1968 Patch submitted by Lars Timman.
1970 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1972 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1973 Requires trusted mode and valid format message id, aborts otherwise.
1974 Patch contributed by Heiko Schlichting.
1976 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1977 certextract with support for various fields. Bug 1358.
1979 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1980 is requested by default, modifiable by smtp transport option
1983 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1984 operate on certificate variables to give certificate fingerprints
1985 Also new ${sha256:cert_variable}.
1987 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1989 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1990 Christian Aistleitner.
1992 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1994 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1995 file. Patch from Wolfgang Breyha.
1997 JH/25 Expand the coverage of the delivery $host and $host_address to
1998 client authenticators run in verify callout. Bug 1476.
2000 JH/26 Port service names are now accepted for tls_on_connect_ports, to
2001 align with daemon_smtp_ports. Bug 72.
2003 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
2004 support and error reporting did not work properly.
2006 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
2007 and is readable. Patch from Andrew Colin Kissa.
2009 TL/14 Enhance documentation of ${run expansion and how it parses the
2010 commandline after expansion, particularly in the case when an
2011 unquoted variable expansion results in an empty value.
2013 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
2015 PP/02 Fix internal collision of T_APL on systems which support RFC3123
2016 by renaming away from it. Addresses GH issue 15, reported by
2019 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
2021 TL/15 SECURITY: prevent double expansion in math comparison functions
2022 (can expand unsanitized data). Not remotely exploitable.
2029 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
2031 PP/02 Make -n do something, by making it not do something.
2032 When combined with -bP, the name of an option is not output.
2034 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
2037 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
2038 $sender_host_name and config options to manage this, and basic check
2041 PP/05 DSCP support for outbound connections and control modifier for inbound.
2043 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
2044 (Only plugin which currently uses this is kerberos4, which nobody should
2045 be using, but we should make it available and other future plugins might
2046 conceivably use it, even though it would break NAT; stuff *should* be
2047 using channel bindings instead).
2049 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
2050 name; added for Sendmail compatibility; requires admin caller.
2051 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
2052 just ignore it); requires trusted caller.
2053 Also parse but ignore: -Ac -Am -X<logfile>
2056 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
2058 TL/02 Add +smtp_confirmation as a default logging option.
2060 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
2061 Patch by Magnus Holmgren from 2007-02-20.
2063 TL/04 Bugzilla 1281 - Spec typo.
2064 Bugzilla 1283 - Spec typo.
2065 Bugzilla 1290 - Spec grammar fixes.
2067 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
2069 TL/06 Add Experimental DMARC support using libopendmarc libraries.
2071 TL/07 Fix an out of order global option causing a segfault. Reported to dev
2072 mailing list by by Dmitry Isaikin.
2074 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
2076 JH/02 Support "G" suffix to numbers in ${if comparisons.
2078 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
2080 NM/01 Bugzilla 1197 - Spec typo
2081 Bugzilla 1196 - Spec examples corrections
2083 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
2085 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
2086 gnutls_enable_pkcs11, but renamed to more accurately indicate its
2089 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
2090 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
2092 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
2093 "acl {{name}{arg}...}", and optional args on acl condition
2096 JH/05 Permit multiple router/transport headers_add/remove lines.
2098 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
2100 JH/07 Avoid using a waiting database for a single-message-only transport.
2101 Performance patch from Paul Fisher. Bugzilla 1262.
2103 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
2106 JH/09 Add $headers_added variable, with content from use of ACL modifier
2107 add_header (but not yet added to the message). Bugzilla 199.
2109 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
2110 Pulled from Bugzilla 817 by Wolfgang Breyha.
2112 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
2114 (nb: this is the same fix as in Exim 4.80.1)
2116 JH/11 Add A= logging on delivery lines, and a client_set_id option on
2119 JH/12 Add optional authenticated_sender logging to A= and a log_selector
2122 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
2124 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
2125 advertise SMTP AUTH mechanism to us, instead of a generic
2126 protocol violation error. Also, make Exim more robust to bad
2127 data from the Dovecot auth socket.
2129 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
2131 When a queue runner is handling a message, Exim first routes the
2132 recipient addresses, during which it prunes them based on the retry
2133 hints database. After that it attempts to deliver the message to
2134 any remaining recipients. It then updates the hints database using
2137 So if a recipient address works intermittently, it can get repeatedly
2138 deferred at routing time. The retry hints record remains fresh so the
2139 address never reaches the final cutoff time.
2141 This is a fairly common occurrence when a user is bumping up against
2142 their storage quota. Exim had some logic in its local delivery code
2143 to deal with this. However it did not apply to per-recipient defers
2144 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
2146 This change adds a proper retry rule check during routing so that the
2147 final cutoff time is checked against the message's age. We only do
2148 this check if there is an address retry record and there is not a
2149 domain retry record; this implies that previous attempts to handle
2150 the address had the retry_use_local_parts option turned on. We use
2151 this as an approximation for the destination being like a local
2152 delivery, as in LMTP.
2154 I suspect this new check makes the old local delivery cutoff check
2155 redundant, but I have not verified this so I left the code in place.
2157 TF/02 Correct gecos expansion when From: is a prefix of the username.
2159 Test 0254 submits a message to Exim with the header
2163 When I ran the test suite under the user fanf2, Exim expanded
2164 the header to contain my full name, whereas it should have added
2165 a Resent-Sender: header. It erroneously treats any prefix of the
2166 username as equal to the username.
2168 This change corrects that bug.
2170 GF/01 DCC debug and logging tidyup
2171 Error conditions log to paniclog rather than rejectlog.
2172 Debug lines prefixed by "DCC: " to remove any ambiguity.
2174 TF/03 Avoid unnecessary rebuilds of lookup-related code.
2176 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
2177 Bug spotted by Jeremy Harris; was flawed since initial commit.
2178 Would have resulted in OCSP responses post-SNI triggering an Exim
2179 NULL dereference and crash.
2181 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
2183 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
2184 Bug detection, analysis and fix by Samuel Thibault.
2185 Bugzilla 1331, Debian bug #698092.
2187 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
2189 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
2190 Server implementation by Todd Lyons, client by JH.
2191 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
2192 config variable "prdr_enable" controls whether the server
2193 advertises the facility. If the client requests PRDR a new
2194 acl_data_smtp_prdr ACL is called once for each recipient, after
2195 the body content is received and before the acl_smtp_data ACL.
2196 The client is controlled by both of: a hosts_try_prdr option
2197 on the smtp transport, and the server advertisement.
2198 Default client logging of deliveries and rejections involving
2199 PRDR are flagged with the string "PRDR".
2201 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
2202 fclose(). Diagnosis by Todd Lyons.
2204 PP/17 Update configure.default to handle IPv6 localhost better.
2205 Patch by Alain Williams (plus minor tweaks).
2208 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
2209 This is now consistent with GnuTLS, and is now documented: the
2210 previous undocumented portable approach to treating the option as
2211 unset was to force an expansion failure. That still works, and
2212 an empty string is now equivalent.
2214 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
2215 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
2216 not performing validation itself.
2218 PP/20 Added force_command boolean option to pipe transport.
2219 Patch from Nick Koston, of cPanel Inc.
2221 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
2224 TF/04 Added udpsend ACL modifier and hexquote expansion operator
2226 PP/21 Fix eximon continuous updating with timestamped log-files.
2227 Broken in a format-string cleanup in 4.80, missed when I repaired the
2228 other false fix of the same issue.
2229 Report and fix from Heiko Schlichting.
2232 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
2233 Report from Prashanth Katuri.
2235 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
2236 It's SecureTransport, so affects any MacOS clients which use the
2237 system-integrated TLS libraries, including email clients.
2239 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
2240 using a MIME ACL for non-SMTP local injection.
2241 Report and assistance in diagnosis by Warren Baker.
2243 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
2245 JH/16 Fix comparisons for 64b. Bugzilla 1385.
2247 TL/09 Add expansion variable $authenticated_fail_id to keep track of
2248 last id that failed so it may be referenced in subsequent ACL's.
2250 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
2253 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
2254 ldap library initialization, allowing self-signed CA's to be
2255 used. Also properly sets require_cert option later in code by
2256 using NULL (global ldap config) instead of ldap handle (per
2257 session). Bug diagnosis and testing by alxgomz.
2259 TL/12 Enhanced documentation in the ratelimit.pl script provided in
2260 the src/util/ subdirectory.
2262 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
2263 renamed to Transport Post Delivery Action by Jeremy Harris, as
2266 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
2267 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
2268 redis_servers = needs to be configured which will be used by the redis
2269 lookup. Patch from Warren Baker, of The Packet Hub.
2271 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
2273 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
2274 hostname or reverse DNS when processing a host list. Used suggestions
2275 from multiple comments on this bug.
2277 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
2279 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
2280 Missed a few lines, added it to make the runtest require no keyboard
2283 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
2284 contains upper case chars. Make router use caseful_local_part.
2286 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
2287 support when GnuTLS has been built with p11-kit.
2293 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
2295 This, or similar/improved, will also be change PP/11 of 4.82.
2301 PP/01 Handle short writes when writing local log-files.
2302 In practice, only affects FreeBSD (8 onwards).
2303 Bugzilla 1053, with thanks to Dmitry Isaikin.
2305 NM/01 Bugzilla 949 - Documentation tweak
2307 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
2310 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
2312 PP/02 Implemented gsasl authenticator.
2314 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
2316 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
2317 `pkg-config foo` for cflags/libs.
2319 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
2320 with rest of GSASL and with heimdal_gssapi.
2322 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
2323 `pkg-config foo` for cflags/libs for the TLS implementation.
2325 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
2326 properties get this fed in as external SSF. A number of robustness
2327 and debugging improvements to the cyrus_sasl authenticator.
2329 PP/08 cyrus_sasl server now expands the server_realm option.
2331 PP/09 Bugzilla 1214 - Log authentication information in reject log.
2332 Patch by Jeremy Harris.
2334 PP/10 Added dbmjz lookup type.
2336 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
2338 PP/12 MAIL args handles TAB as well as SP, for better interop with
2339 non-compliant senders.
2340 Analysis and variant patch by Todd Lyons.
2342 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
2343 Bug report from Lars Müller <lars@samba.org> (via SUSE),
2344 Patch from Dirk Mueller <dmueller@suse.com>
2346 PP/13 tls_peerdn now print-escaped for spool files.
2347 Observed some $tls_peerdn in wild which contained \n, which resulted
2348 in spool file corruption.
2350 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
2351 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
2352 or write after TLS renegotiation, which otherwise led to messages
2355 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
2356 as a tracking header (ie: a signed header comes before the signature).
2357 Patch from Wolfgang Breyha.
2359 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
2360 comma-sep list; embedded commas doubled.
2362 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
2364 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
2366 Report and patch from Dmitry Banschikov.
2368 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
2369 Removed SSL_clear() after SSL_new() which led to protocol negotiation
2370 failures. We appear to now support TLS1.1+ with Exim.
2372 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
2373 lets Exim select keys and certificates based upon TLS SNI from client.
2374 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
2375 before an outbound SMTP session. New log_selector, +tls_sni.
2377 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
2378 NULL dereference. Report and patch from Alun Jones.
2380 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
2381 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
2382 Not seeing resolver debug output on NetBSD, but suspect this is a
2383 resolver implementation change.
2385 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
2386 Left warnings. Added "eximon gdb" invocation mode.
2388 PP/21 Defaulting "accept_8bitmime" to true, not false.
2390 PP/22 Added -bw for inetd wait mode support.
2392 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
2393 locate the relevant includes and libraries. Made this the default.
2395 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
2396 Bugzilla 1246, report and most of solution from Tomasz Kusy.
2398 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
2399 This may cause build issues on older platforms.
2401 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
2402 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
2403 gnutls_require_mac & gnutls_require_protocols (no longer supported).
2404 Added SNI support via GnuTLS too.
2405 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
2407 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
2409 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
2410 Applied second patch from Janne, implementing suggestion to default
2411 multiple-strings-in-record handling to match SPF spec.
2413 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
2415 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
2416 read-only, out of scope).
2417 Patch from Wolfgang Breyha, report from Stuart Northfield.
2419 PP/29 Fix three issues highlighted by clang analyser static analysis.
2420 Only crash-plausible issue would require the Cambridge-specific
2421 iplookup router and a misconfiguration.
2422 Report from Marcin Mirosław.
2424 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
2426 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
2427 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
2428 As part of this, removing so much warning spew let me fix some minor
2429 real issues in debug logging.
2431 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
2432 assignment on my part. Fixed.
2434 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
2435 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
2436 Janne Snabb (who went above and beyond: thank you).
2438 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
2439 string otherwise requires a connection and a bunch more work and it's
2440 relatively easy to get wrong. Should also expose TLS library linkage
2443 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
2444 64-bit ${eval} (JH/03).
2446 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
2447 GNU libc to support some of the 64-bit stuff, should not lead to
2448 conflicts. Defined before os.h is pulled in, so if a given platform
2449 needs to override this, it can.
2451 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
2452 protection layer was required, which is not implemented.
2453 Bugzilla 1254, patch from Wolfgang Breyha.
2455 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
2456 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
2457 tls_dhparam take prime identifiers. Also unbreak combination of
2458 OpenSSL+DH_params+TLSSNI.
2460 PP/39 Disable SSLv2 by default in OpenSSL support.
2466 PP/01 Solaris build fix for Oracle's LDAP libraries.
2467 Bugzilla 1109, patch from Stephen Usher.
2469 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
2471 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
2474 TF/02 Fix a couple more cases where we did not log the error message
2475 when unlink() failed. See also change 4.74-TF/03.
2477 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
2478 lock up or crash if it happened to be inside a call to libc when it
2479 got a SIGUSR1 from exiwhat.
2481 The SIGUSR1 handler appends the current process status to the process
2482 log which is later printed by exiwhat. It used to use the general
2483 purpose logging code to do this, but several functions it calls are
2484 not safe for signals.
2486 The new output code in the SIGUSR1 handler is specific to the process
2487 log, and simple enough that it's easy to inspect for signal safety.
2488 Removing some special cases also simplifies the general logging code.
2489 Removing the spurious timestamps from the process log simplifies
2492 TF/04 Improved ratelimit ACL condition.
2494 The /noupdate option has been deprecated in favour of /readonly which
2495 has clearer semantics. The /leaky, /strict, and /readonly update modes
2496 are mutually exclusive. The update mode is no longer included in the
2497 database key; it just determines when the database is updated. (This
2498 means that when you upgrade Exim will forget old rate measurements.)
2500 Exim now checks that the per_* options are used with an update mode that
2501 makes sense for the current ACL. For example, when Exim is processing a
2502 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
2503 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
2504 must specify per_mail/readonly. If you omit the update mode it defaults to
2505 /leaky where that makes sense (as before) or /readonly where required.
2507 The /noupdate option is now undocumented but still supported for
2508 backwards compatibility. It is equivalent to /readonly except that in
2509 ACLs where /readonly is required you may specify /leaky/noupdate or
2510 /strict/noupdate which are treated the same as /readonly.
2512 A useful new feature is the /count= option. This is a generalization
2513 of the per_byte option, so that you can measure the throughput of other
2514 aggregate values. For example, the per_byte option is now equivalent
2515 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
2517 The per_rcpt option has been generalized using the /count= mechanism
2518 (though it's more complicated than the per_byte equivalence). When it is
2519 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
2520 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
2521 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
2522 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
2523 Note that using per_rcpt with a non-readonly update mode in more than
2524 one ACL will cause the recipients to be double-counted. (The per_mail
2525 and per_byte options don't have this problem.)
2527 The handling of very low rates has changed slightly. If the computed rate
2528 is less than the event's count (usually one) then this event is the first
2529 after a long gap. In this case the rate is set to the same as this event's
2530 count, so that the first message of a spam run is counted properly.
2532 The major new feature is a mechanism for counting the rate of unique
2533 events. The new per_addr option counts the number of different
2534 recipients that someone has sent messages to in the last time period. It
2535 behaves like per_rcpt if all the recipient addresses are different, but
2536 duplicate recipient addresses do not increase the measured rate. Like
2537 the /count= option this is a general mechanism, so the per_addr option
2538 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
2539 example, measure the rate that a client uses different sender addresses
2540 with the options per_mail/unique=$sender_address. There are further
2541 details in the main documentation.
2543 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
2545 TF/06 Removed a few PCRE remnants.
2547 TF/07 Automatically extract Exim's version number from tags in the git
2548 repository when doing development or release builds.
2550 PP/02 Raise smtp_cmd_buffer_size to 16kB.
2551 Bugzilla 879. Patch from Paul Fisher.
2553 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
2554 Heavily based on revision 40f9a89a from Simon Arlott's tree.
2557 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
2559 PP/05 Variable $av_failed, true if the AV scanner deferred.
2560 Bugzilla 1078. Patch from John Horne.
2562 PP/06 Stop make process more reliably on build failure.
2563 Bugzilla 1087. Patch from Heiko Schlittermann.
2565 PP/07 Make maildir_use_size_file an _expandable_ boolean.
2566 Bugzilla 1089. Patch from Heiko Schlittermann.
2568 PP/08 Handle ${run} returning more data than OS pipe buffer size.
2569 Bugzilla 1131. Patch from Holger Weiß.
2571 PP/09 Handle IPv6 addresses with SPF.
2572 Bugzilla 860. Patch from Wolfgang Breyha.
2574 PP/10 GnuTLS: support TLS 1.2 & 1.1.
2576 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
2579 PP/11 match_* no longer expand right-hand-side by default.
2580 New compile-time build option, EXPAND_LISTMATCH_RHS.
2581 New expansion conditions, "inlist", "inlisti".
2583 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
2585 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
2587 PP/14 fix log_write() format string regression from TF/03.
2588 Bugzilla 1152. Patch from Dmitry Isaikin.
2594 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
2596 PP/02 Harmonised TLS library version reporting; only show if debugging.
2597 Layout now matches that introduced for other libraries in 4.74 PP/03.
2599 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
2601 PP/04 New "dns_use_edns0" global option.
2603 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
2606 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
2607 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
2609 TK/01 Updated PolarSSL code to 0.14.2.
2610 Bugzilla 1097. Patch from Andreas Metzler.
2612 PP/07 Catch divide-by-zero in ${eval:...}.
2613 Fixes bugzilla 1102.
2615 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
2618 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
2619 format-string attack -- SECURITY: remote arbitrary code execution.
2621 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
2622 time unintentionally subject to list matching rules, letting the header
2623 cause arbitrary Exim lookups (of items which can occur in lists, *not*
2624 arbitrary string expansion). This allowed for information disclosure.
2626 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
2627 INT_MIN/-1 -- value coerced to INT_MAX.
2633 NM/01 Workaround for PCRE version dependency in version reporting
2636 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
2637 This fixes portability to compilers other than gcc, notably
2638 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
2640 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
2641 makefiles for portability to HP-UX and POSIX correctness.
2643 PP/01 Permit LOOKUP_foo enabling on the make command-line.
2644 Also via indented variable definition in the Makefile.
2645 (Debugging by Oliver Heesakkers).
2647 PP/02 Restore caching of spamd results with expanded spamd_address.
2648 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
2650 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
2651 Improves build reliability. Fix from: Frank Elsner
2653 NM/02 Fix wide character breakage in the rfc2047 coding
2654 Fixes bug 1064. Patch from Andrey N. Oktyabrski
2656 NM/03 Allow underscore in dnslist lookups
2657 Fixes bug 1026. Patch from Graeme Fowler
2659 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
2660 Code patches from Adam Ciarcinski of NetBSD.
2662 NM/04 Fixed exiqgrep to cope with mailq missing size issue
2665 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
2666 is logged, to avoid truncation. Patch from John Horne.
2668 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
2669 Patch from Jakob Hirsch.
2671 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
2672 SQL string expansion failure details.
2673 Patch from Andrey Oktyabrski.
2675 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
2676 Patch from Simon Arlott.
2678 PP/09 New lookups functionality failed to compile on old gcc which rejects
2679 extern declarations in function scope.
2680 Patch from Oliver Fleischmann
2682 PP/10 Use sig_atomic_t for flags set from signal handlers.
2683 Check getgroups() return and improve debugging.
2684 Fixed developed for diagnosis in bug 927 (which turned out to be
2687 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
2688 Patch from Mark Zealey.
2690 PP/12 Bugzilla 1056: Improved spamd server selection.
2691 Patch from Mark Zealey.
2693 PP/13 Bugzilla 1086: Deal with maildir quota file races.
2694 Based on patch from Heiko Schlittermann.
2696 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
2697 Patch from Uwe Doering, sign-off by Michael Haardt.
2699 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
2700 variable declaration deep within a block. Bug and patch from
2703 PP/15 lookups-Makefile IRIX compatibility coercion.
2705 PP/16 Make DISABLE_DKIM build knob functional.
2707 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
2708 Patch by Simon Arlott
2710 TF/03 Fix valgrind.h portability to C89 compilers that do not support
2711 variable argument macros. Our copy now differs from upstream.
2717 TF/01 Failure to get a lock on a hints database can have serious
2718 consequences so log it to the panic log.
2720 TF/02 Log LMTP confirmation messages in the same way as SMTP,
2721 controlled using the smtp_confirmation log selector.
2723 TF/03 Include the error message when we fail to unlink a spool file.
2725 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
2726 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
2727 for maintaining out-of-tree patches for some time.
2729 PP/01 Bugzilla 139: Documentation and portability issues.
2730 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
2731 Handle per-OS dynamic-module compilation flags.
2733 PP/02 Let /dev/null have normal permissions.
2734 The 4.73 fixes were a little too stringent and complained about the
2735 permissions on /dev/null. Exempt it from some checks.
2736 Reported by Andreas M. Kirchwitz.
2738 PP/03 Report version information for many libraries, including
2739 Exim version information for dynamically loaded libraries. Created
2740 version.h, now support a version extension string for distributors
2741 who patch heavily. Dynamic module ABI change.
2743 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
2744 privilege escalation vulnerability whereby the Exim run-time user
2745 can cause root to append content of the attacker's choosing to
2748 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
2751 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
2752 If dropping privileges for untrusted macros, we disabled normal logging
2753 on the basis that it would fail; for the Exim run-time user, this is not
2754 the case, and it resulted in successful deliveries going unlogged.
2755 Fixed. Reported by Andreas Metzler.
2761 PP/01 Date: & Message-Id: revert to normally being appended to a message,
2762 only prepend for the Resent-* case. Fixes regression introduced in
2763 Exim 4.70 by NM/22 for Bugzilla 607.
2765 PP/02 Include check_rfc2047_length in configure.default because we're seeing
2766 increasing numbers of administrators be bitten by this.
2768 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
2770 PP/03 Bugzilla 994: added openssl_options main configuration option.
2772 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
2774 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
2776 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
2778 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
2779 without a peer certificate, leading to a segfault because of an
2780 assumption that peers always have certificates. Be a little more
2781 paranoid. Problem reported by Martin Tscholak.
2783 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
2784 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
2785 NB: ClamAV planning to remove STREAM in "middle of 2010".
2786 CL also introduces -bmalware, various -d+acl logging additions and
2787 more caution in buffer sizes.
2789 PP/09 Implemented reverse_ip expansion operator.
2791 PP/10 Bugzilla 937: provide a "debug" ACL control.
2793 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
2795 PP/12 Bugzilla 973: Implement --version.
2797 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2799 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2801 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2803 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2804 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2805 ignore trailing whitespace.
2807 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2809 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2812 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2813 Notification from Dr Andrew Aitchison.
2815 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2816 ExtendedDetectionInfo response format.
2817 Notification from John Horne.
2819 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2822 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2823 XSL and documented dependency on system catalogs, with examples of how
2826 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2829 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2830 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2831 configuration file which is writeable by the Exim user or group.
2833 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2834 of configuration files to cover files specified with the -C option if
2835 they are going to be used with root privileges, not just the default
2838 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2839 option (effectively making it always true).
2841 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2842 files to be used while preserving root privileges.
2844 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2845 that rogue child processes cannot use them.
2847 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2848 run-time user, instead of root.
2850 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2851 Exim run-time user without dropping privileges.
2853 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2854 result string, instead of calling string_vformat() twice with the same
2857 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2858 for other users. Others should always drop root privileges if they use
2859 -C on the command line, even for a whitelisted configure file.
2861 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2863 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2869 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2870 $data_path, and $header_path variables; fixed documentation bugs and
2873 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2874 exipick to access non-standard spools, including the "frozen" queue
2877 NM/01 Bugzilla 965: Support mysql stored procedures.
2878 Patch from Alain Williams
2880 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2882 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2883 Patch from Andreas Metzler
2885 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2886 Patch from Kirill Miazine
2888 NM/05 Bugzilla 671: Added umask to procmail example.
2890 JJ/03 installed exipick 20100323.0, fixing doc bug
2892 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2893 directory. Notification and patch from Dan Rosenberg.
2895 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2897 TK/02 Improve log output when DKIM signing operation fails.
2899 MH/01 Treat the transport option dkim_domain as a colon separated
2900 list, not as a single string, and sign the message with each element,
2901 omitting multiple occurences of the same signer.
2903 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2904 Bugzilla 985, 986. Patch by Simon Arlott
2906 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2907 Patch by Simon Arlott
2909 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2910 MBX locking. Notification from Dan Rosenberg.
2916 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2918 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2920 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2922 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2924 NM/04 Bugzilla 915: Flag broken perl installation during build.
2930 TK/01 Added patch by Johannes Berg that expands the main option
2931 "spamd_address" if it starts with a dollar sign.
2933 TK/02 Write list of recipients to X-Envelope-Sender header when building
2934 the mbox-format spool file for content scanning (suggested by Jakob
2937 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2938 (http://www.dcc-servers.net/) support via dccifd. Activated by
2939 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2941 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2942 by Mark Daniel Reidel <mr@df.eu>.
2944 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2945 When building exim an external PCRE library is now needed -
2946 PCRE is a system library on the majority of modern systems.
2947 See entry on PCRE_LIBS in EDITME file.
2949 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2950 conversation. Added nologin parameter to request.
2951 Patch contributed by Kirill Miazine.
2953 TF/01 Do not log submission mode rewrites if they do not change the address.
2955 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2957 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2958 log files in place. Contributed by Roberto Lima.
2960 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2962 TF/03 Bugzilla 615: When checking the local_parts router precondition
2963 after a local_part_suffix or local_part_prefix option, Exim now
2964 does not use the address's named list lookup cache, since this
2965 contains cached lookups for the whole local part.
2967 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2968 Robert Millan. Documentation is in experimental-spec.txt.
2970 TF/04 Bugzilla 668: Fix parallel build (make -j).
2972 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2974 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2975 Patch provided by Jan Srzednicki.
2977 TF/05 Leading white space used to be stripped from $spam_report which
2978 wrecked the formatting. Now it is preserved.
2980 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2981 that they are available at delivery time.
2983 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2985 TF/08 TLS error reporting now respects the incoming_interface and
2986 incoming_port log selectors.
2988 TF/09 Produce a more useful error message if an SMTP transport's hosts
2989 setting expands to an empty string.
2991 NM/06 Bugzilla 744: EXPN did not work under TLS.
2992 Patch provided by Phil Pennock.
2994 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2995 Patch provided by Richard Godbee.
2997 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2998 acl_smtp_notquit, added index entry.
3000 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
3001 Patch provided by Eugene Bujak.
3003 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
3004 accept(). Patch provided by Maxim Dounin.
3006 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
3007 Patch provided by Phil Pennock.
3009 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
3011 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
3012 Patch provided by Brad "anomie" Jorsch.
3014 NM/14 Bugzilla 622: Added timeout setting to transport filter.
3015 Patch provided by Dean Brooks.
3017 TK/05 Add native DKIM support (does not depend on external libraries).
3019 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
3020 Patch provided by Graeme Fowler.
3022 NM/16 Bugzilla 851: Documentation example syntax fix.
3024 NM/17 Changed NOTICE file to remove references to embedded PCRE.
3026 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
3029 NM/19 Bugzilla 745: TLS version reporting.
3030 Patch provided by Phil Pennock.
3032 NM/20 Bugzilla 167: bool: condition support.
3033 Patch provided by Phil Pennock.
3035 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
3036 clients. Patch provided by Phil Pennock.
3038 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
3039 Patch provided by Brad "anomie" Jorsch.
3041 NM/23 Bugzilla 687: Fix misparses in eximstats.
3042 Patch provided by Heiko Schlittermann.
3044 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
3045 Patch provided by Heiko Schlittermann.
3047 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
3048 plus update to original patch.
3050 NM/26 Bugzilla 799: Documentation correction for ratelimit.
3052 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
3053 Patch provided by David Brownlee.
3055 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
3057 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
3059 NM/30 Bugzilla 888: TLS documentation bugfixes.
3061 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
3063 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
3064 Unlike the original bugzilla I have changed all shell scripts in src tree.
3066 NM/33 Bugzilla 898: Transport filter timeout fix.
3067 Patch by Todd Rinaldo.
3069 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
3070 Patch by Serge Demonchaux.
3072 NM/35 Bugzilla 39: Base64 decode bug fixes.
3073 Patch by Jakob Hirsch.
3075 NM/36 Bugzilla 909: Correct connect() call in dcc code.
3077 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
3079 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
3081 NM/39 Bugzilla 911: Fixed MakeLinks build script.
3087 TK/01 Add preliminary DKIM support. Currently requires a forked version of
3088 ALT-N's libdkim that I have put here:
3089 http://duncanthrax.net/exim-experimental/
3091 Note to Michael Haardt: I had to rename some vars in sieve.c. They
3092 were called 'true' and it seems that C99 defines that as a reserved
3093 keyword to be used with 'bool' variable types. That means you could
3094 not include C99-style headers which use bools without triggering
3095 build errors in sieve.c.
3097 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
3098 as mailq or other aliases. Changed the --help handling significantly
3099 to do whats expected. exim_usage() emits usage/help information.
3101 SC/01 Added the -bylocaldomain option to eximstats.
3103 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
3105 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
3107 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
3113 PH/01 Another patch from the Sieve maintainer.
3115 PH/02 When an IPv6 address is converted to a string for single-key lookup
3116 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
3117 dots are used instead of colons so that keys in lsearch files need not
3118 contain colons. This was done some time before quoting was made available
3119 in lsearch files. However, iplsearch files do require colons in IPv6 keys
3120 (notated using the quote facility) so as to distinguish them from IPv4
3121 keys. This meant that lookups for IP addresses in host lists did not work
3122 for iplsearch lookups.
3124 This has been fixed by arranging for IPv6 addresses to be expressed with
3125 colons if the lookup type is iplsearch. This is not incompatible, because
3126 previously such lookups could never work.
3128 The situation is now rather anomalous, since one *can* have colons in
3129 ordinary lsearch keys. However, making the change in all cases is
3130 incompatible and would probably break a number of configurations.
3132 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
3135 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
3136 conversion specification without a maximum field width, thereby enabling
3137 a rogue spamd server to cause a buffer overflow. While nobody in their
3138 right mind would setup Exim to query an untrusted spamd server, an
3139 attacker that gains access to a server running spamd could potentially
3140 exploit this vulnerability to run arbitrary code as the Exim user.
3142 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
3143 $primary_hostname instead of what libspf2 thinks the hosts name is.
3145 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
3146 a directory entry by the name of the lookup key. Previously, if a
3147 symlink pointed to a non-existing file or a file in a directory that
3148 Exim lacked permissions to read, a lookup for a key matching that
3149 symlink would fail. Now it is enough that a matching directory entry
3150 exists, symlink or not. (Bugzilla 503.)
3152 PH/03 The body_linecount and body_zerocount variables are now exported in the
3155 PH/04 Added the $dnslist_matched variable.
3157 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
3158 This means they are set thereafter only if the connection becomes
3161 PH/06 Added the client_condition to authenticators so that some can be skipped
3162 by clients under certain conditions.
3164 PH/07 The error message for a badly-placed control=no_multiline_responses left
3165 "_responses" off the end of the name.
3167 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
3169 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
3170 (without spaces) instead of just copying the configuration text.
3172 PH/10 Added the /noupdate option to the ratelimit ACL condition.
3174 PH/11 Added $max_received_linelength.
3176 PH/12 Added +ignore_defer and +include_defer to host lists.
3178 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
3179 way in which PCRE > 7.0 is built.
3181 PH/14 Implemented queue_only_load_latch.
3183 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
3184 MAIL command. The effect was to mangle the value on 64-bit systems.
3186 PH/16 Another patch from the Sieve maintainer.
3188 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
3190 PH/18 If a system quota error occurred while trying to create the file for
3191 a maildir delivery, the message "Mailbox is full" was not appended to the
3192 bounce if the delivery eventually timed out. Change 4.67/27 below applied
3193 only to a quota excession during the actual writing of the file.
3195 PH/19 It seems that peer DN values may contain newlines (and other non-printing
3196 characters?) which causes problems in log lines. The DN values are now
3197 passed through string_printing() before being added to log lines.
3199 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
3200 and InterBase are left for another time.)
3202 PH/21 Added message_body_newlines option.
3204 PH/22 Guard against possible overflow in moan_check_errorcopy().
3206 PH/23 POSIX allows open() to be a macro; guard against that.
3208 PH/24 If the recipient of an error message contained an @ in the local part
3209 (suitably quoted, of course), incorrect values were put in $domain and
3210 $local_part during the evaluation of errors_copy.
3216 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
3217 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
3220 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
3221 issue a MAIL command.
3223 PH/02 In an ACL statement such as
3225 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
3227 if a client was not listed at all, or was listed with a value other than
3228 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
3229 the condition was not true (as it should be), so access was not denied.
3230 The bug was that the ! inversion was incorrectly passed on to the second
3231 item. This has been fixed.
3233 PH/03 Added additional dnslists conditions == and =& which are different from
3234 = and & when the dns lookup returns more than one IP address.
3236 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
3237 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
3239 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
3240 FSYNC, which compiles an option called disable_fsync that allows for
3241 bypassing fsync(). The documentation is heavily laced with warnings.
3243 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
3245 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
3246 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
3247 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
3248 including adding "make clean"; (3) Added -fPIC when compiling the test
3249 dynamically loaded module, to get rid of a warning.
3251 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
3252 message fails, move_frozen_messages = true and ignore_bounce_errors_after
3253 = 0s. The bug is otherwise harmless.
3255 PH/07 There was a bug in the dovecot authenticator such that the value of
3256 $auth1 could be overwritten, and so not correctly preserved, after a
3257 successful authentication. This usually meant that the value preserved by
3258 the server_setid option was incorrect.
3260 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
3262 PH/09 Installed PCRE release 7.0.
3264 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
3265 run for batched SMTP input. It is now run at the start of every message
3266 in the batch. While fixing this I discovered that the process information
3267 (output by running exiwhat) was not always getting set for -bs and -bS
3268 input. This is fixed, and it now also says "batched" for BSMTP.
3270 PH/11 Added control=no_pipelining.
3272 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
3273 patch, slightly modified), and move the expansion of helo_data till after
3274 the connection is made in the smtp transport (so it can use these
3277 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
3279 PH/14 Added log_selector = +pid.
3281 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
3283 PH/16 Add ${if forany and ${if forall.
3285 PH/17 Added dsn_from option to vary the From: line in DSNs.
3287 PH/18 Flush SMTP output before performing a callout, unless control =
3288 no_callout_flush is set.
3290 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
3291 was true (the default) a successful delivery failed to delete the retry
3292 item, thus causing premature timeout of the address. The bug is now
3295 PH/20 Added hosts_avoid_pipelining to the smtp transport.
3297 PH/21 Long custom messages for fakedefer and fakereject are now split up
3298 into multiline responses in the same way that messages for "deny" and
3299 other ACL rejections are.
3301 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
3302 with slight modification.
3304 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
3305 draft, changing the syntax and factoring some duplicate code".
3307 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
3308 for deliveries of the second and subsequent messages over the same SMTP
3311 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
3312 ${reduce, with only minor "tidies".
3314 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
3316 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
3317 expansion side effects.
3319 PH/27 When a message times out after an over-quota error from an Exim-imposed
3320 quota, the bounce message says "mailbox is full". This message was not
3321 being given when it was a system quota that was exceeded. It now should
3324 MH/03 Made $recipients available in local_scan(). local_scan() already has
3325 better access to the recipient list through recipients_list[], but
3326 $recipients can be useful in postmaster-provided expansion strings.
3328 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
3329 in the case of a MAIL command with additional options following the
3330 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
3331 were accidentally chopped off.
3333 PH/29 SMTP synchronization checks are implemented when a command is read -
3334 there is a check that no more input is waiting when there shouldn't be
3335 any. However, for some commands, a delay in an ACL can mean that it is
3336 some time before the response is written. In this time, more input might
3337 arrive, invalidly. So now there are extra checks after an ACL has run for
3338 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
3339 pipelining has not been advertised.
3341 PH/30 MH's patch to allow iscntrl() characters to be list separators.
3343 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
3344 returned in the SMTP response when smtp_return_error_details was false.
3345 This has been fixed.
3347 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
3348 instead of the C I/O that was originally supplied, because problems were
3349 reported on Solaris.
3351 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
3352 Exim which did not show up earlier: it was assuming that a call to
3353 SSL_CTX_set_info_callback() might give an error value. In fact, there is
3354 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
3355 was a macro that became an assignment, so it seemed to work. This has
3356 changed to a proper function call with a void return, hence the compile
3357 error. Exim's code has been fixed.
3359 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
3362 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
3364 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
3366 PH/37 If a message is not accepted after it has had an id assigned (e.g.
3367 because it turns out to be too big or there is a timeout) there is no
3368 "Completed" line in the log. When some messages of this type were
3369 selected by exigrep, they were listed as "not completed". Others were
3370 picked up by some special patterns. I have improved the selection
3371 criteria to be more general.
3373 PH/38 The host_find_failed option in the manualroute router can now be set
3374 to "ignore", to completely ignore a host whose IP address cannot be
3375 found. If all hosts are ignored, the behaviour is controlled by the new
3376 host_all_ignored option.
3378 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
3379 homing or because of multiple MX records with /mx) generated more than
3380 one IP address, and the following item turned out to be the local host,
3381 all the secondary addresses of the first item were incorrectly removed
3382 from the list, along with the local host and any following hosts (which
3383 is what is supposed to happen).
3385 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
3386 whoever called Exim into the -H file. In the case of the daemon it was
3387 behaving confusingly. When first started, it used values for whoever
3388 started the daemon, but after a SIGHUP it used the Exim user (because it
3389 calls itself on a restart). I have changed the code so that it now always
3392 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
3393 message are rejected with the same error (e.g. no authentication or bad
3394 sender address), and a DATA command is nevertheless sent (as can happen
3395 with PIPELINING or a stupid MUA), the error message that was given to the
3396 RCPT commands is included in the rejection of the DATA command. This is
3397 intended to be helpful for MUAs that show only the final error to their
3400 PH/42 Another patch from the Sieve maintainer.
3402 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
3403 Eximstats - Fixed some broken HTML links and added missing column headers
3405 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
3406 columns for Rejects, Temp Rejects, Ham, and Spam rows.
3408 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
3410 PH/43 Yet another patch from the Sieve maintainer.
3412 PH/44 I found a way to check for a TCP/IP connection going away before sending
3413 the response to the final '.' that terminates a message, but only in the
3414 case where the client has not sent further data following the '.'
3415 (unfortunately, this is allowed). However, in many cases there won't be
3416 any further data because there won't be any more messages to send. A call
3417 to select() can be used: if it shows that the input is "ready", there is
3418 either input waiting, or the socket has been closed. An attempt to read
3419 the next input character can distinguish the two cases. Previously, Exim
3420 would have sent an OK response which the client would never have see.
3421 This could lead to message repetition. This fix should cure that, at
3422 least in a lot of common cases.
3424 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
3425 advertised in response to EHLO.
3431 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
3432 fixed by 4.65/MH/01 (is this a record?) are fixed:
3434 (i) An empty string was always treated as zero by the numeric comparison
3435 operators. This behaviour has been restored.
3437 (ii) It is documented that the numeric comparison operators always treat
3438 their arguments as decimal numbers. This was broken in that numbers
3439 starting with 0 were being interpreted as octal.
3441 While fixing these problems I realized that there was another issue that
3442 hadn't been noticed. Values of message_size_limit (both the global option
3443 and the transport option) were treated as octal if they started with 0.
3444 The documentation was vague. These values are now always treated as
3445 decimal, and I will make that clear in the documentation.
3451 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
3452 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
3455 MH/01 Don't check that the operands of numeric comparison operators are
3456 integers when their expansion is in "skipping" mode (fixes bug
3457 introduced by 4.64-PH/07).
3459 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
3460 child addresses, Exim now panics and dies. Previously, because the count
3461 is held in a short int, deliveries were likely to be lost. As such a
3462 large number of recipients for a single message is ridiculous
3463 (performance will be very, very poor), I have chosen to impose a limit
3464 rather than extend the field.
3470 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
3471 leftover -K file (the existence of which was triggered by #402).
3472 While we were at it, introduced process PID as part of the -K
3473 filename. This should rule out race conditions when creating
3476 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
3477 processing considerably. Previous code took too long for large mails,
3478 triggering a timeout which in turn triggers #401.
3480 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
3481 in the DK code in transports.c. sendfile() is not really portable,
3482 hence the _LINUX specificness.
3484 TF/01 In the add_headers option to the mail command in an Exim filter,
3485 there was a bug that Exim would claim a syntax error in any
3486 header after the first one which had an odd number of characters
3489 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
3490 callout verification, Exim cached a "reject" for the entire domain. This
3491 is correct for most verifications, but it is not correct for a recipient
3492 verification with use_sender or use_postmaster set, because in that case
3493 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
3494 case of MAIL FROM:<> rejection from other early rejections (e.g.
3495 rejection of HELO). When verifying a recipient using a non-null MAIL
3496 address, the cache is ignored if it shows MAIL FROM:<> rejection.
3497 Whatever the result of the callout, the value of the domain cache is
3498 left unchanged (for any other kind of callout, getting as far as trying
3499 RCPT means that the domain itself is ok).
3501 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
3504 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
3505 manifest itself as EPIPE rather than ECONNECT. When tidying away a
3506 session, the daemon ignores ECONNECT errors and logs others; it now
3507 ignores EPIPE as well.
3509 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
3510 (quoted-printable decoding).
3512 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
3513 later the small subsequent patch to fix an introduced bug.
3515 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
3517 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
3519 PH/08 An error is now given if message_size_limit is specified negative.
3521 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
3522 to be given (somewhat) arbitrary names.
3524 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
3527 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
3528 miscellaneous code fixes
3530 PH/10 Added the log_reject_target ACL modifier to specify where to log
3533 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
3534 hostname. This is wrong, because it relates to the incoming message (and
3535 probably the interface on which it is arriving) and not to the outgoing
3536 callout (which could be using a different interface). This has been
3537 changed to use the value of the helo_data option from the smtp transport
3538 instead - this is what is used when a message is actually being sent. If
3539 there is no remote transport (possible with a router that sets up host
3540 addresses), $smtp_active_hostname is used.
3542 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
3543 tweaks were necessary in order to get it to work (see also 21 below):
3544 (a) The code assumed that strncpy() returns a negative number on buffer
3545 overflow, which isn't the case. Replaced with Exim's string_format()
3547 (b) There were several signed/unsigned issues. I just did the minimum
3548 hacking in of casts. There is scope for a larger refactoring.
3549 (c) The code used strcasecmp() which is not a standard C function.
3550 Replaced with Exim's strcmpic() function.
3551 (d) The code set only $1; it now sets $auth1 as well.
3552 (e) A simple test gave the error "authentication client didn't specify
3553 service in request". It would seem that Dovecot has changed its
3554 interface. Fortunately there's a specification; I followed it and
3555 changed what the client sends and it appears to be working now.
3557 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
3560 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
3561 address A is aliased to B and C, where B exists and C does not. Without
3562 -v the output is "A verified" because verification stops after a
3563 successful redirection if more than one address is generated. However,
3564 with -v the child addresses are also verified. Exim was outputting "A
3565 failed to verify" and then showing the successful verification for C,
3566 with its parentage. It now outputs "B failed to verify", showing B's
3567 parentage before showing the successful verification of C.
3569 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
3570 look up a TXT record in a specific list after matching in a combined
3573 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
3574 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
3575 they consult the DNS. I had assumed they would set it the way they
3576 wanted; and indeed my experiments on Linux seem to show that in some
3577 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
3578 To be on the safe side, however, I have now made the interface to
3579 host_find_byname() similar to host_find_bydns(), with an argument
3580 containing the DNS resolver options. The host_find_byname() function now
3581 sets these options at its start, just as host_find_bydns() does. The smtp
3582 transport options dns_qualify_single and dns_search_parents are passed to
3583 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
3584 of host_find_byname() use the default settings of RES_DEFNAMES
3585 (qualify_single) but not RES_DNSRCH (search_parents).
3587 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
3588 spool_read_header() do less string testing, by means of a preliminary
3589 switch on the second character of optional "-foo" lines. (This is
3590 overdue, caused by the large number of possibilities that now exist.
3591 Originally there were few.) While I was there, I also converted the
3592 str(n)cmp tests so they don't re-test the leading "-" and the first
3593 character, in the hope this might squeeze out yet more improvement.
3595 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
3596 flag allowing group syntax was set by the header_syntax check but not
3597 turned off, possible causing trouble later; (2) The flag was not being
3598 set at all for the header_verify test, causing "group"-style headers to
3599 be rejected. I have now set it in this case, and also caused header_
3600 verify to ignore an empty address taken from a group. While doing this, I
3601 came across some other cases where the code for allowing group syntax
3602 while scanning a header line wasn't quite right (mostly, not resetting
3603 the flag correctly in the right place). These bugs could have caused
3604 trouble for malformed header lines. I hope it is now all correct.
3606 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
3607 with the "reply" argument non-NULL. The code, however (which originally
3608 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
3609 but it didn't always do it. This confused somebody who was copying the
3610 code for some other use. I have removed all the tests.
3612 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
3613 feature that was used to support insecure browsers during the U.S. crypto
3614 embargo. It requires special client support, and Exim is probably the
3615 only MTA that supported it -- and would never use it because real RSA is
3616 always available. This code has been removed, because it had the bad
3617 effect of slowing Exim down by computing (never used) parameters for the
3618 RSA_EXPORT functionality.
3620 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
3621 authenticator to fail if there's a tab character in the incoming data
3622 (there should never be unless someone is messing about, as it's supposed
3623 to be base64-encoded). Also added, on Timo's advice, the "secured" option
3624 if the connection is using TLS or if the remote IP is the same as the
3625 local IP, and the "valid-client-cert option" if a client certificate has
3628 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
3629 authenticators. This can be used for authorization after authentication
3630 succeeds. (In the case of plaintext, it servers for both authentication
3633 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
3634 if any retry times were supplied.
3636 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
3637 connection, where there is no client IP address to check. In this
3638 situation, the verify now always succeeds.
3640 PH/25 Applied John Jetmore's -Mset patch.
3642 PH/26 Added -bem to be like -Mset, but loading a message from a file.
3644 PH/27 In a string expansion for a processed (not raw) header when multiple
3645 headers of the same name were present, leading whitespace was being
3646 removed from all of them, but trailing whitespace was being removed only
3647 from the last one. Now trailing whitespace is removed from each header
3648 before concatenation. Completely empty headers in a concatenation (as
3649 before) are ignored.
3651 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
3652 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
3654 PH/29 [Removed. This was a change that I later backed out, and forgot to
3655 correct the ChangeLog entry (that I had efficiently created) before
3656 committing the later change.]
3658 PH/30 Exim was sometimes attempting to deliver messages that had suffered
3659 address errors (4xx response to RCPT) over the same connection as other
3660 messages routed to the same hosts. Such deliveries are always "forced",
3661 so retry times are not inspected. This resulted in far too many retries
3662 for the affected addresses. The effect occurred only when there were more
3663 hosts than the hosts_max_try setting in the smtp transport when it had
3664 the 4xx errors. Those hosts that it had tried were not added to the list
3665 of hosts for which the message was waiting, so if all were tried, there
3666 was no problem. Two fixes have been applied:
3668 (i) If there are any address or message errors in an SMTP delivery, none
3669 of the hosts (tried or untried) are now added to the list of hosts
3670 for which the message is waiting, so the message should not be a
3671 candidate for sending over the same connection that was used for a
3672 successful delivery of some other message. This seems entirely
3673 reasonable: after all the message is NOT "waiting for some host".
3674 This is so "obvious" that I'm not sure why it wasn't done
3675 previously. Hope I haven't missed anything, but it can't do any
3676 harm, as the worst effect is to miss an optimization.
3678 (ii) If, despite (i), such a delivery is accidentally attempted, the
3679 routing retry time is respected, so at least it doesn't keep
3680 hammering the server.
3682 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
3683 in ${readsocket because some servers need this prod.
3685 PH/32 Added some extra debug output when updating a wait-xxx database.
3687 PH/33 The hint "could be header name not terminated by colon", which has been
3688 given for certain expansion errors for a long time, was not being given
3689 for the ${if def:h_colon_omitted{... case.
3691 PH/34 The spec says: "With one important exception, whenever a domain list is
3692 being scanned, $domain contains the subject domain." There was at least
3693 one case where this was not true.
3695 PH/35 The error "getsockname() failed: connection reset by peer" was being
3696 written to the panic log as well as the main log, but it isn't really
3697 panic-worthy as it just means the connection died rather early on. I have
3698 removed the panic log writing for the ECONNRESET error when getsockname()
3701 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
3702 runs only) independently of the message's sender address. This meant
3703 that, if the 4xx error was in fact related to the sender, a different
3704 message to the same recipient with a different sender could confuse
3705 things. In particular, this can happen when sending to a greylisting
3706 server, but other circumstances could also provoke similar problems.
3707 I have changed the default so that the retry time for these errors is now
3708 based a combination of the sender and recipient addresses. This change
3709 can be overridden by setting address_retry_include_sender=false in the
3712 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
3713 remote server are returned as part of bounce messages. This was not
3714 happening for LMTP over a pipe (the lmtp transport), but now it is the
3715 same for both kinds of LMTP.
3717 PH/38 Despite being documented as not happening, Exim was rewriting addresses
3718 in header lines that were in fact CNAMEs. This is no longer the case.
3720 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
3721 and queue runs started by the daemon processed all messages. This has
3722 been fixed so that -R and -S can now usefully be given with -q<time>.
3724 PH/40 Import PCRE release 6.7 (fixes some bugs).
3726 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
3728 PH/42 Give an error if -q is specified more than once.
3730 PH/43 Renamed the variables $interface_address and $interface_port as
3731 $received_ip_address and $received_port, to make it clear that these
3732 values apply to message reception, and not to the outgoing interface when
3733 a message is delivered. (The old names remain recognized, of course.)
3735 PH/44 There was no timeout on the connect() call when using a Unix domain
3736 socket in the ${readsocket expansion. There now is.
3738 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
3739 be meaningful with "accept".
3741 SC/01 Eximstats V1.43
3742 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
3744 SC/02 Eximstats V1.44
3745 Use a glob alias rather than an array ref in the generated
3746 parser. This improves both readability and performance.
3748 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
3749 Collect SpamAssassin and rejection statistics.
3750 Don't display local sender or destination tables unless
3751 there is data to show.
3752 Added average volumes into the top table text output.
3754 SC/04 Eximstats V1.46
3755 Collect data on the number of addresses (recipients)
3756 as well as the number of messages.
3758 SC/05 Eximstats V1.47
3759 Added 'Message too big' to the list of mail rejection
3760 reasons (thanks to Marco Gaiarin).
3762 SC/06 Eximstats V1.48
3763 Mainlog lines which have GMT offsets and are too short to
3764 have a flag are now skipped.
3766 SC/07 Eximstats V1.49 (Alain Williams)
3767 Added the -emptyok flag.
3769 SC/08 Eximstats V1.50
3770 Fixes for obtaining the IP address from reject messages.
3772 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
3773 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
3774 whitespace changes from 4.64-PH/27
3776 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
3779 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
3780 are found, allow negative numbers in numeric criteria)
3782 JJ/06 exipick.20061117.2, added new $message_body_missing variable
3784 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
3785 to match changes made in 4.64-PH/43
3787 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
3789 PH/47 Put in an explicit test for a DNS lookup of an address record where the
3790 "domain" is actually an IP address, and force a failure. This locks out
3791 those revolvers/nameservers that support "A-for-A" lookups, in
3792 contravention of the specifications.
3794 PH/48 When a host name was looked up from an IP address, and the subsequent
3795 forward lookup of the name timed out, the host name was left in
3796 $sender_host_name, contrary to the specification.
3798 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3799 restricted to single-key lookups, Exim was not diagnosing an error if
3800 * or *@ was used with a query-style lookup.
3802 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3804 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3805 long ago, but noone interested enough thought of it. Let's just say that
3806 the "1.1" means that there are some new functions that weren't there at
3807 some point in the past.
3809 PH/51 Error processing for expansion failure of helo_data from an smtp
3810 transport during callout processing was broken.
3812 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3813 tested/used via the -bh/-bhc/-bs options.
3815 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3816 bug, fixed in subsequent PCRE releases).
3818 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3819 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3821 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3827 SC/01 Use a glob alias rather than an array ref in eximstats generated
3828 parser. This improves both readability and performance.
3830 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3831 Don't display local sender or destination tables in eximstats unless
3832 there is data to show.
3833 Added average volumes into the eximstats top table text output.
3835 SC/03 Collect data on the number of addresses (recipients) as well
3836 as the number of messages in eximstats.
3838 TF/01 Correct an error in the documentation for the redirect router. Exim
3839 does (usually) call initgroups() when daemonizing.
3841 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3842 with consistent privilege compared to when running as a daemon.
3844 TF/03 Note in the spec that $authenticated_id is not set for local
3845 submissions from trusted users.
3847 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3848 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3850 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3851 by adding some example configuration directives to the default
3852 configuration file. A little bit of work is required to uncomment the
3853 directives and define how usernames and passwords are checked, but
3854 there is now a framework to start from.
3856 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3857 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3858 without this. I don't know how relevant this is to other LDAP libraries.
3860 PH/02 Add the verb name to the "unknown ACL verb" error.
3862 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3864 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3866 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3867 directory not expanded when it should be if an expanded home directory
3868 was set for the address (which is overridden by the transport).
3870 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3873 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3874 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3875 because it is too late at that time, and has no effect.
3877 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3878 security issue with \' (bugzilla #107). I could not use the
3879 PQescapeStringConn() function, because it needs a PGconn value as one of
3882 PH/08 When testing addresses using -bt, indicate those final addresses that
3883 are duplicates that would not cause an additional delivery. At least one
3884 person was confused, thinking that -bt output corresponded to deliveries.
3885 (Suppressing duplicates isn't a good idea as you lose the information
3886 about possibly different redirections that led to the duplicates.)
3888 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3889 systems where poll() doesn't work, in particular OS X.
3891 PH/10 Added more information to debugging output for retry time not reached.
3893 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3894 operations in malware.c.
3896 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3899 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3900 syslog_duplication set false, log lines that would normally be written
3901 both the the main log and to the reject log were not written to syslog at
3904 PH/14 In the default configuration, change the use of "message" in ACL warn
3905 statements to "add_header".
3907 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3908 not followed by a command (e.g. "seen endif").
3910 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3911 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3914 PH/17 Added extra conditions to the default value of delay_warning_condition
3918 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3919 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3920 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3923 The Auto-Submitted: and various List- headers are standardised, whereas I
3924 don't think Precedence: ever was.
3926 PH/18 Refactored debugging code in route_finduser() to show more information,
3927 in particular, the error code if getpwnam() issues one.
3929 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3930 This is apparently needed in addition to the PH/07 change above to avoid
3931 any possible encoding problems.
3933 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3934 but not after initializing Perl.
3936 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3937 output them only if debugging. By default they are written stderr,
3938 apparently, which is not desirable.
3940 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3943 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3946 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3948 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3949 authenticated or an ident call has been made. Suppress the default
3950 values for $authenticated_id and $authenticated_sender (but permit -oMai
3951 and -oMas) when testing with -bh.
3953 PH/24 Re-jigged the order of the tests in the default configuration so that the
3954 tests for valid domains and recipients precede the DNS black list and CSA
3955 tests, on the grounds that those ones are more expensive.
3957 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3958 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3959 This bug exists in every version of Exim that I still have, right back to
3962 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3963 However, an attempt to turn on case-sensitivity in a regex key by
3964 including (?-i) didn't work because the subject string was already
3965 lowercased, and the effects were non-intuitive. It turns out that a
3966 one-line patch can be used to allow (?-i) to work as expected.
3972 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3973 other effects) broke the use of negated acl sub-conditions.
3975 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3978 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3979 "Deny" causes Exim to reject the incoming connection with a 554 error.
3980 Unfortunately, if there is a major crisis, such as a disk failure,
3981 tcp-wrappers gives "deny", whereas what one would like would be some
3982 kind of temporary error. A kludge has been added to help with this.
3983 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3984 554 error is used if errno is still zero or contains ENOENT (which occurs
3985 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3988 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3990 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3991 errors. Otherwise a message that provokes a temporary error (when other
3992 messages do not) can cause a whole host to time out.
3994 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3995 addresses were routed directly to files or pipes from a redirect router.
3996 File deliveries just didn't batch; pipe deliveries might have suffered
3999 PH/06 A failure to get a lock for a hints database would erroneously always say
4000 "Failed to get write lock", even when it was really a read lock.
4002 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
4003 of 0600. This has been changed to use the value of the lockfile_mode
4004 option (which defaults to 0600).
4006 PH/08 Applied small patch from the Sieve maintainer.
4008 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
4009 folder from quota calculations, a direct delivery into this folder messed
4010 up the contents of the maildirsize file. This was because the regex was
4011 used only to exclude .Trash (or whatever) when the size of the mailbox
4012 was calculated. There was no check that a delivery was happening into an
4013 excluded directory. This bug has been fixed by ignoring all quota
4014 processing for deliveries into excluded directories.
4016 PH/10 Added the maildirfolder_create_regex option to appendfile.
4022 PH/01 The code for finding all the local interface addresses on a FreeBSD
4023 system running IPv6 was broken. This may well have applied to all BSD
4024 systems, as well as to others that have similar system calls. The broken
4025 code found IPv4 interfaces correctly, but gave incorrect values for the
4026 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
4027 that it would not match correctly against @[] and not recognize the IPv6
4030 PH/02 The ipliteral router was not recognizing addresses of the form user@
4031 [ipv6:....] because it didn't know about the "ipv6:" prefix.
4033 PH/03 Added disable_ipv6.
4035 PH/04 Changed $reply_address to use the raw form of the headers instead of the
4036 decoded form, because it is most often used to construct To: headers
4037 lines in autoreplies, and the decoded form may well be syntactically
4038 invalid. However, $reply_address has leading white space removed, and all
4039 newlines turned into spaces so that the autoreply transport does not
4042 PH/05 If group was specified without a user on a router, and no group or user
4043 was specified on a transport, the group from the router was ignored.
4045 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
4046 for visible compile-time settings that can be used to change these
4047 numbers, for those that want even more. Backwards compatibility with old
4048 spool files has been maintained. However, going back to a previous Exim
4049 release will lost any variables that are in spool files.
4051 PH/07 Two small changes when running in the test harness: increase delay when
4052 passing a TCP/IP connection to a new process, in case the original
4053 process has to generate a bounce, and remove special handling of
4054 127.0.0.2 (sic), which is no longer necessary.
4056 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
4057 be the same on different OS.
4059 PH/09 Moved a debug statement in filter processing to avoid a race problem when
4062 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
4063 whether --show-vars was specified or not
4065 JJ/02 exipick: Added support for new ACL variable spool format introduced
4068 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
4069 syntactically invalid From: or Reply-to: line, and a filter used this to
4070 generate an autoreply, and therefore failed to obtain an address for the
4071 autoreply, Exim could try to deliver to a non-existent relative file
4072 name, causing unrelated and misleading errors. What now happens is that
4073 it logs this as a hard delivery error, but does not attempt to create a
4076 PH/11 The exinext utility has a -C option for testing purposes, but although
4077 the given file was scanned by exinext itself; it wasn't being passed on
4078 when Exim was called.
4080 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
4081 an end-of-file indication when reading a command response.
4083 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
4084 compiled. In many other places in Exim, IPv6 addresses are always
4085 recognized, so I have changed this. It also means that IPv4 domain
4086 literals of the form [IPV4:n.n.n.n] are now always recognized.
4088 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
4089 used if the router is not running as root, for example, when verifying at
4090 ACL time, or when using -bh. The debugging output from this situation was
4091 non-existent - all you got was a failure to exec. I have made two
4094 (a) Failures to set uid/gid, the current directory, or a process leader
4095 in a subprocess such as that created by queryprogram now generate
4096 suitable debugging output when -d is set.
4098 (b) The queryprogram router detects when it is not running as root,
4099 outputs suitable debugging information if -d is set, and then runs
4100 the subprocess without attempting to change uid/gid.
4102 PH/15 Minor change to Makefile for building test_host (undocumented testing
4105 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
4106 additional section of a DNS packet that returns MX or SRV records.
4107 Instead, it always explicitly searches for A/AAAA records. This avoids
4108 major problems that occur when a DNS server includes only records of one
4109 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
4110 fixed another bug: if SRV records were looked up and the corresponding
4111 address records were *not* found in the additional section, the port
4112 values from the SRV records were lost.
4114 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
4115 using the correct key (the original address) when searching the retry
4116 rules in order to find which one to use for generating the retry hint.
4118 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
4119 adding the default one. Similarly, if it contains a Reply-To: header, the
4120 errors_reply_to option, if set, is not used.
4122 PH/19 When calculating a retry time, Exim used to measure the "time since
4123 failure" by looking at the "first failed" field in the retry record. Now
4124 it does not use this if it is later than than the arrival time of the
4125 message. Instead it uses the arrival time. This makes for better
4126 behaviour in cases where some deliveries succeed, thus re-setting the
4127 "first failed" field. An example is a quota failure for a huge message
4128 when small messages continue to be delivered. Without this change, the
4129 "time since failure" will always be short, possible causing more frequent
4130 delivery attempts for the huge message than are intended.
4131 [Note: This change was subsequently modified - see PH/04 for 4.62.]
4133 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
4134 $1, $2, $3) because the numerical variables can be reset during some
4135 expansion items (e.g. "match"), thereby losing the authentication data.
4137 PH/21 Make -bV show the size of off_t variables so that the test suite can
4138 decide whether to run tests for quotas > 2G.
4140 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
4141 mailbox_size, and mailbox_filecount in the appendfile transport. If a
4142 filecount value is greater than 2G or if a quota value is greater than 2G
4143 on a system where the size of off_t is not greater than 4, a panic error
4146 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
4147 never match. The debug and -bh output now contains an explicit error
4148 message indicating a malformed IPv4 address or mask.
4150 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
4151 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
4152 PH/23 above applies.
4154 PH/25 Do not write to syslog when running in the test harness. The only
4155 occasion when this arises is a failure to open the main or panic logs
4156 (for which there is an explicit test).
4158 PH/26 Added the /no_tell option to "control=freeze".
4160 PH/27 If a host name lookup failed very early in a connection, for example, if
4161 the IP address matched host_lookup and the reverse lookup yielded a name
4162 that did not have a forward lookup, an error message of the form "no IP
4163 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
4164 could be logged. Now it outputs the IP address instead of "NULL".
4166 PH/28 An enabling patch from MH: add new function child_open_exim2() which
4167 allows the sender and the authenticated sender to be set when
4168 submitting a message from within Exim. Since child_open_exim() is
4169 documented for local_scan(), the new function should be too.
4171 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
4172 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
4173 results in an empty string is now treated as unset.
4175 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
4177 PH/31 Added sender_verify_fail logging option.
4179 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
4180 needed by RFC 822 but not by RFC 2822 was commented out. I have now
4181 tidied the source and removed it altogether.
4183 PH/33 When a queue run was abandoned because the load average was too high, a
4184 log line was always written; now it is written only if the queue_run log
4185 selector is set. In addition, the log line for abandonment now contains
4186 information about the queue run such as the pid. This is always present
4187 in "start" and "stop" lines but was omitted from the "abandon" line.
4189 PH/34 Omit spaces between a header name and the colon in the error message that
4190 is given when verify = headers_syntax fails (if there are lots of them,
4191 the message gets confusing).
4193 PH/35 Change the default for dns_check_names_pattern to allow slashes within
4194 names, as there are now some PTR records that contain slashes. This check
4195 is only to protect against broken name servers that fall over on strange
4196 characters, so the fact that it applies to all lookups doesn't matter.
4198 PH/36 Now that the new test suite is complete, we can remove some of the
4199 special code in Exim that was needed for the old test suite. For example,
4200 sorting DNS records because real resolvers return them in an arbitrary
4201 order. The new test suite's fake resolver always returns records in the
4204 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
4205 bounces) except when queue_only is set, to avoid logging races between
4206 the different processes.
4208 PH/38 Panic-die if .include specifies a non-absolute path.
4210 PH/39 A tweak to the "H" retry rule from its user.
4212 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
4213 a label. They prevented compilation on older perls.
4215 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
4216 a warning to be raised on newish perls.
4218 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
4219 on queue. Changes to match documented behaviour of showing count of
4220 messages matching specified criteria.
4222 PH/40 Changed the default ident timeout from 30s to 5s.
4224 PH/41 Added support for the use of login_cap features, on those BSD systems
4225 that have them, for controlling the resources used by pipe deliveries.
4227 PH/42 The content-scanning code uses fopen() to create files in which to put
4228 message data. Previously it was not paying any attention to the mode of
4229 the files. Exim runs with umask(0) because the rest of the code creates
4230 files with open(), and sets the required mode explicitly. Thus, these
4231 files were ending up world-writeable. This was not a big issue, because,
4232 being within the spool directory, they were not world-accessible. I have
4233 created a function called modefopen, which takes an additional mode
4234 argument. It sets umask(777), creates the file, chmods it to the required
4235 mode, then resets the umask. All the relevant calls to fopen() in the
4236 content scanning code have been changed to use this function.
4238 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
4239 to 24 hours. This avoids potential overflow problems when processing G
4240 and H retry rules. I suspect nobody ever tinkers with this value.
4242 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
4244 PH/45 When the plaintext authenticator is running as a client, the server's
4245 challenges are checked to ensure they are valid base64 strings. By
4246 default, the authentication attempt is cancelled if an invalid string is
4247 received. Setting client_ignore_invalid_base64 true ignores these errors.
4248 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
4249 they are received. Thus, the responses can be made to depend on the
4250 challenges. If an invalid string is ignored, an empty string is placed in
4253 PH/46 Messages that are created by the autoreply transport now contains a
4254 References: header, in accordance with RFCs 2822 and 3834.
4256 PH/47 Added authenticated_sender_force to the smtp transport.
4258 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
4260 PH/49 Installed latest patch from the Sieve maintainer.
4262 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
4263 was also set, the appendfile transport was unnecessarily scanning a
4264 directory of message files (e.g. for maildir delivery) to find the count
4265 of files (along with the size), even though it did not need this
4266 information. It now does the scan only if it needs to find either the
4267 size of the count of files.
4269 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
4271 PH/52 Two bugs concerned with error handling when the smtp transport is
4274 (i) Exim was not creating retry information for temporary errors given
4275 for individual recipients after the DATA command when the smtp transport
4276 was used in LMTP mode. This meant that they could be retried too
4277 frequently, and not timed out correctly.
4279 (ii) Exim was setting the flag that allows error details to be returned
4280 for LMTP errors on RCPT commands, but not for LMTP errors for individual
4281 recipients that were returned after the DATA command.
4283 PH/53 This is related to PH/52, but is more general: for any failing address,
4284 when detailed error information was permitted to be returned to the
4285 sender, but the error was temporary, then after the final timeout, only
4286 "retry timeout exceeded" was returned. Now it returns the full error as
4287 well as "retry timeout exceeded".
4289 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
4290 do this, and (what is worse) MTAs that accept it.
4292 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
4293 will now be deprecated.
4295 PH/56 New os.c-cygwin from the Cygwin maintainer.
4297 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
4298 formats (previously only available in exim formats via -bpr, -bpru,
4299 and -bpra. Now also available in native and exiqgrep formats)
4301 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
4302 with very large, slow to parse queues
4304 JJ/08 exipick: added ! as generic prefix to negate any criteria format
4306 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
4308 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
4309 responses to authentication challenges, though it was showing the
4310 challenges; (ii) I've removed the CR characters from the debug output for
4313 PH/58 Allow for the insertion of a newline as well as a space when a string
4314 is turned into more than one encoded-word during RFC 2047 encoding. The
4315 Sieve code now uses this.
4317 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
4318 data_4xx, lost_connection, tls_required.
4320 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
4321 message was being sent as an SMTP response.
4323 PH/61 Add -l and -k options to exicyclog.
4325 PH/62 When verifying, if an address was redirected to one new address, so that
4326 verification continued, and the new address failed or deferred after
4327 having set something in $address_data, the value of $address_data was not
4328 passed back to the ACL. This was different to the case when no
4329 redirection occurred. The value is now passed back in both cases.
4331 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
4332 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
4333 use login_cap.h, so on its own it isn't the distinguishing feature. The
4334 new name refers directly to the setclassresources() function.
4336 PH/65 Added configuration files for NetBSD3.
4338 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
4340 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
4341 is preferred over IPv4.
4343 PH/68 The bounce_return_message and bounce_return_body options were not being
4344 honoured for bounces generated during the reception of non-SMTP messages.
4345 In particular, this applied to messages rejected by the ACL. This bug has
4346 been fixed. However, if bounce_return_message is true and bounce_return_
4347 body is false, the headers that are returned for a non-SMTP message
4348 include only those that have been read before the error was detected.
4349 (In the case of an ACL rejection, they have all been read.)
4351 PH/69 The HTML version of the specification is now built in a directory called
4352 spec_html instead of spec.html, because the latter looks like a path with
4353 a MIME-type, and this confuses some software.
4355 PH/70 Catch two compiler warnings in sieve.c.
4357 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
4358 function verify_get_ident() calls ip_connect() to connect a socket, but
4359 if the "connect()" function timed out, ip_connect() used to close the
4360 socket. However, verify_get_ident() also closes the socket later, and in
4361 between Exim writes to the log, which may get opened at this point. When
4362 the socket was closed in ip_connect(), the log could get the same file
4363 descriptor number as the socket. This naturally causes chaos. The fix is
4364 not to close the socket in ip_connect(); the socket should be closed by
4365 the function that creates it. There was only one place in the code where
4366 this was missing, in the iplookup router, which I don't think anybody now
4367 uses, but I've fixed it anyway.
4369 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
4370 well as to direct DNS lookups. Otherwise the handling of names in host
4371 lists is inconsistent and therefore confusing.
4377 PH/01 Two changes to the default runtime configuration:
4379 (1) Move the checks for relay_from_hosts and authenticated clients from
4380 after to before the (commented out) DNS black list checks.
4382 (2) Add control=submission to the relay_from_hosts and authenticated
4383 clients checks, on the grounds that messages accepted by these
4384 statements are most likely to be submissions.
4386 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
4388 (1) Generate an error if the third argument for the ${prvs expansion is
4391 (2) Treat a missing third argument of ${prvscheck as if it were an empty
4394 (3) Reset the variables that are obtained from the first argument of
4395 ${prvscheck and used in the second argument before leaving the code,
4396 because their memory is reclaimed, so using them afterwards may do
4399 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
4400 one (it's much easier than Tom thought :-).
4402 (5) Because of (4), we can now allow for the use of $prvscheck_result
4403 inside the third argument.
4405 PH/03 For some reason, the default setting of PATH when running a command from
4406 a pipe transport was just "/usr/bin". I have changed it to
4409 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
4410 anything to be listed in the output from -bV.
4412 PH/05 When a filter generated an autoreply, the entire To: header line was
4413 quoted in the delivery log line, like this:
4415 => >A.N.Other <ano@some.domain> <original@ddress> ...
4417 This has been changed so that it extracts the operative address. There
4418 may be more than one such address. If so, they are comma-separated, like
4421 => >ano@some.domain,ona@other.domain <original@ddress> ...
4423 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
4424 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
4425 not being looked up in the rDNS to get a host name, Exim was showing the
4426 IP address twice in Received: lines, even though the IP addresses were
4427 identical. For example:
4429 Received: from [1.2.3.4] (helo=[1.2.3.4])
4431 However, if the real host name was known, it was omitting the HELO data
4432 if it matched the actual IP address. This has been tidied up so that it
4433 doesn't show the same IP address twice.
4435 PH/07 When both +timestamp and +memory debugging was on, the value given by
4436 $tod_xxx expansions could be wrong, because the tod_stamp() function was
4437 called by the debug printing, thereby overwriting the timestamp buffer.
4438 Debugging no longer uses the tod_stamp() function when +timestamp is set.
4440 PH/08 When the original message was included in an autoreply transport, it
4441 always said "this is a copy of the message, including all the headers",
4442 even if body_only or headers_only was set. It now gives an appropriate
4445 PH/09 Applied a patch from the Sieve maintainer which:
4447 o fixes some comments
4448 o adds the (disabled) notify extension core
4449 o adds some debug output for the result of if/elsif tests
4450 o points to the current vacation draft in the documentation
4451 and documents the missing references header update
4455 o fixes a bug in processing the envelope test (when testing
4456 multiple envelope elements, the last element determined the
4459 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
4460 Electronic Mail") by including:
4462 Auto-submitted: auto-generated
4464 in the messages that it generates (bounce messages and others, such as
4465 warnings). In the case of bounce messages for non-SMTP messages, there was
4466 also a typo: it was using "Auto_submitted" (underscore instead of
4467 hyphen). Since every message generated by Exim is necessarily in response
4468 to another message, thes have all been changed to:
4470 Auto-Submitted: auto-replied
4472 in accordance with these statements in the RFC:
4474 The auto-replied keyword:
4476 - SHOULD be used on messages sent in direct response to another
4477 message by an automatic process,
4479 - MUST NOT be used on manually-generated messages,
4481 - MAY be used on Delivery Status Notifications (DSNs) and Message
4482 Disposition Notifications (MDNs),
4484 - MUST NOT be used on messages generated by automatic or periodic
4485 processes, except for messages which are automatic responses to
4488 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
4489 to the default Received: header definition.
4491 PH/12 Added log selector acl_warn_skipped (default on).
4493 PH/13 After a successful wildlsearch lookup, discard the values of numeric
4494 variables because (a) they are in the wrong storage pool and (b) even if
4495 they were copied, it wouldn't work properly because of the caching.
4497 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
4498 checking when decoding. Apparently there are clients that generate
4499 overlong encoded strings. Why am I not surprised?
4501 PH/15 If the first argument of "${if match_address" was not empty, but did not
4502 contain an "@" character, Exim crashed. Now it writes a panic log message
4503 and treats the condition as false.
4505 PH/16 In autoreply, treat an empty string for "once" the same as unset.
4507 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
4508 extension "envelope-auth". The code is finished and in agreement with
4509 other implementations, but there is no documentation so far and in fact,
4510 nobody wrote the draft yet. This extension is currently #undef'ed, thus
4511 not changing the active code.
4513 Print executed "if" and "elsif" statements when debugging is used. This
4514 helps a great deal to understand what a filter does.
4516 Document more things not specified clearly in RFC3028. I had all this
4517 sorted out, when out of a sudden new issues came to my mind. Oops."
4519 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
4522 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
4523 canonical form (as documented). However, after a host name lookup from
4524 the IP address, check_host() was doing a simple string comparison with
4525 addresses acquired from the DNS when checking that the found name did
4526 have the original IP as one of its addresses. Since any found IPv6
4527 addresses are likely to be in abbreviated form, the comparison could
4528 fail. Luckily, there already exists a function for doing the comparison
4529 by converting both addresses to binary, so now that is used instead of
4530 the text comparison.
4532 PH/20 There was another similar case to PH/19, when a complete host name was
4533 given in a host list; looking up its IP address could give an abbreviated
4534 form, whereas the current host's name might or might not be abbreviated.
4535 The same fix has been applied.
4541 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
4542 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
4545 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
4546 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
4548 PH/03 Typo: missing ".o" in src/pcre/Makefile.
4550 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
4551 header line, restrict the check to what is listed in RFCs 2369 and 2929.
4552 Also, for "Auto-Submitted", treat anything other than "no" as
4553 non-personal, in accordance with RFC 3834. (Previously it treated
4554 anything starting "auto-" as non-personal.)
4556 TF/01 The control=submission/name=... option had a problem with syntax
4557 errors if the name included a slash character. The /name= option
4558 now slurps the rest of the string, so it can include any characters
4559 but it must come last in the list of options (after /sender_retain
4562 PH/05 Some modifications to the interface to the fake nameserver for the new
4570 TK/01 Added the "success_on_redirect" address verification option. See
4571 NewStuff for rationale and an example.
4573 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
4575 PH/02 Patch to exigrep to allow it to work on syslog lines.
4577 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
4578 fread() to skip over the body file's header line, because in Cygwin the
4579 header line is locked and is inaccessible.
4581 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
4582 co-exist for some time) to make it clear that it is the Exim ID that is
4583 referenced, not the Message-ID: header line.
4585 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
4586 string_format() function, because snprintf() does not exist on all
4589 PH/06 The use of forbid_filter_existstest now also locks out the use of the
4590 ${stat: expansion item.
4592 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
4593 protocol synchronization error", to keep the pedants happy.
4595 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
4596 well as for IRIX systems, when gcc is being used. See the host.c source
4599 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
4601 PH/10 Named domain lists were not working if used in a queue_smtp_domains
4604 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
4605 transport and to the smtp transport in LMTP mode.
4607 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
4609 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
4610 run a filter in a subprocess. This could lead to confusion in subsequent
4611 lookups in the parent process. There should also be a search_tidyup() at
4612 the end of the subprocess.
4614 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
4615 only if the host matched helo_try_verify_hosts, which caused the
4616 verification to occur when the EHLO/HELO command was issued. The ACL just
4617 tested the remembered result. Now, if a previous verification attempt has
4618 not happened, "verify = helo" does it there and then.
4620 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
4622 TK/03 Fix log output including CR from clamd.
4624 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
4625 exist provoked a memory error which could cause a segfault.
4627 PH/15 Installed PCRE 6.2
4629 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
4631 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
4632 of the problem. Specifically, suggested +O2 rather than +O1 for the
4635 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
4637 PH/20 If a delivery was routed to a non-standard port by means of an SRV
4638 record, the port was not correctly logged when the outgoing_port log
4639 selector was set (it logged the transort's default port).
4641 PH/21 Added support for host-specific ports to manualroute, queryprogram,
4642 fallback_hosts, and "hosts" in the smtp transport.
4644 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
4645 host errors such as "Connection refused".
4647 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
4648 authentication with radiusclient 0.4.9:
4650 - Error returned from rc_read_config was caught wrongly
4651 - Username/password not passed on to radius server due to wrong length.
4653 The presumption is that some radiusclient API changes for 4.51/PH/17
4654 were not taken care of correctly. The code is still untested by me (my
4655 Linux distribution still has 0.3.2 of radiusclient), but it was
4656 contributed by a Radius user.
4658 PH/24 When doing a callout, the value of $domain wasn't set correctly when
4659 expanding the "port" option of the smtp transport.
4661 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
4662 while reading a MIME header. Thanks to Tom Hughes for a patch.
4664 PH/24 Include config.h inside local_scan.h so that configuration settings are
4667 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
4668 that in an ACL for RCPT (for example), you can examine exactly what was
4671 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
4672 commands, but it was not correctly comparing the address with the actual
4673 client host address. Thus, it would show the EHLO address in Received:
4674 header lines when this was not necessary.
4676 PH/27 Added the % operator to ${eval:}.
4678 PH/28 Exim tries to create and chdir to its spool directory when it starts;
4679 it should be ignoring failures (because with -C, for example, it has lost
4680 privilege). It wasn't ignoring creation failures other than "already
4683 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
4686 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
4687 because an input line was too long, either on its own, or by virtue of
4688 too many continuations, the temporary file was not being removed, and the
4689 return code was incorrect.
4691 PH/31 Missing "BOOL" in function definition in filtertest.c.
4693 PH/32 Applied Sieve patches from the maintainer.
4695 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
4697 PH/33 Added "verify = not_blind".
4699 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
4700 Local/Makefile (with some defaults set). These are used in built scripts
4701 such as exicyclog, but they have never been used in the exim_install
4702 script (though there are many overriding facilities there). I have
4703 arranged that the exim_install script now takes note of these two
4706 PH/35 Installed configuration files for Dragonfly.
4708 PH/36 When a locally submitted message by a trusted user did not contain a
4709 From: header, and the sender address was obtained from -f or from an SMTP
4710 MAIL command, and the trusted user did not use -F to supply a sender
4711 name, $originator_name was incorrectly used when constructing a From:
4712 header. Furthermore, $originator_name was used for submission mode
4713 messages from external hosts without From: headers in a similar way,
4714 which is clearly wrong.
4716 PH/37 Added control=suppress_local_fixups.
4718 PH/38 When log_selector = +received_sender was set, and the addition of the
4719 sender made the log line's construction buffer exactly full, or one byte
4720 less than full, an overflow happened when the terminating "\n" was
4723 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
4724 when the result of a list match is failure because a DNS lookup failed.
4726 PH/40 RM_COMMAND is now used in the building process.
4728 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
4729 the "build-* directories that it finds.
4731 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
4732 address in a domain literal was a prefix of an interface address.
4734 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
4735 when verifying a sender address, unless rewrite_headers is false.
4737 PH/44 Wrote a long comment about why errors_to addresses are verified as
4738 recipients, not senders.
4740 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
4741 the ratelimit ACL was added.
4743 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
4745 PH/46 Added extra information about PostgreSQL errors to the error string.
4747 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
4748 avoiding the need to install special zones in a real server. This is
4749 backwards compatible; if it can't find the fake resolver, it drops back.
4750 Thus, both old and new test suites can be run.
4752 TF/02 Added util/ratelimit.pl
4754 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
4755 clock is set back in time.
4757 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
4758 Candler <B.Candler@pobox.com>.
4760 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
4761 OK for addresses that are the result of redirections.
4763 PH/48 A number of further additions for the benefit of the new test suite,
4764 including a fake gethostbyname() that interfaces to the fake DNS resolver
4767 TF/06 The fix for widen_domains has also been applied to qualify_single and
4768 search_parents which are the other dnslookup options that can cause
4771 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
4774 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
4776 TF/07 Exim produced the error message "an SRV record indicated no SMTP
4777 service" if it encountered an MX record with an empty target hostname.
4778 The message is now "an MX or SRV record indicated no SMTP service".
4780 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
4781 if the DNS of the sending site is misconfigured. This is quite a
4782 common situation. This change restores the behaviour of treating a
4783 helo verification defer as a failure.
4785 PH/51 If self=fail was set on a router, the bounce message did not include the
4786 actual error message.
4792 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
4794 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
4795 command itself ended in error, the underlying message about the transport
4796 filter timeout was being overwritten with the pipe command error. Now the
4797 underlying error message should be appended to the second error message.
4799 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4801 PH/02 Reduce the amount of output that "make" produces by default. Full output
4802 can still be requested.
4804 PH/03 The warning log line about a condition test deferring for a "warn" verb
4805 was being output only once per connection, rather than after each
4806 occurrence (because it was using the same function as for successful
4807 "warn" verbs). This seems wrong, so I have changed it.
4809 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4810 it should not have, which might have caused a crash in the right
4811 circumstances, but probably never did.
4813 PH/04 Installed a modified version of Tony Finch's patch to make submission
4814 mode fix the return path as well as the Sender: header line, and to
4815 add a /name= option so that you can make the user's friendly name appear
4818 TF/03 Added the control = fakedefer ACL modifier.
4820 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4821 Mark Lowes for thorough testing.
4823 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4825 TK/03 Merged latest SRS patch from Miles Wilton.
4827 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4828 with the definition in sysexits.h (which is #included earlier).
4829 Fortunately, Exim does not actually use EX_OK. The code used to try to
4830 preserve the sysexits.h value, by assuming that macro definitions were
4831 scanned for macro replacements. I have been disabused of this notion,
4832 so now the code just undefines EX_OK before #including unistd.h.
4834 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4835 in the smtp transport. When a block could not be written in a single
4836 write() function, the timeout was being re-applied to each part-write.
4837 This seems wrong - if the receiver was accepting one byte at a time it
4838 would take for ever. The timeout is now adjusted when this happens. It
4839 doesn't have to be particularly precise.
4841 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4842 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4844 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4845 without a domain if the check to <postmaster@domain> fails.
4847 SC/01 Eximstats: added -xls and the ability to specify output files
4848 (patch written by Frank Heydlauf).
4850 SC/02 Eximstats: use FileHandles for outputting results.
4852 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4854 SC/04 Eximstats: fixed display of large numbers with -nvr option
4856 SC/05 Eximstats: fixed merging of reports with empty tables.
4858 SC/06 Eximstats: added the -include_original_destination flag
4860 SC/07 Eximstats: removed tabs and trailing whitespace.
4862 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4864 TK/06 MBOX spool code: Add real "From " MBOX separator line
4865 so the .eml file is really in mbox format (even though
4866 most programs do not really care). Patch from Alex Miller.
4868 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4869 The latter is generated from $received_to and is only set if the
4870 message has one envelope recipient. SA can use these headers,
4871 obviously out-of-the-box. Patch from Alex Miller.
4873 PH/08 The ${def test on a variable was returning false if the variable's
4874 value was "0", contrary to what the specification has always said!
4875 The result should be true unless the variable is empty.
4877 PH/09 The syntax error of a character other than { following "${if
4878 def:variable_name" (after optional whitespace) was not being diagnosed.
4879 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4880 accidental colon was present, for example, could give incorrect results.
4882 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4883 result is used (not including appendfile, where other changes are about
4886 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4887 This involved changing a lot of size variables from int to off_t. It
4888 should work with maildirs and everything.
4890 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4891 spamd dying while we are connected to it.
4893 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4896 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4897 function for BDB 4.3.
4899 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4901 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4902 the sources for PCRE have been re-arranged and more files are now
4905 PH/15 The code I had for printing potentially long long variables in PH/11
4906 above was not the best (it lost precision). The length of off_t variables
4907 is now inspected at build time, and an appropriate printing format (%ld
4908 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4909 to be "long long int" or "long int". This is needed for the internal
4910 formatting function string_vformat().
4912 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4913 the configuration file to be ":syslog", then the script "guesses" where
4914 the logs files are, rather than using the compiled in default. In our
4915 case the guess is not the same as the compiled default, so the script
4916 suddenly stopped working when I started to use syslog. The patch checks
4917 to see if log_file_path is "". If so, it attempts to read it from exim
4918 with no configuration file to get the compiled in version, before it
4919 falls back to the previous guessing code."
4921 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4922 implementing BATV in an Exim configuration. See NewStuff for the gory
4925 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4926 Makefile that are specific to HP-UX.
4928 PH/18 If the "use_postmaster" option was set for a recipient callout together
4929 with the "random" option, the postmaster address was used as the MAIL
4930 FROM address for the random test, but not for the subsequent recipient
4931 test. It is now used for both.
4933 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4934 patch removes a few documentation additions to RFC 3028, because the
4935 latest draft now contains them. It adds the new en;ascii-case comparator
4936 and a new error check for 8bit text in MIME parts. Comparator and
4937 require names are now matched exactly. I enabled the subaddress
4938 extension, but it is not well tested yet (read: it works for me)."
4940 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4941 rework some of the code of TK/09 above to avoid the hardwired use of
4942 "%lld" and "long long". Replaced the call to snprintf() with a call to
4945 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4946 records point to non-existent hosts", "retry timeout exceeded", and
4947 "retry time not reached for any host after a long failure period".
4949 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4950 experimental DomainKeys support:
4952 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4953 (2) On an error such as an illegally used "control", the wrong name for
4954 the control was given.
4956 These problems did NOT occur unless DomainKeys support was compiled.
4958 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4960 PH/24 Added ${if match_ip condition.
4962 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4963 databases so that it will be absolutely obvious if a crash occurs in the
4964 DB library. This is a regular occurrence (often caused by mis-matched
4967 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4968 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4969 picked up on a user's system that detects such things. There doesn't seem
4970 to be a gcc warning option for this - only an attribute that has to be
4971 put on the function's prototype. It seems that in Fedora Core 4 they have
4972 set this on a number of new functions. No doubt there will be more in due
4975 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4976 specify a transport. However, if an address that was verified by such a
4977 router was the subject of a callout, Exim crashed because it tried to
4978 read the rcpt_include_affixes from the non-existent transport. Now it
4979 just assumes that the setting of that option is false. This bug was
4980 introduced by 4.51/PH/31.
4982 PH/28 Changed -d+all to exclude +memory, because that information is very
4983 rarely of interest, but it makes the output a lot bigger. People tend to
4984 do -d+all out of habit.
4986 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4987 code in os-type was giving problems when libc.so lives in lib64, like on
4990 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4991 aren't the modern standard, and it seems that some systems' include files
4992 don't always have them. Exim was already checking for some of the newer
4993 ones like T_AAAA, and defining it itself. I've added checks for all the
4994 record types that Exim uses.
4996 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4997 not automatically generating a new one, as it is supposed to. This
4998 prevented TLS from working. If the file did exist, but contained invalid
4999 data, a new version was generated, as expected. It was only the case of a
5000 non-existent file that was broken.
5002 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
5003 with a change in libdomainkeys > 0.64.
5005 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
5006 from DNS. If the selector record carries the flag, it now has
5007 precedence over the domain-wide flag.
5009 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
5011 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
5012 the use of a port name that isn't defined in /etc/services) occurred, the
5013 message was deferred as in a normal delivery, and thus remained on the
5014 spool, instead of being failed because of the mua_wrapper setting. This
5015 is now fixed, and I tidied up some of the mua_wrapper messages at the
5018 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
5019 the messages in a hash of arrays rather than using individual hashes.
5020 This is a bit cleaner and results in dramatic memory savings, albeit
5021 at a slight CPU cost.
5023 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
5024 as requested by Marc Sherman.
5026 SC/10 Eximstats: added histograms for user specified patterns as requested
5029 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
5031 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
5032 fopen() in the content-scanning modules that did not already have it.
5038 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
5039 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
5041 TK/02 Fix ACL "control" statement not being available in MIME ACL.
5043 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
5045 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
5046 to test Sieve filters that use "vacation".
5048 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
5049 that changes the way the GnuTLS parameters are stored in the cache file.
5050 The new format can be generated externally. For backward compatibility,
5051 if the data in the cache doesn't make sense, Exim assumes it has read an
5052 old-format file, and it generates new data and writes a new file. This
5053 means that you can't go back to an older release without removing the
5056 PH/03 A redirect router that has both "unseen" and "one_time" set does not
5057 work if there are any delivery delays because "one_time" forces the
5058 parent to be marked "delivered", so its unseen clone is never tried
5059 again. For this reason, Exim now forbids the simultaneous setting of
5062 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
5063 redirected to themselves ("homonym" addresses). Read the long ChangeLog
5064 entry if you want to know the details. The fix, however, neglected to
5065 consider the case when local delivery batching is involved. The test for
5066 "previously delivered" was not happening when checking to see if an
5067 address could be batched with a previous (undelivered) one; under
5068 certain circumstances this could lead to multiple deliveries to the same
5071 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
5072 in its include files, and this causes problems building Exim.
5074 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
5075 header_syntax) but Exim was just ignoring anything given after a slash.
5076 In particular, this caused confusion with an attempt to use "verify =
5077 reverse_host_lookup/defer_ok". An error is now given when options are
5078 supplied for verify items that do not have them. (Maybe reverse_host_
5079 lookup should have a defer_ok option, but that's a different point.)
5081 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
5082 defined by RFC 821) to 2048, because there were problems with some AUTH
5083 commands, and RFC 1869 says the size should be increased for extended
5084 SMTP commands that take arguments.
5086 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
5089 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
5090 "unknown" error; now it says that the functionality isn't in the binary.
5092 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
5093 an address' error message when a string expansion fails (syntax or
5094 whatever). Otherwise the password may appear in the log. Following change
5095 PH/42 below, there is no longer a chance of it appearing in a bounce
5098 PH/11 Installed exipick version 20050225.0 from John Jetmore.
5100 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
5101 of its addresses was ever tried. (Bugzilla bug #2.)
5103 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
5104 the result incorrectly in the debug output. (It correctly added a newline
5105 to what was transported.)
5107 TF/01 Added $received_time.
5109 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
5110 commented out examples of how to interface to a virus scanner and to
5111 SpamAssassin. Also added commented examples of av_scanner and
5112 spamd_address settings.
5114 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
5115 and controls are allowed in which ACLs. There were a couple of minor
5116 errors. Some of the entries in the conditions table (which is a table of
5117 where they are NOT allowed) were getting very unwieldy; rewrote them as a
5118 negation of where the condition IS allowed.
5120 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
5122 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
5123 header file does not have a version number, so I've had to invent a new
5124 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
5125 API. The code is untested by me (my Linux distribution still has 0.3.2 of
5126 radiusclient), but it was contributed by a Radius user.
5128 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
5129 files or directories, for OpenSSL.
5131 PH/19 When an Exim process that is running as root has to create an Exim log
5132 file, it does so in a subprocess that runs as exim:exim so as to get the
5133 ownership right at creation (otherwise, other Exim processes might see
5134 the file with the wrong ownership). There was no test for failure of this
5135 fork() call, which would lead to the process getting stuck as it waited
5136 for a non-existent subprocess. Forks do occasionally fail when resources
5137 run out. I reviewed all the other calls to fork(); they all seem to check
5140 PH/20 When checking for unexpected SMTP input at connect time (before writing
5141 the banner), Exim was not dealing correctly with a non-positive return
5142 from the read() function. If the client had disconnected by this time,
5143 the result was a log entry for a synchronization error with an empty
5144 string after "input=" when read() returned zero. If read() returned -1
5145 (an event I could not check), uninitialized data bytes were printed.
5146 There were reports of junk text (parts of files, etc) appearing after
5149 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
5151 PH/22 Added support for macro redefinition, and (re)definition in between
5152 driver and ACL definitions.
5154 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
5155 forgetting to use the resulting value; it was using the unexpanded value.
5157 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
5158 hadn't been configured. The fix is from Juergen Kreileder, who
5159 understands it better than I do:
5161 "Here's what I see happening with three configured cyrus_sasl
5162 authenticators configured (plain, login, cram-md5):
5164 On startup auth_cyrus_sasl_init() gets called for each of these.
5165 This means three calls to sasl_listmech() without a specified mech_list.
5166 => SASL tests which mechs of all available mechs actually work
5167 => three warnings about OTP not working
5168 => the returned list contains: plain, login, cram-md5, digest-md5, ...
5170 With the patch, sasl_listmech() also gets called three times. But now
5171 SASL's mech_list option is set to the server_mech specified in the the
5172 authenticator. Or in other words, the answer from sasl_listmech()
5173 gets limited to just the mech you're testing for (which is different
5175 => the return list contains just 'plain' or 'login', 'cram-md5' or
5176 nothing depending on the value of ob->server_mech.
5178 I've just tested the patch: Authentication still works fine,
5179 unavailable mechs specified in the exim configuration are still
5180 caught, and the auth.log warnings about OTP are gone."
5182 PH/25 When debugging is enabled, the contents of the command line are added
5183 to the debugging output, even when log_selector=+arguments is not
5186 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
5187 answer is "GNU", and only if the return is "GNU/something" is the answer
5190 PH/27 $acl_verify_message is now set immediately after the failure of a
5191 verification in an ACL, and so is available in subsequent modifiers. In
5192 particular, the message can be preserved by coding like this:
5194 warn !verify = sender
5195 set acl_m0 = $acl_verify_message
5197 Previously, $acl_verify_message was set only while expanding "message"
5198 and "log_message" when a very denied access.
5200 PH/28 Modified OS/os.c-Linux with
5202 -#ifndef OS_LOAD_AVERAGE
5203 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
5205 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
5206 nomenclature these days.)
5208 PH/29 Installed patch from the Sieve maintainer that adds the options
5209 sieve_useraddress and sieve_subaddress to the redirect router.
5211 PH/30 In these circumstances:
5212 . Two addresses routed to the same list of hosts;
5213 . First host does not offer TLS;
5214 . First host accepts first address;
5215 . First host gives temporary error to second address;
5216 . Second host offers TLS and a TLS session is established;
5217 . Second host accepts second address.
5218 Exim incorrectly logged both deliveries with the TLS parameters (cipher
5219 and peerdn, if requested) that were in fact used only for the second
5222 PH/31 When doing a callout as part of verifying an address, Exim was not paying
5223 attention to any local part prefix or suffix that was matched by the
5224 router that accepted the address. It now behaves in the same way as it
5225 does for delivery: the affixes are removed from the local part unless
5226 rcpt_include_affixes is set on the transport.
5228 PH/32 Add the sender address, as F=<...>, to the log line when logging a
5229 timeout during the DATA phase of an incoming message.
5231 PH/33 Sieve envelope tests were broken for match types other than :is. I have
5232 applied a patch sanctioned by the Sieve maintainer.
5234 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
5235 the uid or gid is negative. A case of a negative gid caused this to be
5236 noticed. The fix allows for either to be negative.
5238 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
5239 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
5242 PH/36 The change PH/12 above was broken. Fixed it.
5244 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
5245 the grounds that routing the same address twice would always produce the
5246 same answer. This might have been true once, but it is certainly no
5247 longer true now. Routing a child address may depend on the previous
5248 routing that produced that child. Some complicated redirection strategies
5249 went wrong when messages had multiple recipients, and made Exim's
5250 behaviour dependent on the order in which the addresses were given.
5252 I have moved the duplicate checking until after the routing is complete.
5253 Exim scans the addresses that are assigned to local and remote
5254 transports, and removes any duplicates. This means that more work will be
5255 done, as duplicates will always all be routed, but duplicates are
5256 presumably rare, so I don't expect this is of any significance.
5258 For deliveries to pipes, files, and autoreplies, the duplicate checking
5259 still happens during the routing process, since they are not going to be
5262 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
5263 It corrects a timeout issue with spamd. This is Ian's comment: "The
5264 background is that sometimes spamd either never reads data from a
5265 connection it has accepted, or it never writes response data. The exiscan
5266 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
5267 blindly assumes that writes won't block so it may never time out."
5269 PH/39 Allow G after quota size as well as K and M.
5271 PH/40 The value set for $authenticated_id in an authenticator may not contain
5272 binary zeroes or newlines because the value is written to log lines and
5273 to spool files. There was no check on this. Now the value is run through
5274 the string_printing() function so that such characters are converted to
5275 printable escape sequences.
5277 PH/41 $message_linecount is a new variable that contains the total number of
5278 lines in the message. Compare $body_linecount, which is the count for the
5281 PH/42 Exim no longer gives details of delivery errors for specific addresses in
5282 bounce and delay warning messages, except in certain special cases, which
5285 (a) An SMTP error message from a remote host;
5286 (b) A message specified in a :fail: redirection;
5287 (c) A message specified in a "fail" command in a system filter;
5288 (d) A message specified in a FAIL return from the queryprogram router;
5289 (e) A message specified by the cannot_route_message router option.
5291 In these cases only, Exim does include the error details in bounce and
5292 warning messages. There are also a few cases where bland messages such
5293 as "unrouteable address" or "local delivery error" are given.
5295 PH/43 $value is now also set for the "else" part of a ${run expansion.
5297 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
5298 being worked on, but at least Exim now implements the latest version to
5301 PH/45 In a pipe transport, although a timeout while waiting for the pipe
5302 process to complete was treated as a delivery failure, a timeout while
5303 writing the message to the pipe was logged, but erroneously treated as a
5304 successful delivery. Such timeouts include transport filter timeouts. For
5305 consistency with the overall process timeout, these timeouts are now
5306 treated as errors, giving rise to delivery failures by default. However,
5307 there is now a new Boolean option for the pipe transport called
5308 timeout_defer, which, if set TRUE, converts the failures into defers for
5309 both kinds of timeout. A transport filter timeout is now identified in
5312 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
5313 systems where "make" and "gmake" are different, calling "gmake" at top
5314 level broke things. I've arranged for the value of $(MAKE) to be passed
5315 from the Makefile to this script so that it can call the same version of
5319 A note about Exim versions 4.44 and 4.50
5320 ----------------------------------------
5322 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
5323 changes of various kinds. As a consequence, a big documentation update was
5324 needed. This delayed the release for rather longer than seemed good, especially
5325 in the light of a couple of (minor) security issues. Therefore, the changes
5326 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
5327 release. So 4.44 and 4.50 are in effect two different branches that both start
5330 I have left the 4.50 change log unchanged; it contains all the changes since
5331 4.43. The change log for 4.44 is below; many of its items are identical to
5332 those for 4.50. This seems to be the most sensible way to preserve the
5333 historical information.
5339 1. Minor wording change to the doc/README.SIEVE file.
5341 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
5342 computation of the current number of files was incorrect.
5344 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5345 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5348 4. Give more explanation in the error message when the command for a transport
5349 filter fails to execute.
5351 5. There are several places where Exim runs a non-Exim command in a
5352 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5353 was being done only for the command run by the queryprogram router. It is
5354 now done for all such subprocesses. The other cases are: ${run, transport
5355 filters, and the commands run by the lmtp and pipe transports.
5357 6. Added CONFIGURE_GROUP build-time option.
5359 7. Some older OS have a limit of 256 on the maximum number of file
5360 descriptors. Exim was using setrlimit() to set 1000 as a large value
5361 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5362 systems. I've change it so that if it can't get 1000, it tries for 256.
5364 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5365 was an oversight, and furthermore, ever since the addition of extra
5366 controls (e.g. 4.43/32), the checks on when to allow different forms of
5367 "control" were broken. There should now be diagnostics for all cases when a
5368 control that does not make sense is encountered.
5370 9. Added the /retain_sender option to "control=submission".
5372 10. $recipients is now available in the predata ACL (oversight).
5374 11. Tidy the search cache before the fork to do a delivery from a message
5375 received from the command line. Otherwise the child will trigger a lookup
5376 failure and thereby defer the delivery if it tries to use (for example) a
5377 cached ldap connection that the parent has called unbind on.
5379 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5380 of $address_data from the recipient verification was clobbered by the
5381 sender verification.
5383 13. The value of address_data from a sender verification is now available in
5384 $sender_address_data in subsequent conditions in the ACL statement.
5386 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
5388 15. Added a new option "connect=<time>" to callout options, to set a different
5391 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5392 was its contents. (It was OK if the option was not defined at all.)
5394 17. A "Completed" log line is now written for messages that are removed from
5395 the spool by the -Mrm option.
5397 18. New variables $sender_verify_failure and $recipient_verify_failure contain
5398 information about exactly what failed.
5400 19. Added -dd to debug only the daemon process.
5402 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
5403 handles timeouts, both on the server side and network timeouts. Renamed the
5404 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
5406 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
5407 It is now set to "smtps".
5409 22. $host_address is now set to the target address during the checking of
5410 ignore_target_hosts.
5412 23. When checking ignore_target_hosts for an ipliteral router, no host name was
5413 being passed; this would have caused $sender_host_name to have been used if
5414 matching the list had actually called for a host name (not very likely,
5415 since this list is usually IP addresses). A host name is now passed as
5418 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5419 code that specifies a non-restarting handler (typically sigaction() in
5420 modern systems) in an attempt to fix a rare and obscure crash bug.
5422 25. Narrowed the window for a race in the daemon that could cause it to ignore
5423 SIGCHLD signals. This is not a major problem, because they are used only to
5424 wake it up if nothing else does.
5426 26. A malformed maildirsize file could cause Exim to calculate negative values
5427 for the mailbox size or file count. Odd effects could occur as a result.
5428 The maildirsize information is now recalculated if the size or filecount
5431 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5432 support for a long time. Removed HAVE_SYS_VFS_H.
5434 28. Installed the latest version of exipick from John Jetmore.
5436 29. In an address list, if the pattern was not a regular expression, an empty
5437 subject address (from a bounce message) matched only if the pattern was an
5438 empty string. Non-empty patterns were not even tested. This was the wrong
5439 because it is perfectly reasonable to use an empty address as part of a
5440 database query. An empty address is now tested by patterns that are
5441 lookups. However, all the other forms of pattern expect the subject to
5442 contain a local part and a domain, and therefore, for them, an empty
5443 address still always fails if the pattern is not itself empty.
5445 30. Exim went into a mad DNS loop when attempting to do a callout where the
5446 host was specified on an smtp transport, and looking it up yielded more
5447 than one IP address.
5449 31. Re-factored the code for checking spool and log partition space into a
5450 function that finds that data and another that does the check. The former
5451 is then used to implement four new variables: $spool_space, $log_space,
5452 $spool_inodes, and $log_inodes.
5454 32. The RFC2047 encoding function was originally intended for short strings
5455 such as real names; it was not keeping to the 75-character limit for
5456 encoded words that the RFC imposes. It now respects the limit, and
5457 generates multiple encoded words if necessary. To be on the safe side, I
5458 have increased the buffer size for the ${rfc2047: expansion operator from
5461 33. It is now permitted to omit both strings after an "if" condition; if the
5462 condition is true, the result is "true". As before, when the second string
5463 is omitted, a false condition yields an empty string. This makes it less
5464 cumbersome to write custom ACL and router conditions.
5466 34. Failure to deliver a bounce message always caused it to be frozen, even if
5467 there was an errors_to setting on the router. The errors_to setting is now
5470 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5471 canonical form (fully expanded) before being placed in
5472 $sender_host_address.
5474 36. The table in the code that translates DNS record types into text (T_A to
5475 "A" for instance) was missing entries for NS and CNAME. It is just possible
5476 that this could have caused confusion if both these types were looked up
5477 for the same domain, because the text type is used as part of Exim's
5478 per-process caching. But the chance of anyone hitting this buglet seems
5481 37. The dnsdb lookup has been extended in a number of ways.
5483 (1) There is a new type, "zns", which walks up the domain tree until it
5484 finds some nameserver records. It should be used with care.
5486 (2) There is a new type, "mxh", which is like "mx" except that it returns
5487 just the host names, not the priorities.
5489 (3) It is now possible to give a list of domains (or IP addresses) to be
5490 looked up. The behaviour when one of the lookups defers can be
5491 controlled by a keyword.
5493 (4) It is now possible to specify the separator character for use when
5494 multiple records are returned.
5496 38. The dnslists ACL condition has been extended: it is now possible to supply
5497 a list of IP addresses and/or domains to be looked up in a particular DNS
5500 39. Added log_selector=+queue_time_overall.
5502 40. When running the queue in the test harness, wait just a tad after forking a
5503 delivery process, to get repeatability of debugging output.
5505 41. Include certificate and key file names in error message when GnuTLS fails
5506 to set them up, because the GnuTLS error message doesn't include the name
5507 of the failing file when there is a problem reading it.
5509 42. Allow both -bf and -bF in the same test run.
5511 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
5513 44. The "Exiscan patch" is now merged into the mainline Exim source.
5515 45. Sometimes the final signoff response after QUIT could fail to get
5516 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5517 < 0 before doing a fflush(). This bug looks as though it goes back to the
5518 introduction of TLS in release 3.20, but "sometimes" must have been rare
5519 because the tests only now provoked it.
5521 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
5522 (this can affect the format of dates).
5524 47. exim_tidydb, when checking for the continued existence of a message for
5525 which it has found a message-specific retry record, was not finding
5526 messages that were in split spool directories. Consequently, it was
5527 deleting retry records that should have stayed in existence.
5529 48. Steve fixed some bugs in eximstats.
5531 49. The SPA authentication driver was not abandoning authentication and moving
5532 on to the next authenticator when an expansion was forced to fail,
5533 contradicting the general specification for all authenticators. Instead it
5534 was generating a temporary error. It now behaves as specified.
5536 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
5537 (the order specifies the preference for clients). The order is now AES256,
5538 AES128, 3DES, ARCFOUR128.
5540 51. Small patch to Sieve code - explicitly set From: when generating an
5543 52. Exim crashed if a remote delivery caused a very long error message to be
5544 recorded - for instance if somebody sent an entire SpamAssassin report back
5545 as a large number of 550 error lines. This bug was coincidentally fixed by
5546 increasing the size of one of Exim's internal buffers (big_buffer) that
5547 happened as part of the Exiscan merge. However, to be on the safe side, I
5548 have made the code more robust (and fixed the comments that describe what
5551 53. Now that there can be additional text after "Completed" in log lines (if
5552 the queue_time_overall log selector is set), a one-byte patch to exigrep
5553 was needed to allow it to recognize "Completed" as not the last thing in
5556 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
5557 patch that reportedly fixes this has been added. I am not expert enough to
5558 create a test for it. This is what the patch creator wrote:
5560 "I found a little strange behaviour of ldap code when working with
5561 Windows 2003 AD Domain, where users was placed in more than one
5562 Organization Units. When I tried to give exim partial DN, the exit code
5563 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
5564 But simultaneously result of request was absolutely normal ldap result,
5565 so I produce this patch..."
5567 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
5568 REFERENCE, so I have modified the code to exclude the patch when that macro
5571 55. Some experimental protocols are using DNS PTR records for new purposes. The
5572 keys for these records are domain names, not reversed IP addresses. The
5573 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5574 leaves it alone. Component reversal etc. now happens only for IP addresses.
5577 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
5579 57. Double the size of the debug message buffer (to 2048) so that more of very
5580 long debug lines gets shown.
5582 58. The exicyclog utility now does better if the number of log files to keep
5583 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
5585 59. Two changes related to the smtp_active_hostname option:
5587 (1) $smtp_active_hostname is now available as a variable.
5588 (2) The default for smtp_banner uses $smtp_active_hostname instead
5589 of $primary_hostname.
5591 60. The host_aton() function is supposed to be passed a string that is known
5592 to be a valid IP address. However, in the case of IPv6 addresses, it was
5593 not checking this. This is a hostage to fortune. Exim now panics and dies
5594 if the condition is not met. A case was found where this could be provoked
5595 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5596 components; fortuitously, this particular loophole had already been fixed
5597 by change 4.50/55 above.
5599 If there are any other similar loopholes, the new check in host_aton()
5600 itself should stop them being exploited. The report I received stated that
5601 data on the command line could provoke the exploit when Exim was running as
5602 exim, but did not say which command line option was involved. All I could
5603 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5604 running as the user.
5607 61. There was a buffer overflow vulnerability in the SPA authentication code
5608 (which came originally from the Samba project). I have added a test to the
5609 spa_base64_to_bits() function which I hope fixes it.
5612 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
5613 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
5615 63. The daemon start-up calls getloadavg() while still root for those OS that
5616 need the first call to be done as root, but it missed one case: when
5617 deliver_queue_load_max is set with deliver_drop_privilege. This is
5618 necessary for the benefit of the queue runner, because there is no re-exec
5619 when deliver_drop_privilege is set.
5621 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
5622 This has been fixed.
5624 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5625 was in use, was not putting the data itself into the right store pool;
5626 consequently, it could be overwritten for a subsequent message in the same
5627 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5630 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
5632 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
5633 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
5634 the function were treating the return as a boolean value, which happened to
5635 work because 0=false and not-0=true, but is not correct code.
5637 68. The host_aton() function was not handling scoped IPv6 addresses (those
5638 with, for example, "%eth0" on the end) correctly.
5640 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
5641 negated items (that is, ~something) in unsigned ints. Some compilers
5642 apparently mutter when there is no cast.
5644 70. If an address verification called from an ACL failed, and did not produce a
5645 user-specific message (i.e. there was only a "system" message), nothing was
5646 put in $acl_verify_message. In this situation, it now puts the system
5649 71. Change 4.23/11 added synchronization checking at the start of an SMTP
5650 session; change 4.31/43 added the unwanted input to the log line - except
5651 that it did not do this in the start of session case. It now does.
5653 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
5654 This is wrong and can cause the other end to generate a synchronization
5655 error if it is another Exim or anything else that does the synchronization
5656 check. A QUIT command is no longer sent after a timeout.
5658 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
5659 during host lookups.
5661 74. The defer_ok option of callout verification was not working if it was used
5662 when verifying addresses in header lines, that is, for this case:
5664 verify = header_sender/callout=defer_ok
5666 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
5667 those file descriptors could be used for SMTP connections. If anything
5668 wrote to stderr (the example that came up was "warn" in embedded Perl), it
5669 could be sent to the SMTP client, causing chaos. The daemon now opens
5670 stdin, stdout, and stderr to /dev/null when it puts itself into the
5673 76. Arrange for output from Perl's "warn" command to be written to Exim's main
5674 log by default. The user can override this with suitable Perl magic.
5676 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
5677 the log message when discard triggers, was not working for the DATA ACL or
5678 for the non-SMTP ACL.
5680 78. Error message wording change in sieve.c.
5682 79. If smtp_accept_max_per_host was set, the number of connections could be
5683 restricted to fewer than expected, because the daemon was trying to set up
5684 a new connection before checking whether the processes handling previous
5685 connections had finished. The check for completed processes is now done
5686 earlier. On busy systems, this bug wouldn't be noticed because something
5687 else would have woken the daemon, and it would have reaped the completed
5690 80. If a message was submitted locally by a user whose login name contained one
5691 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
5692 It caused a spool format error. I have fixed the spool reading code. A
5693 related problem was that the "from" clause in the Received: line became
5694 illegal because of the space(s). It is now covered by ${quote_local_part.
5696 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
5699 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
5701 83. Patch from Sieve maintainer to fix unterminated string problem in
5702 "vacation" handling.
5704 84. Some minor changes to the Linux configuration files to help with other
5705 OS variants using glibc.
5707 85. One more patch for Sieve to update vacation handling to latest spec.
5710 ----------------------------------------------------
5711 See the note above about the 4.44 and 4.50 releases.
5712 ----------------------------------------------------
5718 1. Change 4.43/35 introduced a bug that caused file counts to be
5719 incorrectly computed when quota_filecount was set in an appendfile
5722 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5723 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5726 3. Give more explanation in the error message when the command for a transport
5727 filter fails to execute.
5729 4. There are several places where Exim runs a non-Exim command in a
5730 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5731 was being done only for the command run by the queryprogram router. It is
5732 now done for all such subprocesses. The other cases are: ${run, transport
5733 filters, and the commands run by the lmtp and pipe transports.
5735 5. Some older OS have a limit of 256 on the maximum number of file
5736 descriptors. Exim was using setrlimit() to set 1000 as a large value
5737 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5738 systems. I've change it so that if it can't get 1000, it tries for 256.
5740 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5741 was an oversight, and furthermore, ever since the addition of extra
5742 controls (e.g. 4.43/32), the checks on when to allow different forms of
5743 "control" were broken. There should now be diagnostics for all cases when a
5744 control that does not make sense is encountered.
5746 7. $recipients is now available in the predata ACL (oversight).
5748 8. Tidy the search cache before the fork to do a delivery from a message
5749 received from the command line. Otherwise the child will trigger a lookup
5750 failure and thereby defer the delivery if it tries to use (for example) a
5751 cached ldap connection that the parent has called unbind on.
5753 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5754 of $address_data from the recipient verification was clobbered by the
5755 sender verification.
5757 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5758 was its contents. (It was OK if the option was not defined at all.)
5760 11. A "Completed" log line is now written for messages that are removed from
5761 the spool by the -Mrm option.
5763 12. $host_address is now set to the target address during the checking of
5764 ignore_target_hosts.
5766 13. When checking ignore_target_hosts for an ipliteral router, no host name was
5767 being passed; this would have caused $sender_host_name to have been used if
5768 matching the list had actually called for a host name (not very likely,
5769 since this list is usually IP addresses). A host name is now passed as
5772 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5773 code that specifies a non-restarting handler (typically sigaction() in
5774 modern systems) in an attempt to fix a rare and obscure crash bug.
5776 15. Narrowed the window for a race in the daemon that could cause it to ignore
5777 SIGCHLD signals. This is not a major problem, because they are used only to
5778 wake it up if nothing else does.
5780 16. A malformed maildirsize file could cause Exim to calculate negative values
5781 for the mailbox size or file count. Odd effects could occur as a result.
5782 The maildirsize information is now recalculated if the size or filecount
5785 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5786 support for a long time. Removed HAVE_SYS_VFS_H.
5788 18. Updated exipick to current release from John Jetmore.
5790 19. Allow an empty sender to be matched against a lookup in an address list.
5791 Previously the only cases considered were a regular expression, or an
5794 20. Exim went into a mad DNS lookup loop when doing a callout where the
5795 host was specified on the transport, if the DNS lookup yielded more than
5798 21. The RFC2047 encoding function was originally intended for short strings
5799 such as real names; it was not keeping to the 75-character limit for
5800 encoded words that the RFC imposes. It now respects the limit, and
5801 generates multiple encoded words if necessary. To be on the safe side, I
5802 have increased the buffer size for the ${rfc2047: expansion operator from
5805 22. Failure to deliver a bounce message always caused it to be frozen, even if
5806 there was an errors_to setting on the router. The errors_to setting is now
5809 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5810 canonical form (fully expanded) before being placed in
5811 $sender_host_address.
5813 24. Updated eximstats to version 1.33
5815 25. Include certificate and key file names in error message when GnuTLS fails
5816 to set them up, because the GnuTLS error message doesn't include the name
5817 of the failing file when there is a problem reading it.
5819 26. Expand error message when OpenSSL has problems setting up cert/key files.
5822 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5823 (this can affect the format of dates).
5825 28. exim_tidydb, when checking for the continued existence of a message for
5826 which it has found a message-specific retry record, was not finding
5827 messages that were in split spool directories. Consequently, it was
5828 deleting retry records that should have stayed in existence.
5830 29. eximstats updated to version 1.35
5831 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5832 1.35 - bugfix such that pie charts by volume are generated correctly
5834 30. The SPA authentication driver was not abandoning authentication and moving
5835 on to the next authenticator when an expansion was forced to fail,
5836 contradicting the general specification for all authenticators. Instead it
5837 was generating a temporary error. It now behaves as specified.
5839 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5840 (the order specifies the preference for clients). The order is now AES256,
5841 AES128, 3DES, ARCFOUR128.
5843 31. Small patch to Sieve code - explicitly set From: when generating an
5846 32. Exim crashed if a remote delivery caused a very long error message to be
5847 recorded - for instance if somebody sent an entire SpamAssassin report back
5848 as a large number of 550 error lines. This bug was coincidentally fixed by
5849 increasing the size of one of Exim's internal buffers (big_buffer) that
5850 happened as part of the Exiscan merge. However, to be on the safe side, I
5851 have made the code more robust (and fixed the comments that describe what
5854 33. Some experimental protocols are using DNS PTR records for new purposes. The
5855 keys for these records are domain names, not reversed IP addresses. The
5856 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5857 leaves it alone. Component reversal etc. now happens only for IP addresses.
5860 34. The host_aton() function is supposed to be passed a string that is known
5861 to be a valid IP address. However, in the case of IPv6 addresses, it was
5862 not checking this. This is a hostage to fortune. Exim now panics and dies
5863 if the condition is not met. A case was found where this could be provoked
5864 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5865 components; fortuitously, this particular loophole had already been fixed
5866 by change 4.50/55 or 4.44/33 above.
5868 If there are any other similar loopholes, the new check in host_aton()
5869 itself should stop them being exploited. The report I received stated that
5870 data on the command line could provoke the exploit when Exim was running as
5871 exim, but did not say which command line option was involved. All I could
5872 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5873 running as the user.
5876 35. There was a buffer overflow vulnerability in the SPA authentication code
5877 (which came originally from the Samba project). I have added a test to the
5878 spa_base64_to_bits() function which I hope fixes it.
5881 36. The daemon start-up calls getloadavg() while still root for those OS that
5882 need the first call to be done as root, but it missed one case: when
5883 deliver_queue_load_max is set with deliver_drop_privilege. This is
5884 necessary for the benefit of the queue runner, because there is no re-exec
5885 when deliver_drop_privilege is set.
5887 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5888 was in use, was not putting the data itself into the right store pool;
5889 consequently, it could be overwritten for a subsequent message in the same
5890 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5893 38. Sometimes the final signoff response after QUIT could fail to get
5894 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5895 < 0 before doing a fflush(). This bug looks as though it goes back to the
5896 introduction of TLS in release 3.20, but "sometimes" must have been rare
5897 because the tests only now provoked it.
5903 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5904 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5905 Now it returns an integer. A number of calls were still expecting a T/F
5906 return. Fortuitously, in all cases, the tests worked in OK situations,
5907 which is the norm. However, things would have gone wrong on any write
5908 failures on the smtp file descriptor. This function is used when sending
5909 messages over SMTP and also when doing verify callouts.
5911 2. When Exim is called to do synchronous delivery of a locally submitted
5912 message (the -odf or -odi options), it no longer closes stderr before doing
5915 3. Implemented the mua_wrapper option.
5917 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5919 5. Implemented the functions header_remove(), header_testname(),
5920 header_add_at_position(), and receive_remove_recipient(), and exported them
5923 6. If an ACL "warn" statement specified the addition of headers, Exim already
5924 inserted X-ACL-Warn: at the start if there was no header name. However, it
5925 was not making this test for the second and subsequent header lines if
5926 there were newlines in the string. This meant that an invalid header could
5927 be inserted if Exim was badly configured.
5929 7. Allow an ACL "warn" statement to add header lines at the start or after all
5930 the Received: headers, as well as at the end.
5932 8. Added the rcpt_4xx retry error code.
5934 9. Added postmaster_mailfrom=xxx to callout verification option.
5936 10. Added mailfrom=xxxx to the callout verification option, for verify=
5939 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5940 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5942 12. Inserted some casts to stop certain compilers warning when using pointer
5943 differences as field lengths or precisions in printf-type calls (mostly
5944 affecting debugging statements).
5946 13. Added optional readline() support for -be (dynamically loaded).
5948 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5949 same clock tick as a message's arrival, so that its received time was the
5950 same as the "first fail" time on the retry record, and that message
5951 remained on the queue past the ultimate address timeout, every queue runner
5952 would try a delivery (because it was past the ultimate address timeout) but
5953 after another failure, the ultimate address timeout, which should have then
5954 bounced the address, did not kick in. This was a "< instead of <=" error;
5955 in most cases the first failure would have been in the next clock tick
5956 after the received time, and all would be well.
5958 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5959 being recognized when the domain list was tested by the match_domain
5960 condition in an expansion string.
5962 16. Added the ${str2b64: operator.
5964 17. Exim was always calling setrlimit() to set a large limit for the number of
5965 processes, without checking whether the existing limit was already
5966 adequate. (It did check for the limit on file descriptors.) Furthermore,
5967 errors from getrlimit() and setrlimit() were being ignored. Now they are
5968 logged to the main and panic logs, but Exim does carry on, to try to do its
5969 job under whatever limits there are.
5971 18. Imported PCRE 5.0.
5973 19. Trivial typo in log message " temporarily refused connection" (the leading
5976 20. If the log selector return_path_on_delivery was set and an address was
5977 redirected to /dev/null, the delivery process crashed because it assumed
5978 that a return path would always be set for a "successful" delivery. In this
5979 case, the whole delivery is bypassed as an optimization, and therefore no
5982 21. Internal re-arrangement: the function for sending a challenge and reading
5983 a response while authentication was assuming a zero-terminated challenge
5984 string. It's now changed to take a pointer and a length, to allow for
5985 binary data in such strings.
5987 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5989 23. Exim was not respecting finduser_retries when seeking the login of the
5990 uid under which it was called; it was always trying 10 times. (The default
5991 setting of finduser_retries is zero.) Also, it was sleeping after the final
5992 failure, which is pointless.
5994 24. Implemented tls_on_connect_ports.
5996 25. Implemented acl_smtp_predata.
5998 26. If the domain in control=submission is set empty, Exim assumes that the
5999 authenticated id is a complete email address when it generates From: or
6000 Sender: header lines.
6002 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
6003 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
6004 chown and chgrp in /bin and hostname in /usr/bin.
6006 28. Exim was keeping the "process log" file open after each use, just as it
6007 does for the main log. This opens the possibility of it remaining open for
6008 long periods when the USR1 signal hits a daemon. Occasional processlog
6009 errors were reported, that could have been caused by this. Anyway, it seems
6010 much more sensible not to leave this file open at all, so that is what now
6013 29. The long-running daemon process does not normally write to the log once it
6014 has entered its main loop, and it closes the log before doing so. This is
6015 so that log files can straightforwardly be renamed and moved. However,
6016 there are a couple of unusual error situations where the daemon does write
6017 log entries, and I had neglected to close the log afterwards.
6019 30. The text of an SMTP error response that was received during a remote
6020 delivery was being truncated at 512 bytes. This is too short for some of
6021 the long messages that one sometimes sees. I've increased the limit to
6024 31. It is now possible to make retry rules that apply only when a message has a
6025 specific sender, in particular, an empty sender.
6027 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
6028 it possible to be selective about when SMTP synchronization is enforced.
6030 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
6032 32. Implemented hosts_connection_nolog.
6034 33. Added an ACL for QUIT.
6036 34. Setting "delay_warning=" to disable warnings was not working; it gave a
6039 35. Added mailbox_size and mailbox_filecount to appendfile.
6041 36. Added control = no_multiline_responses to ACLs.
6043 37. There was a bug in the logic of the code that waits for the clock to tick
6044 in the case where the clock went backwards by a substantial amount such
6045 that the microsecond fraction of "now" was more than the microsecond
6046 fraction of "then" (but the whole seconds number was less).
6048 38. Added support for the libradius Radius client library this is found on
6049 FreeBSD (previously only the radiusclient library was supported).
6055 1. When certain lookups returned multiple values in the form name=value, the
6056 quoting of the values was not always being done properly. Specifically:
6057 (a) If the value started with a double quote, but contained no whitespace,
6059 (b) If the value contained whitespace other than a space character (i.e.
6060 tabs or newlines or carriage returns) it was not quoted.
6061 This fix has been applied to the mysql and pgsql lookups by writing a
6062 separate quoting function and calling it from the lookup code. The fix
6063 should probably also be applied to nisplus, ibase and oracle lookups, but
6064 since I cannot test any of those, I have not disturbed their existing code.
6066 2. A hit in the callout cache for a specific address caused a log line with no
6067 reason for rejecting RCPT. Now it says "Previous (cached) callout
6068 verification failure".
6070 3. There was an off-by-one bug in the queryprogram router. An over-long
6071 return line was truncated at 256 instead of 255 characters, thereby
6072 overflowing its buffer with the terminating zero. As well as fixing this, I
6073 have increased the buffer size to 1024 (and made a note to document this).
6075 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
6076 when Exim is waiting for an SMTP response from a remote server, Exim
6077 restarts its select() call on the socket, thereby resetting its timeout.
6078 This is not a problem when such interrupts are rare. Somebody set up a cron
6079 job to run exiwhat every 2 minutes, which is less than the normal select()
6080 timeout (5 or 10 minutes). This meant that the select() timeout never
6081 kicked in because it was always reset. I have fixed this by comparing the
6082 time when an interrupt arrives with the time at the start of the first call
6083 to select(). If more time than the timeout has elapsed, the interrupt is
6084 treated as a timeout.
6086 5. Some internal re-factoring in preparation for the addition of Sieve
6087 extensions (by MH). In particular, the "personal" test is moved to a
6088 separate function, and given an option for scanning Cc: and Bcc: (which is
6089 not set for Exim filters).
6091 6. When Exim created an email address using the login of the caller as the
6092 local part (e.g. when creating a From: or Sender: header line), it was not
6093 quoting the local part when it contained special characters such as @.
6095 7. Installed new OpenBSD configuration files.
6097 8. Reworded some messages for syntax errors in "and" and "or" conditions to
6098 try to make them clearer.
6100 9. Callout options, other than the timeout value, were being ignored when
6101 verifying sender addresses in header lines. For example, when using
6103 verify = header_sender/callout=no_cache
6105 the cache was (incorrectly) being used.
6107 10. Added a missing instance of ${EXE} to the exim_install script; this affects
6108 only the Cygwin environment.
6110 11. When return_path_on_delivery was set as a log selector, if different remote
6111 addresses in the same message used different return paths and parallel
6112 remote delivery occurred, the wrong values would sometimes be logged.
6113 (Whenever a remote delivery process finished, the return path value from
6114 the most recently started remote delivery process was logged.)
6116 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
6117 lines when AUTH and/or TLS are in use. This is the "received protocol"
6118 field. Exim used to use "asmtp" for authenticated SMTP, without any
6119 indication (in the protocol name) for TLS use. Now it follows the RFC and
6120 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
6121 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
6122 names appear in log lines as well as in Received: header lines.
6124 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
6125 extensions, and comparison tests, and to fix some bugs.
6127 14. Changes to the "personal" filter test:
6129 (1) The test was buggy in that it was just doing the equivalent of
6130 "contains" tests on header lines. For example, if a user's address was
6131 anne@some.where, the "personal" test would incorrectly be true for
6133 To: susanne@some.where
6135 This test is now done by extracting each address from the header in turn,
6136 and checking the entire address. Other tests that are part of "personal"
6137 are now done using regular expressions (for example, to check local parts
6138 of addresses in From: header lines).
6140 (2) The list of non-personal local parts in From: addresses has been
6141 extended to include "listserv", "majordomo", "*-request", and "owner-*",
6142 taken from the Sieve specification recommendations.
6144 (3) If the message contains any header line starting with "List-" it is
6145 treated as non-personal.
6147 (4) The test for "circular" in the Subject: header line has been removed
6148 because it now seems ill-conceived.
6150 15. Minor typos in src/EDITME comments corrected.
6152 16. Installed latest exipick from John Jetmore.
6154 17. If headers_add on a router specified a text string that was too long for
6155 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
6156 of string_sprintf() is now avoided.
6158 18. $message_body_size was not set (it was always zero) when running the DATA
6159 ACL and the local_scan() function.
6161 19. For the "mail" command in an Exim filter, no default was being set for
6162 the once_repeat time, causing a random time value to be used if "once" was
6163 specified. (If the value happened to be <= 0, no repeat happened.) The
6164 default is now 0s, meaning "never repeat". The "vacation" command was OK
6165 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
6166 (I found it when inspecting the code).
6168 20. There is now an overall timeout for performing a callout verification. It
6169 defaults to 4 times the callout timeout, which applies to individual SMTP
6170 commands during the callout. The overall timeout applies when there is more
6171 than one host that can be tried. The timeout is checked before trying the
6172 next host. This prevents very long delays if there are a large number of
6173 hosts and all are timing out (e.g. when the network connections are timing
6174 out). The value of the overall timeout can be changed by specifying an
6175 additional sub-option for "callout", called "maxwait". For example:
6177 verify = sender/callout=5s,maxwait=20s
6179 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
6180 to the end before writing, but this should make it even safer).
6182 22. Exim was forgetting that it had advertised PIPELINING for the second and
6183 subsequent messages on an SMTP connection. It was also not resetting its
6184 memory on STARTTLS and an internal HELO.
6186 23. When Exim logs an SMTP synchronization error within a session, it now
6187 records whether PIPELINING has been advertised or not.
6189 24. Added 3 instances of "(long int)" casts to time_t variables that were being
6190 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
6191 rather than long int.
6193 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
6195 26. Added the never_mail option to autoreply.
6201 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
6202 crash if the getsockname() call failed; this can happen if a connection is
6203 closed very soon after it is established. The problem was simply in the
6204 order in which certain operations were done, causing Exim to try to write
6205 to the SMTP stream before it had set up the file descriptor. The bug has
6206 been fixed by making things happen in the correct order.
6212 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
6213 before the connection was closed, thus losing the rejection response.
6215 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
6216 some early Solaris releases, but causes trouble in current releases where
6217 socklen_t is defined.
6219 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
6222 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
6225 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
6226 that releasing the top of it at the end releases what was used for sub-
6227 expansions (unless the block got too big). However, discard this block if
6228 the first thing is a variable or header, so that we can use its block when
6229 it is dynamic (useful for very large $message_headers, for example).
6231 6. Lookups now cache *every* query, not just the most recent. A new, separate
6232 store pool is used for this. It can be recovered when all lookup caches are
6233 flushed. Lookups now release memory at the end of their result strings.
6234 This has involved some general refactoring of the lookup sources.
6236 7. Some code has been added to the store_xxx() functions to reduce the amount
6237 of flapping under certain conditions.
6239 8. log_incoming_interface used to affect only the <= reception log lines. Now
6240 it causes the local interface and port to be added to several more SMTP log
6241 lines, for example "SMTP connection from", and rejection lines.
6243 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
6245 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
6247 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
6248 could be overwritten at the end of the current message (or the start of a
6249 new message if it was set in a HELO ACL). The value is now preserved for
6250 the duration of the SMTP connection.
6252 12. If a transport had a headers_rewrite setting, and a matching header line
6253 contained an unqualified address, that address was qualified, even if it
6254 did not match any rewriting rules. The underlying bug was that the values
6255 of the flags that permit the existence of unqualified sender and recipient
6256 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
6257 non-local messages, and by -bnq for local messages) were not being
6258 preserved with the message after it was received.
6260 13. When Exim was logging an SMTP synchronization error, it could sometimes log
6261 "next input=" as part of the text comprising the host identity instead of
6262 the correct text. The code was using the same buffer for two different
6263 strings. However, depending on which order the printing function evaluated
6264 its arguments, the bug did not always show up. Under Linux, for example, my
6265 test suite worked just fine.
6267 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
6268 doesn't work with some older versions of Perl. It has been changed to "my",
6269 which in any case is probably the better facility to use.
6271 15. A really picky compiler found some instances of statements for creating
6272 error messages that either had too many or two few arguments for the format
6275 16. The size of the buffer for calls to the DNS resolver has been increased
6276 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
6277 for addresses that have a lot of PTR records. This alleviates a problem; it
6278 does not fully solve it.
6280 17. A dnsdb lookup for PTR records that receives more data than will fit in the
6281 buffer now truncates the list and logs the incident, which is the same
6282 action as happens when Exim is looking up a host name and its aliases.
6283 Previously in this situation something unpredictable would happen;
6284 sometimes it was "internal error: store_reset failed".
6286 18. If a server dropped the connection unexpectedly when an Exim client was
6287 using GnuTLS and trying to read a response, the client delivery process
6288 crashed while trying to generate an error log message.
6290 19. If a "warn" verb in an ACL added multiple headers to a message in a single
6291 string, for example:
6293 warn message = H1: something\nH2: something
6295 the text was added as a single header line from Exim's point of view
6296 though it ended up OK in the delivered message. However, searching for the
6297 second and subsequent header lines using $h_h2: did not work. This has been
6298 fixed. Similarly, if a system filter added multiple headers in this way,
6299 the routers could not see them.
6301 20. Expanded the error message when iplsearch is called with an invalid key to
6302 suggest using net-iplsearch in a host list.
6304 21. When running tests using -bh, any delays imposed by "delay" modifiers in
6305 ACLs are no longer actually imposed (and a message to that effect is
6308 22. If a "gecos" field in a passwd entry contained escaped characters, in
6309 particular, if it contained a \" sequence, Exim got it wrong when building
6310 a From: or a Sender: header from that name. A second bug also caused
6311 incorrect handling when an unquoted " was present following a character
6312 that needed quoting.
6314 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
6315 was not being matched caselessly.
6317 24. Arranged for all hyphens in the exim.8 source to be escaped with
6320 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
6321 properly. Recipient callout cache records were still being keyed to include
6322 the sender, even when use_sender was set false. This led to far more
6323 callouts that were necessary. The sender is no longer included in the key
6324 when use_sender is false.
6326 26. Added "control = submission" modifier to ACLs.
6328 27. Added the ${base62d: operator to decode base 62 numbers.
6330 28. dnsdb lookups can now access SRV records.
6332 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
6333 the configuration file.
6335 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
6336 (-v) mode. This makes the output for a verbose queue run more intelligible.
6338 31. Added a use_postmaster feature to recipient callouts.
6340 32. Added the $body_zerocount variable, containing the number of binary zero
6341 bytes in the message body.
6343 33. The time of last modification of the "new" subdirectory is now used as the
6344 "mailbox time last read" when there is a quota error for a maildir
6347 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
6349 35. Added +ignore_unknown as a special item in host lists.
6351 36. Code for decoding IPv6 addresses in host lists is now included, even if
6352 IPv6 support is not being compiled. This fixes a bug in which an IPv6
6353 address was recognized as an IP address, but was then not correctly decoded
6354 into binary, causing unexpected and incorrect effects when compared with
6361 1. Very minor rewording of debugging text in manualroute to say "list of
6362 hosts" instead of "hostlist".
6364 2. If verify=header_syntax was set, and a header line with an unqualified
6365 address (no domain) and a large number of spaces between the end of the
6366 name and the colon was received, the reception process suffered a buffer
6367 overflow, and (when I tested it) crashed. This was caused by some obsolete
6368 code that should have been removed. The fix is to remove it!
6370 3. When running in the test harness, delay a bit after writing a bounce
6371 message to get a bit more predictability in the log output.
6373 4. Added a call to search_tidyup() just before forking a reception process. In
6374 theory, someone could use a lookup in the expansion of smtp_accept_max_
6375 per_host which, without the tidyup, could leave open a database connection.
6377 5. Added the variables $recipient_data and $sender_data which get set from a
6378 lookup success in an ACL "recipients" or "senders" condition, or a router
6379 "senders" option, similar to $domain_data and $local_part_data.
6381 6. Moved the writing of debug_print from before to after the "senders" test
6384 7. Change 4.31/66 (moving the time when the Received: is generated) caused
6385 problems for message scanning, either using a data ACL, or using
6386 local_scan() because the Received: header was not generated till after they
6387 were called (in order to set the time as the time of reception completion).
6388 I have revised the way this works. The header is now generated after the
6389 body is received, but before the ACL or local_scan() are called. After they
6390 are run, the timestamp in the header is updated.
6396 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
6397 before starting a queue runner without re-exec. This happened only when
6398 deliver_drop_privilege was set or when the Exim user was set to root. The
6399 effect of the bug was that timeouts during subsequent deliveries caused
6400 crashes instead of being properly handled. The handler is now left at its
6401 default (and expected) setting.
6403 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
6404 message, again when deliver_drop_privilege is set or Exim is run as root.
6405 The bug described in (1) was not present in this case, but the tidying up
6406 of the other signals was missing. I have made the two cases consistent.
6408 3. The ignore_target_hosts setting on a manualroute router was being ignored
6409 for hosts that were looked up using the /MX notation.
6411 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
6414 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
6415 operated on the sender address. After changing the $sender_address to <>
6416 for the sender address verify, Exim was re-instated it as the original
6417 (before rewriting) address, but remembering that it had rewritten it, so it
6418 wasn't rewriting it again. This bug also had the effect of breaking the
6419 sender address verification caching when the sender address was rewritten.
6421 6. The ignore_target_hosts option was being ignored by the ipliteral router.
6422 This has been changed so that if the ip literal address matches
6423 ignore_target_hosts, the router declines.
6425 7. Added expansion conditions match_domain, match_address, and match_local_
6426 part (NOT match_host).
6428 8. The placeholder for the Received: header didn't have a length field set.
6430 9. Added code to Exim itself and to exim_lock to test for a specific race
6431 condition that could lead to file corruption when using MBX delivery. The
6432 issue is with the lockfile that is created in /tmp. If this file is removed
6433 after a process has opened it but before that process has acquired a lock,
6434 there is the potential for a second process to recreate the file and also
6435 acquire a lock. This could lead to two Exim processes writing to the file
6436 at the same time. The added code performs the same test as UW imapd; it
6437 checks after acquiring the lock that its file descriptor still refers to
6438 the same named file.
6440 10. The buffer for building added header lines was of fixed size, 8192 bytes.
6441 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
6444 11. Added the smtp_active_hostname option. If used, this will typically be made
6445 to depend on the incoming interface address. Because $interface_address is
6446 not set up until the daemon has forked a reception process, error responses
6447 that can happen earlier (such as "too many connections") no longer contain
6450 12. If an expansion in a condition on a "warn" statement fails because a lookup
6451 defers, the "warn" statement is abandoned, and the next ACL statement is
6452 processed. Previously this caused the whole ACL to be aborted.
6454 13. Added the iplsearch lookup type.
6456 14. Added ident_timeout as a log selector.
6458 15. Added tls_certificate_verified as a log selector.
6460 16. Added a global option tls_require_ciphers (compare the smtp transport
6461 option of the same name). This controls incoming TLS connections.
6463 17. I finally figured out how to make tls_require_ciphers do a similar thing
6464 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
6465 before starting the TLS session.
6467 18. Tabs are now shown as \t in -bP output.
6469 19. If the log selector return_path_on_delivery was set, Exim crashed when
6470 bouncing a message because it had too many Received: header lines.
6472 20. If two routers both had headers_remove settings, and the first one included
6473 a superfluous trailing colon, the final name in the first list and the
6474 first name in the second list were incorrectly joined into one item (with a
6475 colon in the middle).
6481 1. Added -C and -D options to the exinext utility, mainly to make it easier
6482 to include in the automated testing, but these could be helpful when
6483 multiple configurations are in use.
6485 2. The exinext utility was not formatting the output nicely when there was
6486 an alternate port involved in the retry record key, nor when there was a
6487 message id as well (for retries that were specific to a specific message
6488 and a specific host). It was also confused by IPv6 addresses, because of
6489 the additional colons they contain. I have fixed the IPv4 problem, and
6490 patched it up to do a reasonable job for IPv6.
6492 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
6493 delivery, the log line now contains "pipelined" if PIPELINING was used.
6495 4. An SMTP transport process used to panic and die if the bind() call to set
6496 an explicit outgoing interface failed. This has been changed; it is now
6497 treated in the same way as a connect() failure.
6499 5. A reference to $sender_host_name in the part of a conditional expansion
6500 that was being skipped was still causing a DNS lookup. This no longer
6503 6. The def: expansion condition was not recognizing references to header lines
6504 that used bh_ and bheader_.
6506 7. Added the _cache feature to named lists.
6508 8. The code for checking quota_filecount in the appendfile transport was
6509 allowing one more file than it should have been.
6511 9. For compatibility with Sendmail, the command line option
6519 and sets the incoming protocol and host name (for trusted callers). The
6520 host name and its colon can be omitted when only the protocol is to be set.
6521 Note the Exim already has two private options, -pd and -ps, that refer to
6522 embedded Perl. It is therefore impossible to set a protocol value of "d" or
6523 "s", but I don't think that's a major issue.
6525 10. A number of refactoring changes to the code, none of which should affect
6528 (a) The number of logging options was getting close to filling up the
6529 32-bit word that was used as a bit map. I have split them into two classes:
6530 those that are passed in the argument to log_write(), and those that are
6531 only ever tested independently outside of that function. These are now in
6532 separate 32-bit words, so there is plenty of room for expansion again.
6533 There is no change in the user interface or the logging behaviour.
6535 (b) When building, for example, log lines, the code previously used a
6536 macro that called string_cat() twice, in order to add two strings. This is
6537 not really sufficiently general. Furthermore, there was one instance where
6538 it was actually wrong because one of the argument was used twice, and in
6539 one call a function was used. (As it happened, calling the function twice
6540 did not affect the overall behaviour.) The macro has been replaced by a
6541 function that can join an arbitrary number of extra strings onto a growing
6544 (c) The code for expansion conditions now uses a table and a binary chop
6545 instead of a serial search (which was left over from when there were very
6546 few conditions). Also, it now recognizes conditions like "pam" even when
6547 the relevant support is not compiled in: a suitably worded error message is
6548 given if an attempt is made to use such a condition.
6550 11. Added ${time_interval:xxxxx}.
6552 12. A bug was causing one of the ddress fields not to be passed back correctly
6553 from remote delivery subprocesses. The field in question was not being
6554 subsequently used, so this caused to problems in practice.
6556 13. Added new log selectors queue_time and deliver_time.
6558 14. Might have fixed a bug in maildirsizefile handling that threw up
6559 "unexpected character" debug warnings, and recalculated the data
6560 unnecessarily. In any case, I expanded the warning message to give more
6563 15. Added the message "Restricted characters in address" to the statements in
6564 the default ACL that block characters like @ and % in local parts.
6566 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
6567 Three changes have been made:
6569 (a) There was a serious bug; a negative response to MAIL caused the whole
6570 recipient domain to be cached as invalid, thereby blocking all messages
6571 to all local parts at the same domain, from all senders. This bug has
6572 been fixed. The domain is no longer cached after a negative response to
6573 MAIL if the sender used is not empty.
6575 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
6578 (c) A new callout option, "use_sender" has been added for people who want
6579 the modified behaviour.
6585 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
6588 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
6589 indeed breaks things for older releases.
6591 3. Added additional logging to the case where there is a problem reading data
6592 from a filter that is running in a subprocess using a pipe, in order to
6593 try to track down a specific problem.
6595 4. Testing facility fudge: when running in the test harness and attempting
6596 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
6597 getting "No route to host". Convert this to a timeout.
6599 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
6602 6. Some OS don't have socklen_t but use size_t instead. This affects the
6603 fifth argument of getsockopt() amongst other things. This is now
6604 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
6605 can be set for individual OS. I have set it for SunOS5, OSF1, and
6606 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
6607 some earlier ones do not.
6609 7. Change 4.30/15 was not doing the test caselessly.
6611 8. The standard form for an IPv6 address literal was being rejected by address
6612 parsing in, for example, MAIL and RCPT commands. An example of this kind of
6613 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6614 this, as well as the form without the "IPv6" on the front (but only when
6615 address literals are enabled, of course).
6617 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
6619 10. Exim crashed if a message with an empty sender address specified by -f
6620 encountered a router with an errors_to setting. This could be provoked only
6621 by a command such as
6625 where an empty string was supplied; "<>" did not hit this bug.
6627 11. Installed PCRE release 4.5.
6629 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
6630 remained set. It is now erased.
6632 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
6633 times from message ids (which are base 36 rather than the normal 62).
6635 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
6636 were being counted as actual protocol errors, and logged if the log
6637 selector +smtp_protocol_error was set. One cannot be perfect in this test,
6638 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
6639 and DATA following a set of rejected RCPTs do not count as protocol errors.
6640 In other words, Exim assumes they were pipelined, though this may not
6641 actually be the case. Of course, in all cases the client gets an
6642 appropriate error code.
6644 15. If a lookup fails in an ACL condition, a message about the failure may
6645 be available; it is used if testing the ACL cannot continue, because most
6646 such messages specify what the cause of the deferral is. However, some
6647 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
6648 that caused an old message to be retained and used if a later statement
6649 caused a defer, replacing the real cause of the deferral.
6651 16. If an IP address had so many PTR records that the DNS lookup buffer
6652 was not large enough to hold them, Exim could crash while trying to process
6653 the truncated data. It now detects and logs this case.
6655 17. Further to 4.21/58, another change has been made: if (and only if) the
6656 first line of a message (the first header line) ends with CRLF, a bare LF
6657 in a subsequent header line has a space inserted after it, so as not to
6658 terminate the header.
6660 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
6661 unnecessarily, used atoi() instead of strtol(), and didn't check the
6662 termination when getting file sizes from file names by regex.
6664 19. Completely re-implemented the support for maildirsize files, in the light
6665 of a number of problems with the previous contributed implementation
6666 (4.30/29). In particular:
6668 . If the quota is zero, the maildirsize file is maintained, but no quota is
6671 . If the maildir directory does not exist, it is created before any attempt
6672 to write a maildirsize file.
6674 . The quota value in the file is just a cache; if the quota is changed in
6675 the transport, the new value overrides.
6677 . A regular expression is available for excluding directories from the
6680 20. The autoreply transport checks the characters in options that define the
6681 message's headers; it allows continued headers, but it was checking with
6682 isspace() after an embedded newline instead of explicitly looking for a
6685 21. If all the "regular" hosts to which an address was routed had passed their
6686 expiry times, and had not reached their retry times, the address was
6687 bounced, even if fallback hosts were defined. Now Exim should go on to try
6690 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
6691 equivalent code in the SMTP transport. Some hosts send humungous responses
6692 to HELO/EHLO, more than 1024 it seems.
6694 23. Refactoring: code in filter.c used (void *) for "any old type" but this
6695 gives compiler warnings in some environments. I've now done it "properly",
6698 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
6699 (because of problems with the built-in one) was declared to return uschar *
6700 instead of char *, causing compiler failure.
6702 25. Fixed a file descriptor leak when processing alias/forward files.
6704 26. Fixed a minor format string issue in dbfn.c.
6706 27. Typo in exim.c: ("dmbnz" for "dbmnz").
6708 28. If a filter file refered to $h_xxx or $message_headers, and the headers
6709 contained RFC 2047 "words", Exim's memory could, under certain conditions,
6712 29. When a sender address is verified, it is cached, to save repeating the test
6713 when there is more than one recipient in a message. However, when the
6714 verification involves a callout, it is possible for different callout
6715 options to be set for different recipients. It is too complicated to keep
6716 track of this in the cache, so now Exim always runs a verification when a
6717 callout is required, relying on the callout cache for the optimization.
6718 The overhead is duplication of the address routing, but this should not be
6721 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
6722 to be verified with callout=postmaster, and the main callout worked but the
6723 postmaster check failed, the verification correctly failed. However, if a
6724 subsequent RCPT command asked for sender verification *without* the
6725 postmaster check, incorrect caching caused this verification also to fail,
6728 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
6729 it was not caching the DNS options (qualify_single, search_parents) that
6730 were used when the lookup failed. A subsequent lookup with different
6731 options therefore always gave the same answer, though there were cases
6732 where it should not have. (Example: a "domains = !$mx_any" option on a
6733 dnslookup router: the "domains" option is always processed without any
6734 widening, but the router might have qualify_single set.) Now Exim uses the
6735 cached value only when the same options are set.
6737 32. Added John Jetmore's "exipick" utility to the distribution.
6739 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
6740 than a timeout (e.g. a certificate is required, and is not provided), an
6741 Exim server now closes the connection immediately. Previously it waited for
6742 the client to close - but if the client is SSL, it seems that they each
6743 wait for each other, leading to a delay before one of them times out.
6745 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
6746 maintained 0.8.x compatibility because I don't think many are using it, and
6747 it is clearly obsolete.
6749 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
6752 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
6753 client certificate was expired. A simple patch fixes this, though I don't
6754 understand the full logic of why the verify callback is called multiple
6757 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
6758 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
6759 which causes problems with some clients (such as the Certicom SSL Plus
6760 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
6761 disables the coutermeasure allowing Eudora to connect."
6763 38. Exim was not checking that a write() to a log file succeeded. This could
6764 lead to Bad Things if a log got too big, in particular if it hit a file
6765 size limit. Exim now panics and dies if it cannot write to a log file, just
6766 as it does if it cannot open a log file.
6768 39. Modified OS/Makefile-Linux so that it now contains
6770 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
6772 The two -D definitions ensure that Exim is compiled with large file
6773 support, which makes it possible to handle log files that are bigger than
6776 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
6777 instance) a domain was checked against a named list that involved a lookup,
6778 causing $domain_data to be set, then another domain was checked against the
6779 same list, then the first domain was re-checked, the value of $domain_data
6780 after the final check could be wrong. In particular, if the second check
6781 failed, it could be set empty. This bug probably also applied to
6784 41. The strip_trailing_dot option was not being applied to the address given
6785 with the -f command-line option.
6787 42. The code for reading a message's header from the spool was incrementing
6788 $received_count, but never initializing it. This meant that the value was
6789 incorrect (doubled) while delivering a message in the same process in which
6790 it was received. In the most common configuration of Exim, this never
6791 happens - a fresh exec is done - but it can happen when
6792 deliver_drop_privilege is set.
6794 43. When Exim logs an SMTP synchronization error - client data sent too soon -
6795 it now includes up to 150 characters of the unexpected data in the log
6798 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6799 and building data strings. The size of both of these buffers was 10 000
6800 bytes - far larger than anybody would *ever* want, thought I. Needless to
6801 say, somebody hit the limit. I have increased the maximum line length to
6802 20 000 and the maximum data length of concatenated lines to 100 000. I have
6803 also fixed two bugs, because there was no checking on these buffers. Tsk,
6804 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6805 buffer is too small.
6807 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6808 lsearch lookups. Now it does.
6810 46. When parsing a route_list item in a manualroute router, a fixed-length
6811 buffer was used for the list of hosts. I made this 1024 bytes long,
6812 thinking that nobody would ever have a list of hosts that long. Wrong.
6813 Somebody had a whole pile of complicated expansion conditions, and the
6814 string was silently truncated, leading to an expansion error. It turns out
6815 that it is easier to change to an unlimited length (owing to other changes
6816 that have happened since this code was originally written) than to build
6817 structure for giving a limitation error. The length of the item that
6818 expands into the list of hosts is now unlimited.
6820 47. The lsearch lookup could not handle data where the length of text line was
6821 more than 4095 characters. Such lines were truncated, leading to shortened
6822 data being returned. It should now handle lines of any length.
6824 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6825 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6828 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6829 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6830 no longer happens in comment lines. A list of replacements is now placed
6831 at the head of all of the source files, except those whose only change is
6832 to replace PERL_COMMAND in the very first #! line.
6834 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6835 messages on the queue, with a bottom-up merge sort, using code contributed
6836 by Michael Haardt. This should make operations like -bp somewhat faster on
6837 large queues. It won't affect queue runners, except when queue_run_in_order
6840 51. Installed eximstats 1.31 in the distribution.
6842 52. Added support for SRV lookups to the dnslookup router.
6844 53. If an ACL referred to $message_body or $message_body_end, the value was not
6845 reset for any messages that followed in the same SMTP session.
6847 54. The store-handling optimization for building very long strings was not
6848 differentiating between the different store pools. I don't think this
6849 actually made any difference in practice, but I've tidied it.
6851 55. While running the routers to verify a sender address, $sender_address
6852 was still set to the sender address. This is wrong, because when routing to
6853 send a bounce to the sender, it would be empty. Therefore, I have changed
6854 it so that, while verifying a sender address, $sender_address is set to <>.
6855 (There is no change to what happens when verifying a recipient address.)
6857 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6858 target A or AAAA records (if not already returned) without resetting the
6859 qualify_single or search_parents options of the DNS resolver. These are
6860 inappropriate in this case because the targets of MX and SRV records must
6861 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6862 target that, when qualified, matched something in the local domain. These
6863 two options are now turned off when doing these lookups.
6865 57. It seems that at least some releases of Reiserfs (which does not have the
6866 concept of a fixed number of inodes) returns zero and not -1 for the
6867 number of available inodes. This interacted badly with check_spool_inodes,
6868 which assumed that -1 was the "no such thing" setting. What I have done is
6869 to check that the total number of inodes is greater than zero before doing
6870 the test of how many are available.
6872 58. When a "warn" ACL statement has a log_message modifier, the message is
6873 remembered, and not repeated. This is to avoid a lot of repetition when a
6874 message has many recipients that cause the same warning to be written.
6875 However, Exim was preserving the list of already written lines for an
6876 entire SMTP session, which doesn't seem right. The memory is now reset if a
6877 new message is started.
6879 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6880 debugging output unless log_rewrite was also set.
6882 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6883 of (int)(handle) when we know that handle contains (void *)(-1).
6885 61. The Exim daemon panic-logs an error return when it closes the incoming
6886 connection. However "connection reset by peer" seems to be common, and
6887 isn't really an error worthy of noting specially, so that particular error
6890 62. When Exim is trying to find all the local interfaces, it used to panic and
6891 die if the ioctl to get the interface flags failed. However, it seems that
6892 on at least one OS (Solaris 9) it is possible to have an interface that is
6893 included in the list of interfaces, but for which you get a failure error
6894 for this call. This happens when the interface is not "plumbed" into a
6895 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6896 failure of the "get flags" call assumes that the interface is down.
6898 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6899 makes life easier for people who are doing arithmetic on fields extracted
6900 from dates, where you often get leading zeros that should not be
6901 interpreted as octal.
6903 64. Added qualify_domain to the redirect router, to override the global
6906 65. If a pathologically long header line contained very many addresses (the
6907 report of this problem mentioned 10 000) and each of them was rewritten,
6908 Exim could use up a very large amount of memory. (It kept on making new
6909 copies of the header line as it rewrote, and never released the old ones.)
6910 At the expense of a bit more processing, the header rewriting function has
6911 been changed so that it no longer eats memory in this way.
6913 66. The generation of the Received: header has been moved from the time that a
6914 message starts to be received, to the time that it finishes. The timestamp
6915 in the Received: header should now be very close to that of the <= log
6916 line. There are two side-effects of this change:
6918 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6919 logged header lines no longer include the local Received: line, because
6920 it has not yet been created. The same applies to a copy of the message
6921 that is returned to a non-SMTP sender when a message is rejected.
6923 (b) When a filter file is tested using -bf, no additional Received: header
6924 is added to the test message. After some thought, I decided that this
6927 This change does not affect the value of $received_for. It is still set
6928 after address rewriting, but before local_scan() is called.
6930 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6932 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6933 gave an unhelpful panic error message, and a defer error. I have managed to
6934 change this behaviour so that it now rejects any supplied certificate,
6935 which seems right, as the list of acceptable certificates is empty.
6937 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6938 gave an unhelpful defer error. I have not managed to make this reject any
6939 supplied certificates, but the error message it gives is "no certificate
6940 supplied", which is not helpful.
6942 70. exigrep's output now also includes lines that are not associated with any
6943 message, but which match the given pattern. Implemented by a patch from
6944 Martin Sluka, which also tidied up the Perl a bit.
6946 71. Recipient callout verification, like sender verification, was using <> in
6947 the MAIL FROM command. This isn't really the right thing, since the actual
6948 sender may affect whether the remote host accepts the recipient or not. I
6949 have changed it to use the actual sender in the callout; this means that
6950 the cache record is now keyed on a recipient/sender pair, not just the
6951 recipient address. There doesn't seem to be a real danger of callout loops,
6952 since a callout by the remote host to check the sender would use <>.
6953 [SEE ABOVE: changed after hitting problems.]
6955 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6956 temporary errors. However, in the case of such a code being given after
6957 the end of a data transmission (i.e. after ".") Exim was failing to write
6958 a retry record for the message. (Yes, there was some broken host that was
6959 actually sending 8xx at this point.)
6961 73. An unknown lookup type in a host list could cause Exim to panic-die when
6962 the list was checked. (An example that provoked this was putting <; in the
6963 middle of a list instead of at the start.) If this happened during a DATA
6964 ACL check, a -D file could be left lying around. This kind of configuration
6965 error no longer causes Exim to die; instead it causes a defer error. The
6966 incident is still logged to the main and panic logs.
6968 74. Buglet left over from Exim 3 conversion. The message "too many messages
6969 in one connection" was written to the rejectlog but not the mainlog, except
6970 when address rewriting (yes!) was being logged.
6972 75. Added write_rejectlog option.
6974 76. When a system filter was run not as root (that is, when system_filter_user
6975 was set), the values of the $n variables were not being returned to the
6976 main process; thus, they were not subsequently available in the $sn
6979 77. Added +return_path_on_delivery log selector.
6981 78. A connection timeout was being treated differently from recipients deferred
6982 when testing hosts_max_try with a message that was older than the host's
6983 retry timeout. (The host should not be counted, thus allowing all hosts to
6984 be tried at least once before bouncing.) This may have been the cause of an
6985 occasionally reported bug whereby a message would remain on the queue
6986 longer than the retry timeout, but would be bounced if a delivery was
6987 forced. I say "may" because I never totally pinned down the problem;
6988 setting up timeout/retry tests is difficult. See also the next item.
6990 79. The ultimate address timeout was not being applied to errors that involved
6991 a combination of host plus message (for example, a timeout on a MAIL
6992 command). When an address resolved to a number of possible hosts, and they
6993 were not all tried for each delivery (e.g. because of hosts_max_try), a
6994 message could remain on the queue longer than the retry timeout.
6996 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6999 81. Fixed an obscure SMTP outgoing bug which required at least the following
7000 conditions: (a) there was another message waiting for the same server;
7001 (b) the server returned 5xx to all RCPT commands in the first message so
7002 that the message was not completed; (c) the server dropped the connection
7003 or gave a negative response to the RSET that Exim sends to abort the
7004 transaction. The observed case was a dropped connection after DATA that had
7005 been sent in pipelining mode. That is, the server had advertised PIPELINING
7006 but was not implementing it correctly. The effect of the bug was incorrect
7007 behaviour, such as trying another host, and this could lead to a crash.
7013 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
7014 and daemon.c were passed as pointers to ints; they should have been
7015 pointers to socklen_t variables (which are typically unsigned ints).
7017 2. Some signed/unsigned type warnings in the os.c file for Linux have been
7020 3. Fixed a really odd bug that affected only the testing scheme; patching a
7021 certain fixed string in the binary changed the value of another string that
7022 happened to be identical to the end of the original first string.
7024 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
7025 name", it returns that address as the IP address. On some operating
7026 systems (e.g. Solaris), it also passes back the IP address string as the
7027 "host name". However, on others (e.g. Linux), it passes back an empty
7028 string. Exim wasn't checking for this, and was changing the host name to an
7029 empty string, assuming it had been canonicalized.
7031 5. Although rare, it is permitted to have more than one PTR record for a given
7032 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
7033 all the names associated with an address, because they do in Solaris.
7034 However, it seems that they do not in Linux for data that comes from the
7035 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
7036 I found this out when I moved to a new Linux workstation and tried to run
7037 the Exim test suite.
7039 To get round this problem I have changed the code so that it now does its
7040 own call to the DNS to look up PTR records when searching for a host name.
7041 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
7042 addresses that are only in /etc/hosts are still found.
7044 This behaviour is, however, controlled by an option called host_lookup_
7045 order, which defaults to "bydns:byaddr". If people want to use the other
7046 order, or indeed, just use one or the other means of lookup, they can
7047 specify it in this variable.
7049 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
7050 some operating systems, this comes back from gethostbyaddr() as an empty
7051 string, and this is what Exim used to test for. However, it seems that in
7052 other systems, "." is yielded. Exim now tests for this case too.
7054 7. The values of check_spool_space and check_log_space are now held internally
7055 as a number of kilobytes instead of an absolute number of bytes. If a
7056 numbers is specified without 'K' or 'M', it is rounded up to the nearest
7057 kilobyte. This means that much larger values can be stored.
7059 8. Exim monitor: an attempt to get the action menu when not actually pointing
7060 at a message produces an empty menu entitled "No message selected". This
7061 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
7062 no entries in it ("Shell widget menu has zero width and/or height"). So I
7063 have added a single, blank menu entry in this case.
7065 9. Added ${quote_local_part.
7067 10. MIME decoding is now applied to the contents of Subject: header lines when
7070 11. Now that a reference to $sender_host_address automatically causes a reverse
7071 lookup to occur if necessary (4.13/18), there is no need to arrange for a
7072 host lookup before query-style lookups in lists that might use this
7073 variable. This has therefore been abolished, and the "net-" prefix is no
7074 longer necessary for query-style lookups.
7076 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
7077 have been a typo for LFLAGS, so it has been changed.
7079 13. The install script calls Exim with "-C /dev/null" in order to find the
7080 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
7081 to be output. However, since Exim outputs its version number before the
7082 error, it didn't break the script. It just looked ugly. I fixed this by
7083 always allowing "-C /dev/null" if the caller is root.
7085 14. Ignore overlarge ACL variable number when reading spool file - insurance
7086 against a later release with more variables having written the file.
7088 15. The standard form for an IPv6 address literal was being rejected by EHLO.
7089 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
7090 this, as well as the form without the "IPv6" on the front.
7092 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
7093 OS/Makefile-Darwin file.
7095 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
7096 only to LDAP libraries that do not have LDAP_OPT_DEREF.
7098 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
7099 variable. However, some OS use ints for inodes. I've added cast to long int
7100 to get rid of the compiler warning.
7102 19. I had forgotten to lock out "/../" in configuration file names when
7103 ALT_CONFIG_PREFIX was set.
7105 20. Routers used for verification do not need to specify transports. However,
7106 if such a router generated a host list, and callout was configured, Exim
7107 crashed, because it could not find a port number from the (non-existent)
7108 transport. It now assumes port 25 in this circumstance.
7110 21. Added the -t option to exigrep.
7112 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
7113 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
7114 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
7115 them is set, LOOKUP_LSEARCH is forced.
7117 23. "exim -bV" now outputs a list of lookups that are included in the binary.
7119 24. Added sender and host information to the "rejected by local_scan()" log
7120 line; previously there was no indication of these.
7122 25. Added .include_if_exists.
7124 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
7125 Linux. It turns out that not all file systems support this. Apparently some
7126 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
7127 do it.) To cope with this, the error EINVAL, which means that sync-ing is
7128 not supported on the file descriptor, is now ignored when Exim is trying to
7129 sync a directory. This applies only to Linux.
7131 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
7133 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
7134 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
7137 29. Added support for maildirsize files from supplied patch (modified a bit).
7139 30. The use of :fail: followed by an empty string could lead Exim to respond to
7140 sender verification failures with (e.g.):
7142 550 Verification failed for <xxx>
7143 550 Sender verify failed
7145 where the first response line was missing the '-' that indicates it is not
7146 the final line of the response.
7148 31. The loop for finding the name of the user that called Exim had a hardwired
7149 limit of 10; it now uses the value of finduser_retries, which is used for
7150 all other user lookups.
7152 32. Added $received_count variable, available in data and not_smtp ACLs, and at
7155 33. Exim was neglecting to zero errno before one call of strtol() when
7156 expanding a string and expecting an integer value. On some systems this
7157 resulted in spurious "integer overflow" errors. Also, it was casting the
7158 result into an int without checking.
7160 34. Testing for a connection timeout using "timeout_connect" in the retry rules
7161 did not work. The code looks as if it has *never* worked, though it appears
7162 to have been documented since at least release 1.62. I have made it work.
7164 35. The "timeout_DNS" error in retry rules, also documented since at least
7165 1.62, also never worked. As it isn't clear exactly what this means, and
7166 clearly it isn't a major issue, I have abolished the feature by treating it
7167 as "timeout", and writing a warning to the main and panic logs.
7169 36. The display of retry rules for -brt wasn't always showing the error code
7172 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
7173 timeout_connect_A, timeout_connect_MX.
7175 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
7176 to the empty sender.
7178 39. The daemon was not analysing the content of -oX till after it had closed
7179 stderr and disconnected from the controlling terminal. This meant that any
7180 syntax errors were only noted on the panic log, and the return code from
7181 the command was 0. By re-arranging the code a little, I've made the
7182 decoding happen first, so such errors now appear on stderr, and the return
7183 code is 1. However, the actual setting up of the sockets still happens in
7184 the disconnected process, so errors there are still only recorded on the
7187 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
7188 connections (as happens on some IP stacks) was logged at start up time as
7189 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
7190 it from "IPv6 and IPv4", which means that two separate sockets are being
7193 41. The debug output for gethostbyname2() or getipnodebyname() failures now
7194 says whether AF_INET or AF_INET6 was passed as an argument.
7196 42. Exiwhat output was messed up when time zones were included in log
7199 43. Exiwhat now gives more information about the daemon's listening ports,
7200 and whether -tls-on-connect was used.
7202 44. The "port" option of the smtp transport is now expanded.
7204 45. A "message" modifier in a "warn" statement in a non-message ACL was being
7205 silently ignored. Now an error message is written to the main and panic
7208 46. There's a new ACL modifier called "logwrite" which writes to a log file
7209 as soon as it is encountered.
7211 47. Added $local_user_uid and $local_user_gid at routing time.
7213 48. Exim crashed when trying to verify a sender address that was being
7216 49. Exim was recognizing only a space character after ".include". It now also
7217 recognizes a tab character.
7219 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
7220 extracting the relevant information from the specification. The man page no
7221 longer contains scrambled data for the -d option, and I've added a section
7222 at the front about calling Exim under different names.
7224 51. Added "extra_headers" argument to the "mail" command in filter files.
7226 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
7229 53. Installed eximstats 1.29.
7231 54. Added transport_filter_timeout as a generic transport option.
7233 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
7234 Cc: header lines. This was required by RFC 822, but it not required by RFC
7237 56. Exim used to add From:, Date:, and Message-Id: header lines to any
7238 incoming messages that did not have them. Now it does so only if the
7239 message originates locally, that is, if there is no associated remote host
7240 address. When Resent- header lines are present, this applies to the Resent-
7241 lines rather than the non-Resent- lines.
7243 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
7244 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
7246 58. Messages for configuration errors now include the name of the main
7247 configuration file - useful now that there may be more than one file in a
7248 list (.included file names were always shown).
7250 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
7251 for those rare installations that do not start the daemon as root or run it
7252 setuid root. I've cut out the call to initgroups() if the daemon is not
7255 60. The Exim user and group can now be bound into the binary as text strings
7256 that are looked up at the start of Exim's processing.
7258 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
7260 62. Added $mailstore_basename variable.
7262 63. Installed patch to sieve.c from Michael Haardt.
7264 64. When Exim failed to open the panic log after failing to open the main log,
7265 the original message it was trying to log was written to stderr and debug
7266 output, but if they were not available (the usual case in production), it
7267 was lost. Now it is written to syslog before the two lines that record the
7268 failures to open the logs.
7270 65. Users' Exim filters run in subprocesses under the user's uid. It is
7271 possible for a "deliver" command or an alias in a "personal" command to
7272 provoke an address rewrite. If logging of address rewriting is configured,
7273 this fails because the process is not running as root or exim. There may be
7274 a better way of dealing with this, but for the moment (because 4.30 needs
7275 to be released), I have disabled address rewrite logging when running a
7276 filter in a non-root, non-exim process.
7282 1. The buildconfig auxiliary program wasn't quoting the value set for
7283 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
7284 not defined. This bug was masked in 4.22 by the effect that was fixed in
7287 2. Some messages that were rejected after a message id was allocated were
7288 shown as "incomplete" by exigrep. It no longer does this for messages that
7289 are rejected by local_scan() or the DATA or non-SMTP ACLs.
7291 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
7292 have allow_domain_literals set, the ID did not get logged in the <= line.
7293 Domain literals are now always recognized in Message-ID: header lines.
7295 4. The first argument for a ${extract expansion item is the key name or field
7296 number. Leading and trailing spaces in this item were not being ignored,
7297 causing some misleading effects.
7299 5. When deliver_drop_privilege was set, single queue runner processes started
7300 manually (i.e. by the command "exim -q") or by the daemon (which uses the
7301 same command in the process it spins off) were not dropping privilege.
7303 6. When the daemon running as "exim" started a queue runner, it always
7304 re-executed Exim in the spun-off process. This is a waste of effort when
7305 deliver_drop_privilege is set. The new process now just calls the
7306 queue-runner function directly.
7312 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
7315 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
7316 ignored. Though the use of -oP was forcing the writing of a pid file, it
7317 was always written to the default place.
7319 3. If the message "no IP address found for host xxxx" is generated during
7320 incoming verification, it is now followed by identification of the incoming
7321 connection (so you can more easily find what provoked it).
7323 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
7325 5. Added some features to "harden" Exim a bit more against certain attacks:
7327 (a) There is now a build-time option called FIXED_NEVER_USERS that can
7328 be put in Local/Makefile. This is like the never_users runtime option,
7329 but it cannot be overridden. The default setting is "root".
7331 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
7332 prefix string with which any file named in a -C command line option
7335 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
7336 is retained for -C and -D only if the caller of Exim is root. Without
7337 it, the exim user may also use -C and -D and retain privilege.
7339 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
7340 command line option is disabled.
7342 6. Macro names set by the -D option must start with an upper case letter, just
7343 like macro names defined in the configuration file.
7345 7. Added "dereference=" facility to LDAP.
7347 8. Two instances of the typo "uknown" in the source files are fixed.
7349 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
7350 the Configure-Makefile script screwed up while processing it.
7352 10. Incorporated PCRE 4.4.
7354 11. The SMTP synchronization check was not operating right at the start of an
7355 SMTP session. For example, it could not catch a HELO sent before the client
7356 waited for the greeting. There is now a check for outstanding input at the
7357 point when the greeting is written. Because of the duplex, asynchronous
7358 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
7359 way, but not yet received, when the check is performed.
7361 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
7362 on TCP/IP sockets, because this apparently causes some broken clients to
7365 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
7366 unchanged) from the Cygwin maintainer.
7368 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
7369 was supported instead of testing for mbx. Effectively a typo.
7371 15. The spa authenticator server code was not checking that the input it
7372 received was valid base64.
7374 16. The debug output line for the "set" modifier in ACLs was not showing the
7375 name of the variable that was being set.
7377 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
7379 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
7380 lookup on its own. Something else was needed to trigger the lookup. For
7381 example, a match in host_lookup or the need for a host name in a host list.
7382 Now, if $sender_host_name is referenced and the host name has not yet been
7383 looked up, a lookup is performed. If the lookup fails, the variable remains
7384 empty, and $host_lookup_failed is set to "1".
7386 19. Added "eqi" as a case-independent comparison operator.
7388 20. The saslauthd authentication condition could segfault if neither service
7389 nor realm was specified.
7391 21. If an overflowing value such as "2048M" was set for message_size_limit, the
7392 error message that was logged was misleading, and incoming SMTP
7393 connections were dropped. The message is now more accurate, and temporary
7394 errors are given to SMTP connections.
7396 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
7397 (except RSET) with a 421 error, until QUIT is received. However, it was
7398 failing to send a response to QUIT.
7400 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
7401 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
7402 ACL is now run after the helo_try_verify_hosts code.
7404 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
7405 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
7406 case-independent, so other case variants are also recognized). Apparently
7407 some systems use these upper case variants.
7409 25. If more than two messages were waiting for the same host, and a transport
7410 filter was specified for the transport, Exim sent two messages over the
7411 same TCP/IP connection, and then failed with "socket operation on non-
7412 socket" when it tried to send the third.
7414 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
7416 27. The extern definition of crypt16() in expand.c was not being excluded when
7417 the OS had its own crypt16() function.
7419 28. Added bounce_return_body as a new option, and bounce_return_size_limit
7420 as a preferred synonym for return_size_limit, both as an option and as an
7423 29. Added LIBS=-liconv to OS/Makefile-OSF1.
7425 30. Changed the default configuration ACL to relax the local part checking rule
7426 for addresses that are not in any local domains. For these addresses,
7427 slashes and pipe symbols are allowed within local parts, but the sequence
7428 /../ is explicitly forbidden.
7430 31. SPA server authentication was not clearing the challenge buffer before
7433 32. log_message in a "warn" ACL statement was writing to the reject log as
7434 well as to the main log, which contradicts the documentation and doesn't
7435 seem right (because no rejection is happening). So I have stopped it.
7437 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
7438 However, I am unable to do any testing of this.
7440 34. Fixed an infelicity in the appendfile transport. When checking directories
7441 for a mailbox, to see if any needed to be created, it was accidentally
7442 using path names with one or more superfluous leading slashes; tracing
7443 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
7445 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
7446 contents are added to the log line that is written for every discarded
7447 recipient. (Previously a log_message setting was ignored.)
7449 36. The ${quote: operator now quotes the string if it is empty.
7451 37. The install script runs exim in order to find its version number. If for
7452 some reason other than non-existence or emptiness, which it checks, it
7453 could not run './exim', it was installing it with an empty version number,
7454 i.e. as "exim-". This error state is now caught, and the installation is
7457 38. An argument was missing from the function that creates an error message
7458 when Exim fails to connect to the socket for saslauthd authentication.
7459 This could cause Exim to crash, or give a corrupted message.
7461 39. Added isip, isip4, and isip6 to ${if conditions.
7463 40. The ACL variables $acl_xx are now saved with the message, and can be
7464 accessed later in routers, transports, and filters.
7466 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
7467 strings in the file are not string-expanded.
7469 42. If a MAIL command specified a SIZE value that was too large to fit into an
7470 int variable, the check against message_size_limit failed. Such values are
7471 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
7472 day this will have to be increased, but I don't think I want to be around
7473 when emails are that large.
7480 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
7481 iconv() is not standard in FreeBSD.
7483 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
7484 IPv6 enabled. The observed symptom was a segmentation fault on return from
7485 the function os_common_find_running_interfaces() in src/os.c.
7487 3. In the check_special_case() function in daemon.c I had used "errno" as an
7488 argument name, which causes warnings on some systems. This was basically a
7489 typo, since it was named "eno" in the comments!
7491 4. The code that waits for the clock to tick (at a resolution of some fraction
7492 of a second) so as to ensure message-id uniqueness was always waiting for
7493 at least one whole tick, when it could have waited for less. [This is
7494 almost certainly not relevant at current processor speeds, where it is
7495 unlikely to ever wait at all. But we try to future-proof.]
7497 5. The function that sleeps for a time interval that includes fractions of a
7498 second contained a race. It did not block SIGALRM between setting the
7499 timer, and suspending (a couple of lines later). If the interval was short
7500 and the sigsuspend() was delayed until after it had expired, the suspension
7501 never ended. On busy systems this could lead to processes getting stuck for
7504 6. Some uncommon configurations may cause a lookup to happen in a queue runner
7505 process, before it forks any delivery processes. The open lookup caching
7506 mechanism meant that the open file or database connection was passed into
7507 the delivery process. The problem was that delivery processes always tidy
7508 up cached lookup data. This could cause a problem for the next delivery
7509 process started by the queue runner, because the external queue runner
7510 process does not know about the closure. So the next delivery process
7511 still has data in the lookup cache. In the case of a file lookup, there was
7512 no problem because closing a file descriptor in a subprocess doesn't affect
7513 the parent. However, if the lookup was caching a connection to a database,
7514 the connection was closed, and the second delivery process was likely to
7515 see errors such as "PGSQL: query failed: server closed the connection
7516 unexpectedly". The problem has been fixed by closing all cached lookups
7517 in a queue runner before running a delivery process.
7519 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
7520 seem to have the "const" qualifier which it has on other OS. I've
7523 8. Change 4.21/2 was too strict. It is only if there are two authenticators
7524 *of the same type* (client or server) with the same public name that an
7525 error should be diagnosed.
7527 9. When Exim looked up a host name for an IP address, but failed to find the
7528 original IP address when looking up the host name (a safety check), it
7529 output the message "<ip address> does not match any IP for NULL", which was
7530 confusing, to say the least. The bug was that the host name should have
7531 appeared instead of "NULL".
7533 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
7534 user that is built into the binary, and the -C or -D options is used, root
7535 privilege is dropped before the configuration file is read. In addition,
7536 logging is switched to stderr instead of the normal log files. If the
7537 configuration then re-defines the Exim user, the unprivileged environment
7538 is probably not what is expected, so Exim logs a panic warning message (but
7541 However, if deliver_drop_privilege is set, the unprivileged state may well
7542 be exactly what is intended, so the warning has been cut out in that case,
7543 and Exim is allowed to try to write to its normal log files.
7549 1. smtp_return_error_details was not giving details for temporary sender
7550 or receiver verification errors.
7552 2. Diagnose a configuration error if two authenticators have the same public
7555 3. Exim used not to create the message log file for a message until the first
7556 delivery attempt. This could be confusing when incoming messages were held
7557 for policy or load reasons. The message log file is now created at the time
7558 the message is received, and an initial "Received" line is written to it.
7560 4. The automatically generated man page for command line options had a minor
7561 bug that caused no ill effects; however, a more serious problem was that
7562 the procedure for building the man page automatically didn't always
7563 operate. Consequently, release 4.20 contains an out-of-date version. This
7564 shouldn't happen again.
7566 5. When building Exim with embedded Perl support, the script that builds the
7567 Makefile was calling 'perl' to find its compile-time parameters, ignoring
7568 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
7570 6. The freeze_tell option was not being used for messages that were frozen on
7571 arrival, either by an ACL or by local_scan().
7573 7. Added the smtp_incomplete_transaction log selector.
7575 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
7576 was accepting AUTH without a new EHLO.
7578 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
7579 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
7582 10. Logging of TCP/IP connections (when configured) now happens in the main
7583 daemon process instead of the child process, so that the TCP/IP connection
7584 count is more accurate (but it can never be perfect).
7586 11. The use of "drop" in a nested ACL was not being handled correctly in the
7587 outer ACL. Now, if condition failure induced by the nested "drop" causes
7588 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
7589 or "require"), the connection is dropped.
7591 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
7592 that yield "discard" can now be used with an "accept" or a "discard" verb,
7593 but an error is generated for any others (because I can't see a useful way
7594 to define what should happen).
7596 13. When an ACL is read dynamically from a file (or anywhere else), the lines
7597 are now processed in the same way as lines in the Exim configuration file.
7598 In particular, continuation lines are supported.
7600 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
7602 15. Added -ti meaning -t -i.
7604 16. Check for letters, digits, hyphens, and dots in the names of dnslist
7605 domains, and warn by logging if others are found.
7607 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
7608 returned from GIFCONF when Exim is trying to find the list of interfaces on
7609 a host. The code in os.c has been modified to copy each ifreq to an aligned
7610 structure in all cases.
7612 Also, in some cases, the returned ifreq's were being copied to a 'struct
7613 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
7614 means the last couple of bytes of an IPv6 address could be chopped if the
7615 ifreq contained only a normal sockaddr (14 bytes storage).
7617 18. Named domain lists were not supported in the hosts_treat_as_local option.
7618 An entry such as +xxxx was not recognized, and was treated as a literal
7621 19. Ensure that header lines added by a DATA ACL are included in the reject log
7622 if the ACL subsequently rejects the message.
7624 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
7625 MD5 (which is deprecated).
7627 21. When testing a filter file using -bf, Exim was writing a message when it
7628 took the sender from a "From " line in the message, but it was not doing so
7629 when it took $return_path from a Return-Path: header line. It now does.
7631 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
7632 with a valid header line field name (a series of printing characters
7633 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
7635 23. Changed "disc" in the source to "disk" to conform to the documentation and
7636 the book and for uniformity.
7638 24. Ignore Sendmail's -Ooption=value command line item.
7640 25. When execve() failed while trying to run a command in a pipe transport,
7641 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
7642 could be confused with a return value of 69 from the command itself. This
7643 has been changed to 127, the value the shell returns if it is asked to run
7644 a non-existent command. The wording for the related log line suggests a
7645 non-existent command as the problem.
7647 26. If received_header_text expands to an empty string, do not add a Received:
7648 header line to the message. (Well, it adds a token one on the spool, but
7649 marks it "old" so that it doesn't get used or transmitted.)
7651 27. Installed eximstats 1.28 (addition of -nt option).
7653 28. There was no check for failure on the call to getsockname() in the daemon
7654 code. This can fail if there is a shortage of resources on the system, with
7655 ENOMEM, for example. A temporary error is now given on failure.
7657 29. Contrary to the C standard, it seems that in some environments, the
7658 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
7659 program. Exim now does this explicitly; it affects the formatting of
7660 timestamps using strftime().
7662 30. If exiqsumm was given junk data, it threw up some uninitialized variable
7663 complaints. I've now initialized all the variables, to avoid this.
7665 32. Header lines added by a system filter were not being "seen" during
7666 transport-time rewrites.
7668 33. The info_callback() function passed to OpenSSL is set up with type void
7669 (*)(SSL *, int, int), as described somewhere. However, when calling the
7670 function (actually a macro) that sets it up, the type void(*)() is
7671 expected. I've put in a cast to prevent warnings from picky compilers.
7673 34. If a DNS black list lookup found a CNAME record, but there were no A
7674 records associated with the domain it pointed at, Exim crashed.
7676 35. If a DNS black list lookup returned more than one A record, Exim ignored
7677 all but the first. It now scans all returned addresses if a particular IP
7678 value is being sought. In this situation, the contents of the
7679 $dnslist_value variable are a list of all the addresses, separated by a
7682 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
7683 to accept a host name and all its aliases if the forward lookup for any of
7684 them yielded the IP address of the incoming connection. Now it accepts only
7685 those names whose forward lookup yields the correct IP address. Any other
7686 names are discarded. This closes a loophole whereby a rogue DNS
7687 administrator could create reverse DNS records to break through a
7688 wildcarded host restriction in an ACL.
7690 37. If a user filter or a system filter that ran in a subprocess used any of
7691 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
7692 the wrong values were passed to the pipe command ($thisaddress had the
7693 value of $0, $0 had the value of $1, etc). This bug was introduced by
7694 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
7696 38. Improved the line breaking for long SMTP error messages from ACLs.
7697 Previously, if there was no break point between 40 and 75 characters, Exim
7698 left the rest of the message alone. Two changes have been made: (a) I've
7699 reduced the minimum length to 35 characters; (b) if it can't find a break
7700 point between 35 and 75 characters, it looks ahead and uses the first one
7701 that it finds. This may give the occasional overlong line, but at least the
7702 remaining text gets split now.
7704 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
7705 file descriptors might be low, and that setting 1000 would always raise it.
7706 It turns out that in some environments, the limit is already over 1000 and
7707 that lowering it causes trouble. So now Exim takes care not to decrease it.
7709 40. When delivering a message, the value of $return_path is set to $sender_
7710 address at the start of routing (routers may change the value). By an
7711 oversight, this default was not being set up when an address was tested by
7712 -bt or -bv, which affected the outcome if any router or filter referred to
7715 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
7716 log only when "message" or "log_message" are set. However, if one of the
7717 conditions was an address verification, or a call to a nested ACL, the
7718 messages generated by the underlying test were being passed through. This
7719 no longer happens. The underlying message is available in $acl_verify_
7720 message for both "message" and "log_message" expansions, so it can be
7721 passed through if needed.
7723 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
7724 new expansion $bh_ to give the encoded byte string without charset
7725 translation. Translation happens only if iconv() is available; HAVE_ICONV
7726 indicates this at build time. HEADERS_CHARSET gives the charset to
7727 translate to; headers_charset can change it in the configuration, and
7728 "headers charset" can change it in an individual filter file.
7730 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
7731 that creates RFC 2047 encoded "words" labels them as that charset instead
7732 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
7733 expansion operator; (ii) when Exim creates a From: line for a local
7734 message; (iii) when a header line is rewritten to include a "phrase" part.
7736 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
7737 buggy, causing it to skip the first lines of messages whose message ID
7738 ended in 'D'. This would not have bitten before Exim release 4.14, because
7739 message IDs were unlikely to end in 'D' before then. The effect was to have
7740 incorrect size information for certain domains.
7742 45. #include "config.h" was missing at the start of the crypt16.c module. This
7743 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
7746 46. If there was a timeout during a "random" callout check, Exim treated it as
7747 a failure of the random address, and carried on sending RSET and the real
7748 address. If the delay was just some slowness somewhere, the response to the
7749 original RCPT would be taken as a response to RSET and so on, causing
7750 mayhem of various kinds.
7752 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
7753 when I implemented it. It didn't allow for the fact that some option values
7754 may legitimately be negative (e.g. size_addition), and it didn't even do
7755 the right test for positive values.
7757 48. Domain names in DNS records are case-independent. Exim always looks them up
7758 in lower case. Some resolvers return domain names in exactly the case they
7759 appear in the zone file, that is, they may contain uppercase letters. Not
7760 all resolvers do this - some return always lower case. Exim was treating a
7761 change of case by a resolver as a change of domain, similar to a widening
7762 of a domain abbreviation. This triggered its re-routing code and so it was
7763 trying to route what was effectively the same domain again. This normally
7764 caused routing to fail (because the router wouldn't handle the domain
7765 twice). Now Exim checks for this case specially, and just changes the
7766 casing of the domain that it ultimately uses when it transmits the message
7769 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
7772 50. If a filter generated a file delivery with a non-absolute name (possible if
7773 no home directory exists for the router), the forbid_file option was not
7776 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
7777 the existing equality matching.
7779 52. Exim was using ints instead of ino_t variables in some places where it was
7780 dealing with inode numbers.
7782 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
7783 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
7784 called TMPDIR, and if it finds it is different, it changes its value.
7786 54. The smtp_printf() function is now made available to local_scan() so
7787 additional output lines can be written before returning. There is also an
7788 smtp_fflush() function to enable the detection of a dropped connection.
7789 The variables smtp_input and smtp_batched_input are exported to
7792 55. Changed the default runtime configuration: the message "Unknown user"
7793 has been removed from the ACL, and instead placed on the localuser router,
7794 using the cannot_route_message feature. This means that any verification
7795 failures that generate their own messages won't get overridden. Similarly,
7796 the "Unrouteable address" message that was in the ACL for unverifiable
7797 relay addresses has also been removed.
7799 56. Added hosts_avoid_esmtp to the smtp transport.
7801 57. The exicyclog script was not checking for the esoteric option
7802 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7803 will work only if exicyclog is run under the appropriate euid.
7805 58. Following a discussion on the list, the rules by which Exim recognises line
7806 endings on incoming messages have been changed. The -dropcr and drop_cr
7807 options are now no-ops, retained only for backwards compatibility. The
7808 following line terminators are recognized: LF CRLF CR. However, special
7809 processing applies to CR:
7811 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7812 nor a local message in the state where . is a terminator.
7814 (ii) If a bare CR is encountered in a header line, an extra space is added
7815 after the line terminator so as not to end the header. The reasoning
7816 behind this is that bare CRs in header lines are most likely either
7817 to be mistakes, or people trying to play silly games.
7819 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7820 was being incorrectly given as 18 bytes larger than it should have been.
7821 This is a VOB (very old bug).
7823 60. This may never have affected anything current, but just in case it has:
7824 When the local host is found other than at the start of a list of hosts,
7825 the local host, those with the same MX, and any that follow, are discarded.
7826 When the list in question was part of a longer list of hosts, the following
7827 hosts (not currently being processed) were also being discarded. This no
7828 longer happens. I'm not sure if this situation could ever has previously
7831 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7834 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7835 header. This is something that is recommended in a new Internet Draft, and
7836 is something that is documented as being done by Sendmail. There are two
7837 possible values. For messages generated by the autoreply transport, Exim
7840 Auto-Submitted: auto-replied
7842 whereas for all other generated messages (e.g. bounces) it adds
7844 Auto-Submitted: auto-generated
7846 63. The "personal" condition in filters now includes a test for the
7847 Auto-Submitted: header. If it contains the string "auto-" the message it
7848 not considered personal.
7850 64. Added rcpt_include_affixes as a generic transport option.
7852 65. Added queue_only_override (default true).
7854 66. Added the syslog_duplication option.
7856 67. If what should have been the first header line of a message consisted of
7857 a space followed by a colon, Exim was mis-interpreting it as a header line.
7858 It isn't of course - it is syntactically invalid and should therefore be
7859 treated as the start of the message body. The misbehaviour could have
7860 caused a number of strange effects, including loss of data in subsequent
7861 header lines, and spool format errors.
7863 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7864 client host had authenticated. This control can now be exercised by an ACL
7865 for more flexibility.
7867 69. By default, callouts do not happen when testing with -bh. There is now a
7868 variant, -bhc, which does actually run the callout code, including
7869 consulting and updating the callout cache.
7871 70. Added support for saslauthd authentication, courtesy of Alexander
7874 71. If statvfs() failed on the spool or log directories while checking their
7875 size for availability, Exim confusingly gave the error "space shortage".
7876 Furthermore, in debugging mode it crashed with a floating point exception.
7877 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7878 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7879 serious problem, Exim now writes to the main and panic logs when this
7880 happens, with details of the failure. It then refuses to accept the
7881 incoming message, giving the message "spool directory problem" or "log
7882 directory problem" with a 421 code for SMTP messages.
7884 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7885 0, 1, and 2 exist, because some OS complain for execs without them (see
7886 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7887 descriptors. However, the code omitted to check that the open succeeded,
7888 causing mysterious errors if for some reason the permissions on /dev/null
7889 got screwed. Now Exim writes a message to the main and panic logs, and
7890 bombs out if it can't open /dev/null.
7892 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7893 interact so that it is all more flexible. It is supposed to remain
7894 backwards compatible. Also added extra_local_interfaces.
7896 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7897 to bomb out with an assertion failure - to the client this appears as a
7898 connection drop. This problem occurs in the part of the code that was taken
7899 from the Samba project. Fortunately, the assertion is in a very simple
7900 function, so I have fixed this by reproducing the function inline in the
7901 one place where it is called, and arranging for authentication to fail
7902 instead of killing the process with assert().
7904 75. The SPA client code was not working when the server requested OEM rather
7905 than Unicode encoding.
7907 76. Added code to make require_files with a specific uid setting more usable in
7908 the case where statting the file as root fails - usually a non-root-mounted
7909 NFS file system. When this happens and the failure is EACCES, Exim now
7910 forks a subprocess and does the per-uid checking as the relevant uid.
7912 77. Added process_log_path.
7914 78. If log_file_path was not explicitly set, a setting of check_log_space or
7915 check_log_inodes was ignored.
7917 79. If a space check for the spool or log partitions fails, the incident is now
7918 logged. Of course, in the latter case the data may get lost...
7920 80. Added the %p formatting code to string_format() so that it can be used to
7921 print addresses in debug_print(). Adjusted all the address printing in the
7922 debugging in store.c to use %p rather than %d.
7924 81. There was a concern that a line of code in smtp_in.c could overflow a
7925 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7926 initially expressed, the concern was not well-founded, because trailing
7927 spaces are removed early. However, if the trailing spaces were followed by
7928 a NULL, they did not get removed, so the overflow was possible. Two fixes
7931 (a) I re-wrote the offending code in a cleaner fashion.
7932 (b) If an incoming SMTP command contains a NULL character, it is rejected
7935 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7936 runs initgroups(), so that if the Exim user is in any additional groups,
7937 they will be used during message reception.
7943 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob