1 This document describes *changes* to previous versions, that might
2 affect Exim's operation, with an unchanged configuration file. For new
3 options, and new features, see the NewStuff file next to this ChangeLog.
6 Since Exim version 4.94
7 -----------------------
9 JH/02 Bug 2587: Fix pam expansion condition. Tainted values are commonly used
10 as arguments, so an implementation trying to copy these into a local
11 buffer was taking a taint-enforcement trap. Fix by using dynamically
12 created buffers. Similar fix for radius expansion condition.
14 JH/03 Bug 2586: Fix listcount expansion operator. Using tainted arguments is
15 reasonable, eg. to count headers. Fix by using dynamically created
16 buffers rather than a local. Do similar fixes for ACL actions "dcc",
17 "log_reject_target", "malware" and "spam"; the arguments are expanded
18 so could be handling tainted values.
20 JH/04 Bug 2590: Fix -bi (newaliases). A previous code rearrangement had
21 broken the (no-op) support for this sendmail command. Restore it
22 to doing nothing, silently, and returning good status.
24 JH/05 Bug 2593: Fix "vacation" in Exim filter. Previously, when a "once"
25 record path was given (or the default used) without a leading directory
26 path, an error occurred on trying to open it. Use the transport's working
29 JH/06 Bug 2594: Change the name used for certificate name checks in the smtp
30 transport. Previously it was the name on the DNS A-record; use instead
31 the head of the CNAME chain leading there (if there is one). This seems
32 to align better with RFC 6125.
34 JH/07 Bug 2597: Fix a resource leak. Using a lookup in obtaining a value for
35 smtp_accept_max_per_host allocated resources which were not released
36 when the limit was exceeded. This eventually crashed the daemon. Fix
37 by adding a relase action in that path.
39 JH/08 Bug 2598: Fix verify ACL condition. The options for the condition are
40 expanded; previously using tainted values was rejected. Fix by using
41 dynamically-created buffers.
43 JH/10 Bug 2603: Fix coding of string copying to only evaluate arguments once.
44 Previously a macro used one argument twice; when called with the
45 argument as an expression having side-effects, incorrect operation
46 resulted. Use an inlineable function.
48 JH/11 Bug 2604: Fix request to cutthrough-deliver when a connection is already
49 held open for a verify callout. Previously this wan not accounted for
50 and a corrupt onward SMTP conversation resulted.
52 JH/13 Fix dsearch "subdir" filter to ignore ".". Previously only ".." was
53 excluded, not matching the documentation.
55 JH/14 Bug 2606: Fix a segfault in sqlite lookups. When no, or a bad, filename
56 was given for the sqlite_dbfile a trap resulted.
58 JH/15 Bug 2620: Fix "spam" ACL condition. Previously, tainted values for the
59 "name" argument resulted in a trap. There is no reason to disallow such;
60 this was a coding error.
62 JH/16 Bug 2615: Fix pause during message reception, on systems that have been
63 suspended/resumed. The Linux CLOCK_MONOTONIC does not account for time
64 spent suspended, ignoring the Posix definition. Previously we assumed
65 it did and a constant offset from real time could be used as a correction.
66 Change to using the same clock source for the start-of-message and the
67 post-message next-tick-wait. Also change to using CLOCK_BOOTTIME if it
68 exists, just to get a clock slightly more aligned to reality.
70 JH/17 Bug 2295: Fix DKIM signing to always semicolon-terminate. Although the
71 RFC says it is optional some validators care. The missing char was not
72 intended but triggered by a line-wrap alignement. Discovery and fix by
73 Guillaume Outters, hacked on by JH.
75 JH/18 Bug 2617: Fix a taint trap in parse_fix_phrase(). Previously when the
76 name being quoted was tainted a trap would be taken. Fix by using
77 dynamicaly created buffers. The routine could have been called by a
78 rewrite with the "h" flag, by using the "-F" command-line option, or
79 by using a "name=" option on a control=submission ACL modifier.
81 JH/21 Bug 2630: Fix eol-replacement string for the ${readsocket } expansion.
82 Previously when a whitespace character was specified it was not inserted
83 after removing the newline.
85 JH/24 Bug 2634: Fix a taint trap seen on NetBSD: the testing coded for
86 is_tainted() had an off-by-one error in the overenthusiastic direction.
87 Find and fix by Gavan. Although NetBSD is not a supported platform for
88 4.94 this bug could affect other platforms.
90 JH/24 Bug 2634: Fix a taint trap seen on NetBSD: the testing coded for
91 is_tainted() had an off-by-one error in the overenthusiastic direction.
92 Find and fix by Gavan. Although NetBSD is not a supported platform for
93 4.94 this bug could affect other platforms.
94 JH/21 Bug 2630: Fix eol-replacement string for the ${readsocket } expansion.
95 Previously when a whitespace character was specified it was not inserted
96 after removing the newline.
98 JH/22 Bug 2265: Force SNI usage for smtp transport DANE'd connections, to be
99 the domain part of the recipient address. This overrides any tls_sni
100 option set, which was previously used.
102 JH/23 Logging: with the +tls_sni log_selector, do not wrap the received SNI
105 JH/26 Bug 2646: fix a memory usage issue in ldap lookups. Previously, when more
106 than one server was defined and depending on the platform memory layout
107 details, an internal consistency trap could be hit while walking the list
110 JH/27 Bug 2648: fix the passing of an authenticator public-name through spool
111 files. The value is used by the authresults expansion item. Previously
112 if this was used in a router or transport, a crash could result.
114 JH/30 Bug 2677: fix matching of long addresses. Since 4.93 a limit of 256 was
115 applied. This resulted, if any header-line rewrite rules were configured,
116 in a panic-log trigerrable by sending a message with a long address in
117 a header. Fix by increaing the arbitrary limit to larger than a single
118 (dewrapped) 5322 header line maximum size.
120 JH/31 The ESMTP option name advertised for the SUPPORT_EARLY_PIPE build option
121 is changed from X_PIPE_CONNECT to PIPE_CONNECT. This is in line with
122 RFC 6648 which deprecates X- options in protocols as a general practice.
123 Changeover between the implementations is handled by the mechanisms
126 JH/32 Bug 2599: fix delay of delivery to a local address where there is also
127 a remote which uses callout/hold. Previously the local was queued.
129 JH/33 Fix a taint trap in the ${listextract } expansion when the source data
132 JH/35 Bug 2343: Harden exim_tidydb against corrupt wait- files.
134 JH/36 Bug 2687: Fix interpretation of multiple ^ chars in a plaintext
135 authenticator client_send option. Previously the next char, after a pair
136 was collapsed, was taken verbatim (so ^^^foo became ^^foo; ^^^^foo became
137 ^^\x00foo). Fixed to get ^\x00foo and ^^foo respectively to match the
138 documentation. There is still no way to get a leading ^ immediately
139 after a NUL (ie. for the password of a PLAIN method authenticator.
141 JH/39 Bug 2691: fix $local_part_data. When the matching list element
142 referred to a file, bad data was returned. This likely also affected
145 JH/41 Fix daemon SIGHUP on FreeBSD. Previously, a named socket for IPC was
146 left undeleted; the attempt to re-create it then failed - resulting in
147 the usual "SIGHUP tp have daemon reload configuration" to not work.
148 This affected any platform not supporting "abstract" Unix-domain
149 sockets (i.e. not Linux).
151 JH/42 Bug 2692: Harden against a peer which reneges on a 452 "too many
152 recipients" response to RCPT in a later response, with a 250. The
153 previous coding assumed this would not happen, and under PIPELINING
154 would result in both lost and duplicate recipients for a message.
156 JH/43 Bug 2694: Fix weighted distribution of work to multiple spamd servers.
157 Previously the weighting was incorrectly applied. Similar fix for socks
158 proxies. Found and fixed by Heiko Schlichting.
160 JH/44 Bug 2701: Fix list-expansion of dns_ipv4_lookup. Previously, it did
161 not handle sub-lists included using the +namedlist syntax. While
162 investigating, the same found for dns_trust_aa, dns_again_means_nonexist,
163 dnssec_require_domains, dnssec_request_domains, srv_fail_domains,
166 HS/01 Enforce absolute PID file path name.
168 HS/02 Handle SIGINT as we handle SIGTERM: terminate the Exim process.
170 PP/01 Add a too-many-bad-recipients guard to the default config's RCPT ACL.
172 PP/02 Bug 2643: Correct TLS DH constants.
173 A missing NUL termination in our code-generation tool had led to some
174 incorrect Diffie-Hellman constants in the Exim source.
175 Reported by kylon94, code-gen tool fix by Simon Arlott.
177 PP/03 Impose security length checks on various command-line options.
178 Fixes CVE-2020-SPRSS reported by Qualys.
180 PP/04 Fix Linux security issue CVE-2020-SLCWD and guard against PATH_MAX
181 better. Reported by Qualys.
183 PP/05 Fix security issue CVE-2020-PFPSN and guard against cmdline invoker
184 providing a particularly obnoxious sender full name.
187 PP/06 Fix CVE-2020-28016 (PFPZA): Heap out-of-bounds write in parse_fix_phrase()
189 PP/07 Refuse to allocate too little memory, block negative/zero allocations.
192 PP/08 Change default for recipients_max from unlimited to 50,000.
194 PP/09 Fix security issue with too many recipients on a message (to remove a
195 known security problem if someone does set recipients_max to unlimited,
196 or if local additions add to the recipient list).
197 Fixes CVE-2020-RCPTL reported by Qualys.
199 PP/10 Fix security issue in SMTP verb option parsing
200 Fixes CVE-2020-EXOPT reported by Qualys.
206 JH/01 Avoid costly startup code when not strictly needed. This reduces time
207 for some exim process initialisations. It does mean that the logging
208 of TLS configuration problems is only done for the daemon startup.
210 JH/02 Early-pipelining support code is now included unless disabled in Makefile.
212 JH/03 DKIM verification defaults no long accept sha1 hashes, to conform to
213 RFC 8301. They can still be enabled, using the dkim_verify_hashes main
216 JH/04 Support CHUNKING from an smtp transport using a transport_filter, when
217 DKIM signing is being done. Previously a transport_filter would always
218 disable CHUNKING, falling back to traditional DATA.
220 JH/05 Regard command-line receipients as tainted.
222 JH/06 Bug 340: Remove the daemon pid file on exit, whe due to SIGTERM.
224 JH/07 Bug 2489: Fix crash in the "pam" expansion condition. It seems that the
225 PAM library frees one of the arguments given to it, despite the
226 documentation. Therefore a plain malloc must be used.
228 JH/08 Bug 2491: Use tainted buffers for the transport smtp context. Previously
229 on-stack buffers were used, resulting in a taint trap when DSN information
230 copied from a received message was written into the buffer.
232 JH/09 Bug 2493: Harden ARC verify against Outlook, whick has been seen to mix
233 the ordering of its ARC headers. This caused a crash.
235 JH/10 Bug 2492: Use tainted memory for retry record when needed. Previously when
236 a new record was being constructed with information from the peer, a trap
239 JH/11 Bug 2494: Unset the default for dmarc_tld_file. Previously a naiive
240 installation would get error messages from DMARC verify, when it hit the
241 nonexistent file indicated by the default. Distros wanting DMARC enabled
242 should both provide the file and set the option.
243 Also enforce no DMARC verification for command-line sourced messages.
245 JH/12 Fix an uninitialised flag in early-pipelining. Previously connections
246 could, depending on the platform, hang at the STARTTLS response.
248 JH/13 Bug 2498: Reset a counter used for ARC verify before handling another
249 message on a connection. Previously if one message had ARC headers and
250 the following one did not, a crash could result when adding an
251 Authentication-Results: header.
253 JH/14 Bug 2500: Rewind some of the common-coding in string handling between the
254 Exim main code and Exim-related utities. The introduction of taint
255 tracking also did many adjustments to string handling. Since then, eximon
256 frequently terminated with an assert failure.
258 JH/15 When PIPELINING, synch after every hundred or so RCPT commands sent and
259 check for 452 responses. This slightly helps the inefficieny of doing
260 a large alias-expansion into a recipient-limited target. The max_rcpt
261 transport option still applies (and at the current default, will override
262 the new feature). The check is done for either cause of synch, and forces
263 a fast-retry of all 452'd recipients using a new MAIL FROM on the same
264 connection. The new facility is not tunable at this time.
266 JH/16 Fix the variables set by the gsasl authenticator. Previously a pointer to
267 library live data was being used, so the results became garbage. Make
268 copies while it is still usable.
270 JH/17 Logging: when the deliver_time selector ise set, include the DT= field
271 on delivery deferred (==) and failed (**) lines (if a delivery was
272 attemtped). Previously it was only on completion (=>) lines.
274 JH/18 Authentication: the gsasl driver not provides the $authN variables in time
275 for the expansion of the server_scram_iter and server_scram_salt options.
277 WB/01 SPF: DNS lookups for the obsolete SPF RR type done by the libspf2 library
278 are now specifically given a NO_DATA response without hitting the system
279 resolver. The library goes on to do the now-standard TXT lookup.
280 Use of dnsdb lookups is not affected.
282 JH/19 Bug 2507: Modules: on handling a dynamic-module (lookups) open failure,
283 only retrieve the errormessage once. Previously two calls to dlerror()
284 were used, and the second one (for mainlog/paniclog) retrieved null
287 JH/20 Taint checking: disallow use of tainted data for
288 - the appendfile transport file and directory options
289 - the pipe transport command
290 - the autoreply transport file, log and once options
291 - file names used by the redirect router (including filter files)
293 - paths used by single-key lookups
294 Previously this was permitted.
296 JH/21 Bug 2501: Fix init call in the heimdal authenticator. Previously it
297 adjusted the size of a major service buffer; this failed because the
298 buffer was in use at the time. Change to a compile-time increase in the
299 buffer size, when this authenticator is compiled into exim.
301 JH/22 Taint-checking: move to safe-mode taint checking on all platforms. The
302 previous fast-mode was untenable in the face of glibs using mmap to
303 support larger malloc requests.
305 PP/01 Update the openssl_options possible values through OpenSSL 1.1.1c.
306 New values supported, if defined on system where compiled:
307 allow_no_dhe_kex, cryptopro_tlsext_bug, enable_middlebox_compat,
308 no_anti_replay, no_encrypt_then_mac, prioritize_chacha, tlsext_padding
310 JH/23 Performance improvement in the initial phase of a two-pass queue run. By
311 running a limited number of proceses in parallel, a benefit is gained. The
312 amount varies with the platform hardware and load. The use of the option
313 queue_run_in_order means we cannot do this, as ordering becomes
316 JH/24 Bug 2524: fix the cyrus_sasl auth driver gssapi usage. A previous fix
317 had introduced a string-copy (for ensuring NUL-termination) which was not
318 appropriate for that case, which can include embedded NUL bytes in the
319 block of data. Investigation showed the copy to actually be needless, the
320 data being length-specified.
322 JH/25 Fix use of concurrent TLS connections under GnuTLS. When a callout was
323 done during a receiving connection, and both used TLS, global info was
324 used rather than per-connection info for tracking the state of data
325 queued for transmission. This could result in a connection hang.
327 JH/26 Fix use of the SIZE parameter on MAIL commands, on continued connections.
328 Previously, when delivering serveral messages down a single connection
329 only the first would provide a SIZE. This was due to the size information
330 not being properly tracked.
332 JH/27 Bug 2530: When operating in a timezone with sub-minute offset, such as
333 TAI (at 37 seconds currently), pretend to be in UTC for time-related
334 expansion and logging. Previously, spurious values such as a future
335 minute could be seen.
337 JH/28 Bug 2533: Fix expansion of ${tr } item. When called in some situations
338 it could crash from a null-deref. This could also affect the
339 ${addresses: } operator and ${readsock } item.
341 JH/29 Bug 2537: Fix $mime_part_count. When a single connection had a non-mime
342 message following a mime one, the variable was not reset.
344 JH/30 When an pipelined-connect fails at the first response, assume incorrect
345 cached capability (perhaps the peer reneged?) and immediately retry in
348 JH/31 Fix spurious detection of timeout while writing to transport filter.
350 JH/32 Bug 2541: Fix segfault on bad cmdline -f (sender) argument. Previously
351 an attempt to copy the string was made before checking it.
353 JH/33 Fix the dsearch lookup to return an untainted result. Previously the
354 taint of the lookup key was maintained; we now regard the presence in the
355 filesystem as sufficient validation.
357 JH/34 Fix the readsocket expansion to not segfault when an empty "options"
358 argument is supplied.
360 JH/35 The dsearch lookup now requires that the directory is an absolute path.
361 Previously this was not checked, and nonempty relative paths made an
362 access under Exim's current working directory.
364 JH/36 Bug 2554: Fix msg:defer event for the hosts_max_try_hardlimit case.
365 Previously no event was raised.
367 JH/37 Bug 2552: Fix the check on spool space during reception to use the SIZE
368 parameter supplied by the sender MAIL FROM command. Previously it was
369 ignored, and only the check_spool_space option value for the required
372 JH/38 Fix $dkim_key_length. This should, after a DKIM verification, present
373 the size of the signing public-key. Previously it was instead giving
374 the size of the signature hash.
376 JH/39 DKIM verification: the RFC 8301 restriction on sizes of RSA keys is now
377 the default. See the (new) dkim_verify_min_keysizes option.
379 JH/40 Fix a memory-handling bug: when a connection carried multiple messages
380 and an ACL use a lookup for checking either the local_part or domain,
381 stale data could be accessed. Ensure that variable references are
382 dropped between messages.
384 JH/41 Bug 2571: Fix SPA authenticator. Running as a server, an offset supplied
385 by the client was not checked as pointing within response data before
386 being used. A malicious client could thus cause an out-of-bounds read and
387 possibly gain authentication. Fix by adding the check.
389 JH/42 Internationalisation: change the default for downconversion in the smtp
390 transport to be "if needed". Previously it was "as previously set" for
391 the message, which usually meant "if needed" for message-submission but
392 "no" for everything else. However, MTAs have been seen using SMTPUTF8
393 even when the envelope addresses did not need it, resulting in forwarding
394 failures to non-supporting MTAs. A downconvert in such cases will be
395 a no-op on the addresses, merely dropping the use of SMTPUTF8 by the
396 transport. The change does mean that addresses needing conversion will
397 be converted when previously a delivery failure would occur.
399 JH/43 Fix possible long line in DSN. Previously when a very long SMTP error
400 response was received it would be used unchecked in a fail-DSN, violating
401 standards on line-length limits. Truncate if needed.
403 HS/01 Remove parameters of the link to www.open-spf.org. The linked form
404 doesn't work. (Additionally add a new main config option to configure the
411 JH/01 OpenSSL: With debug enabled output keying information sufficient, server
412 side, to decode a TLS 1.3 packet capture.
414 JH/02 OpenSSL: Suppress the sending of (stateful) TLS1.3 session tickets.
415 Previously the default library behaviour applied, sending two, each in
418 JH/03 Debug output for ACL now gives the config file name and line number for
421 JH/04 The default received_header_text now uses the RFC 8314 tls cipher clause.
423 JH/05 DKIM: ensure that dkim_domain elements are lowercased before use.
425 JH/06 Fix buggy handling of autoreply bounce_return_size_limit, and a possible
426 buffer overrun for (non-chunking) other transports.
428 JH/07 GnuTLS: Our use of late (post-handshake) certificate verification, under
429 TLS1.3, means that a server rejecting a client certificate is not visible
430 to the client until the first read of encrypted data (typically the
431 response to EHLO). Add detection for that case and treat it as a failed
432 TLS connection attempt, so that the normal retry-in-clear can work (if
433 suitably configured).
435 JB/01 Bug 2375: fix expansions of 822 addresses having comments in local-part
436 and/or domain. Found and fixed by Jason Betts.
438 JH/08 Add hardening against SRV & TLSA lookups the hit CNAMEs (a nonvalid
439 configuration). If a CNAME target was not a wellformed name pattern, a
442 JH/09 Logging: Fix initial listening-on line for multiple ports for an IP when
443 the OS reports them interleaved with other addresses.
445 JH/10 OpenSSL: Fix aggregation of messages. Previously, when PIPELINING was
446 used both for input and for a verify callout, both encrypted, SMTP
447 responses being sent by the server could be lost. This resulted in
448 dropped connections and sometimes bounces generated by a peer sending
451 JH/11 Harden plaintext authenticator against a badly misconfigured client-send
452 string. Previously it was possible to cause undefined behaviour in a
453 library routine (usually a crash). Found by "zerons".
455 JH/12 Bug 2384: fix "-bP smtp_receive_timeout". Previously it returned no
458 JH/13 Bug 2386: Fix builds with Dane under LibreSSL 2.9.0 onward. Some old
459 API was removed, so update to use the newer ones.
461 JH/14 Bug 1891: Close the log file if receiving a non-smtp message, without
462 any timeout set, is taking a long time. Previously we would hang on to a
463 rotated logfile "forever" if the input was arriving with long gaps
464 (a previous attempt to fix addressed lack, for a long time, of initial
467 HS/01 Bug 2390: Use message_id for tempfile creation to avoid races in a
468 shared (NFS) environment. The length of the tempfile name is now
469 4 + 16 ("hdr.$message_exim_id") which might break on file
470 systems which restrict the file name length to lower values.
473 HS/02 Bug 2390: Use message_id for tempfile creation to avoid races in a
474 shared (NFS) environment.
476 HS/03 Bug 2392: exigrep does case sensitive *option* processing (as it
477 did for all versions <4.90). Notably -M, -m, --invert, -I may be
480 JH/15 Use unsigned when creating bitmasks in macros, to avoid build errors
481 on some platforms for bit 31.
483 JH/16 GnuTLS: rework ciphersuite strings under recent library versions. Thanks
484 to changes apparently associated with TLS1.3 handling some of the APIs
485 previously used were either nonfunctional or inappropriate. Strings
486 like TLS1.3:ECDHE_SECP256R1__RSA_PSS_RSAE_SHA256__AES_256_GCM__AEAD:256
487 and TLS1.2:ECDHE_SECP256R1__RSA_SHA256__AES_128_CBC__SHA256:128 replace
488 the previous TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256 .
489 This affects log line X= elements, the $tls_{in,out}_cipher variables,
490 and the use of specific cipher names in the encrypted= ACL condition.
492 JH/17 OpenSSL: the default openssl_options now disables ssl_v3.
494 JH/18 GnuTLS: fix $tls_out_ocsp under hosts_request_ocsp. Previously the
495 verification result was not updated unless hosts_require_ocsp applied.
497 JH/19 Bug 2398: fix listing of a named-queue. Previously, even with the option
498 queue_list_requires_admin set to false, non-admin users were denied the
501 JH/20 Bug 2389: fix server advertising of usable certificates, under GnuTLS in
502 directory-of-certs mode. Previously they were advertised despite the
505 JH/21 The smtp transport option "hosts_noproxy_tls" is now unset by default.
506 A single TCP connection by a client will now hold a TLS connection open
507 for multiple message deliveries, by default. Previously the default was to
510 JH/22 The smtp transport option "hosts_try_dane" now enables all hosts by
511 default. If built with the facility, DANE will be used. The facility
512 SUPPORT_DANE is now enabled in the prototype build Makefile "EDITME".
514 JH/23 The build default is now for TLS to be included; the SUPPORT_TLS define
515 is replaced with DISABLE_TLS. Either USE_GNUTLS or (the new) USE_OPENSSL
516 must be defined and you must still, unless you define DISABLE_TLS, manage
517 the the include-dir and library-file requirements that go with that
518 choice. Non-TLS builds are still supported.
520 JH/24 Fix duplicated logging of peer name/address, on a transport connection-
523 JH/25 The smtp transport option "hosts_try_fastopen" now enables all hosts by
524 default. If the platform supports and has the facility enabled, it will
525 be requested on all coneections.
527 JH/26 The PIPE_CONNECT facility is promoted from experimental status and is now
528 controlled by the build-time option SUPPORT_PIPE_CONNECT.
530 PP/01 Unbreak heimdal_gssapi, broken in 4.92.
532 JH/27 Bug 2404: Use the main-section configuration option "dsn_from" for
533 success-DSN messages. Previously the From: header was always the default
534 one for these; the option was ignored.
536 JH/28 Fix the timeout on smtp response to apply to the whole response.
537 Previously it was reset for every read, so a teergrubing peer sending
538 single bytes within the time limit could extend the connection for a
539 long time. Credit to Qualsys Security Advisory Team for the discovery.
541 JH/29 Fix DSN Final-Recipient: field. Previously it was the post-routing
542 delivery address, which leaked information of the results of local
543 forwarding. Change to the original envelope recipient address, per
546 JH/30 Bug 2411: Fix DSN generation when RFC 3461 failure notification is
547 requested. Previously not bounce was generated and a log entry of
548 error ignored was made.
550 JH/31 Avoid re-expansion in ${sort } expansion. (CVE-2019-13917)
552 JH/32 Introduce a general tainting mechanism for values read from the input
553 channel, and values derived from them. Refuse to expand any tainted
554 values, to catch one form of exploit.
556 JH/33 Bug 2413: Fix dkim_strict option. Previously the expansion result
557 was unused and the unexpanded text used for the test. Found and
558 fixed by Ruben Jenster.
560 JH/34 Fix crash after TLS shutdown. When the TCP/SMTP channel was left open,
561 an attempt to use a TLS library read routine dereffed a nul pointer,
564 JH/35 Bug 2409: filter out-of-spec chars from callout response before using
565 them in our smtp response.
567 JH/36 Have the general router option retry_use_local_part default to true when
568 any of the restrictive preconditions are set (to anything). Previously it
569 was only for check_local user. The change removes one item of manual
570 configuration which is required for proper retries when a remote router
571 handles a subset of addresses for a domain.
573 JH/37 Appendfile: when evaluating quota use (non-quota_size_regex) take the file
574 link count into consideration.
576 HS/04 Fix handling of very log lines in -H files. If a -<key> <value> line
577 caused the extension of big_buffer, the following lines were ignored.
579 JH/38 Bug 1395: Teach the DNS negative-cache about TTL value from the SOA in
580 accordance with RFC 2308. Previously there was no expiry, so a longlived
581 receive process (eg. due to ACL delays) versus a short SOA value could
584 HS/05 Handle trailing backslash gracefully. (CVE-2019-15846)
586 JH/39 Promote DMARC support to mainline.
588 JH/40 Bug 2452: Add a References: header to DSNs.
590 JH/41 With GnuTLS 3.6.0 (and later) do not attempt to manage Diffie-Hellman
591 parameters. The relevant library call is documented as "Deprecated: This
592 function is unnecessary and discouraged on GnuTLS 3.6.0 or later. Since
593 3.6.0, DH parameters are negotiated following RFC7919."
595 HS/06 Change the default of dnssec_request_domains to "*"
597 JH/42 Bug 2545: Fix CHUNKING for all RCPT commands rejected. Previously we
598 carried on and emitted a BDAT command, even when PIPELINING was not
601 JH/43 Bug 2465: Fix taint-handling in dsearch lookup. Previously a nontainted
602 buffer was used for the filename, resulting in a trap when tainted
603 arguments (eg. $domain) were used.
605 JH/44 With OpenSSL 1.1.1 (onwards) disable renegotiation for TLS1.2 and below;
606 recommended to avoid a possible server-load attack. The feature can be
607 re-enabled via the openssl_options main cofiguration option.
609 JH/45 local_scan API: documented the current smtp_printf() call. This changed
610 for version 4.90 - adding a "more data" boolean to the arguments.
611 Bumped the ABI version number also, this having been missed previously;
612 release versions 4.90 to 4.92.3 inclusive were effectively broken in
613 respect of usage of smtp_printf() by either local_scan code or libraries
614 accessed via the ${dlfunc } expansion item. Both will need coding
615 adjustment for any calls to smtp_printf() to match the new function
616 signature; a FALSE value for the new argument is always safe.
618 JH/46 FreeBSD: fix use of the sendfile() syscall. The shim was not updating
619 the file-offset (which the Linux syscall does, and exim expects); this
620 resulted in an indefinite loop.
622 JH/47 ARC: fix crash in signing, triggered when a configuration error failed
623 to do ARC verification. The Authentication-Results: header line added
624 by the configuration then had no ARC item.
630 JH/01 Remove code calling the customisable local_scan function, unless a new
631 definition "HAVE_LOCAL_SCAN=yes" is present in the Local/Makefile.
633 JH/02 Bug 1007: Avoid doing logging from signal-handlers, as that can result in
634 non-signal-safe functions being used.
636 JH/03 Bug 2269: When presented with a received message having a stupidly large
637 number of DKIM-Signature headers, disable DKIM verification to avoid
638 a resource-consumption attack. The limit is set at twenty.
640 JH/04 Add variables $arc_domains, $arc_oldest_pass for ARC verify. Fix the
641 report of oldest_pass in ${authres } in consequence, and separate out
642 some descriptions of reasons for verification fail.
644 JH/05 Bug 2273: Cutthrough delivery left a window where the received messsage
645 files in the spool were present and unlocked. A queue-runner could spot
646 them, resulting in a duplicate delivery. Fix that by doing the unlock
647 after the unlink. Investigation by Tim Stewart. Take the opportunity to
648 add more error-checking on spoolfile handling while that code is being
651 PP/01 Refuse to open a spool data file (*-D) if it's a symlink.
652 No known attacks, no CVE, this is defensive hardening.
654 JH/06 Bug 2275: The MIME ACL unlocked the received message files early, and
655 a queue-runner could start a delivery while other operations were ongoing.
656 Cutthrough delivery was a common victim, resulting in duplicate delivery.
657 Found and investigated by Tim Stewart. Fix by using the open message data
658 file handle rather than opening another, and not locally closing it (which
659 releases a lock) for that case, while creating the temporary .eml format
660 file for the MIME ACL. Also applies to "regex" and "spam" ACL conditions.
662 JH/07 Bug 177: Make a random-recipient callout success visible in ACL, by setting
663 $sender_verify_failure/$recipient_verify_failure to "random".
665 JH/08 When generating a selfsigned cert, use serial number 1 since zero is not
668 JH/09 Bug 2274: Fix logging of cmdline args when starting in an unlinked cwd.
669 Previously this would segfault.
671 JH/10 Fix ARC signing for case when DKIM signing failed. Previously this would
674 JH/11 Bug 2264: Exim now only follows CNAME chains one step by default. We'd
675 like zero, since the resolver should be doing this for us, But we need one
676 as a CNAME but no MX presence gets the CNAME returned; we need to check
677 that doesn't point to an MX to declare it "no MX returned" rather than
678 "error, loop". A new main option is added so the older capability of
679 following some limited number of chain links is maintained.
681 JH/12 Add client-ip info to non-pass iprev ${authres } lines.
683 JH/13 For receent Openssl versions (1.1 onward) use modern generic protocol
684 methods. These should support TLS 1.3; they arrived with TLS 1.3 and the
685 now-deprecated earlier definitions used only specified the range up to TLS
686 1.2 (in the older-version library docs).
688 JH/14 Bug 2284: Fix DKIM signing for body lines starting with a pair of dots.
690 JH/15 Rework TLS client-side context management. Stop using a global, and
691 explicitly pass a context around. This enables future use of TLS for
692 connections to service-daemons (eg. malware scanning) while a client smtp
693 connection is using TLS; with cutthrough connections this is quite likely.
695 JH/16 Fix ARC verification to do AS checks in reverse order.
697 JH/17 Support a "tls" option on the ${readsocket } expansion item.
699 JH/18 Bug 2287: Fix the protocol name (eg utf8esmtp) for multiple messages
700 using the SMTPUTF8 option on their MAIL FROM commands, in one connection.
701 Previously the "utf8" would be re-prepended for every additional message.
703 JH/19 Reject MAIL FROM commands with SMTPUTF8 when the facility was not advertised.
704 Previously thery were accepted, resulting in issues when attempting to
705 forward messages to a non-supporting MTA.
707 PP/02 Let -n work with printing macros too, not just options.
709 JH/20 Bug 2296: Fix cutthrough for >1 address redirection. Previously only
710 one parent address was copied, and bogus data was used at delivery-logging
711 time. Either a crash (after delivery) or bogus log data could result.
712 Discovery and analysis by Tim Stewart.
714 PP/03 Make ${utf8clean:} expansion operator detect incomplete final character.
715 Previously if the string ended mid-character, we did not insert the
716 promised '?' replacement.
718 PP/04 Documentation: current string operators work on bytes, not codepoints.
720 JH/21 Change as many as possible of the global flags into one-bit bitfields; these
721 should pack well giving a smaller memory footprint so better caching and
722 therefore performance. Group the declarations where this can't be done so
723 that the byte-sized flag variables are not interspersed among pointer
724 variables, giving a better chance of good packing by the compiler.
726 JH/22 Bug 1896: Fix the envelope from for DMARC forensic reports to be possibly
727 non-null, to avoid issues with sites running BATV. Previously reports were
728 sent with an empty envelope sender so looked like bounces.
730 JH/23 Bug 2318: Fix the noerror command within filters. It wasn't working.
731 The ignore_error flag wasn't being returned from the filter subprocess so
732 was not set for later routers. Investigation and fix by Matthias Kurz.
734 JH/24 Bug 2310: Raise a msg:fail:internal event for each undelivered recipient,
735 and a msg:complete for the whole, when a message is manually removed using
736 -Mrm. Developement by Matthias Kurz, hacked on by JH.
738 JH/25 Avoid fixed-size buffers for pathnames in DB access. This required using
739 a "Gnu special" function, asprintf() in the DB utility binary builds; I
740 hope that is portable enough.
742 JH/26 Bug 2311: Fix DANE-TA verification under GnuTLS. Previously it was also
743 requiring a known-CA anchor certificate; make it now rely entirely on the
744 TLSA as an anchor. Checking the name on the leaf cert against the name
745 on the A-record for the host is still done for TA (but not for EE mode).
747 JH/27 Fix logging of proxy address. Previously, a pointless "PRX=[]:0" would be
748 included in delivery lines for non-proxied connections, when compiled with
749 SUPPORT_SOCKS and running with proxy logging enabled.
751 JH/28 Bug 2314: Fire msg:fail:delivery event even when error is being ignored.
752 Developement by Matthias Kurz, tweaked by JH. While in that bit of code,
753 move the existing event to fire before the normal logging of message
754 failure so that custom logging is bracketed by normal logging.
756 JH/29 Bug 2322: A "fail" command in a non-system filter (file) now fires the
757 msg:fail:internal event. Developement by Matthias Kurz.
759 JH/30 Bug 2329: Increase buffer size used for dns lookup from 2k, which was
760 far too small for todays use of crypto signatures stored there. Go all
761 the way to the max DNS message size of 64kB, even though this might be
762 overmuch for IOT constrained device use.
764 JH/31 Fix a bad use of a copy function, which could be used to pointlessly
765 copy a string over itself. The library routine is documented as not
766 supporting overlapping copies, and on MacOS it actually raised a SIGABRT.
768 JH/32 For main options check_spool_space and check_inode_space, where the
769 platform supports 64b integers, support more than the previous 2^31 kB
770 (i.e. more than 2 TB). Accept E, P and T multipliers in addition to
771 the previous G, M, k.
773 JH/33 Bug 2338: Fix the cyrus-sasl authenticator to fill in the
774 $authenticated_fail_id variable on authentication failure. Previously
777 JH/34 Increase RSA keysize of autogen selfsign cert from 1024 to 2048. RHEL 8.0
778 OpenSSL didn't want to use such a weak key. Do for GnuTLS also, and for
779 more-modern GnuTLS move from GNUTLS_SEC_PARAM_LOW to
780 GNUTLS_SEC_PARAM_MEDIUM.
782 JH/35 OpenSSL: fail the handshake when SNI processing hits a problem, server
783 side. Previously we would continue as if no SNI had been received.
785 JH/36 Harden the handling of string-lists. When a list consisted of a sole
786 "<" character, which should be a list-separator specification, we walked
787 off past the nul-terimation.
789 JH/37 Bug 2341: Send "message delayed" warning MDNs (restricted to external
790 causes) even when the retry time is not yet met. Previously they were
791 not, meaning that when (say) an account was over-quota and temp-rejecting,
792 and multiple senders' messages were queued, only one sender would get
793 notified on each configured delay_warning cycle.
795 JH/38 Bug 2351: Log failures to extract envelope addresses from message headers.
797 JH/39 OpenSSL: clear the error stack after an SSL_accept(). With anon-auth
798 cipher-suites, an error can be left on the stack even for a succeeding
799 accept; this results in impossible error messages when a later operation
802 AM/01 Bug 2359: GnuTLS: repeat lowlevel read and write operations while they
803 return error codes indicating retry. Under TLS1.3 this becomes required.
805 JH/40 Fix the feature-cache refresh for EXPERIMENTAL_PIPE_CONNECT. Previously
806 it only wrote the new authenticators, resulting in a lack of tracking of
807 peer changes of ESMTP extensions until the next cache flush.
809 JH/41 Fix the loop reading a message header line to check for integer overflow,
810 and more-often against header_maxsize. Previously a crafted message could
811 induce a crash of the recive process; now the message is cleanly rejected.
813 JH/42 Bug 2366: Fix the behaviour of the dkim_verify_signers option. It had
814 been totally disabled for all of 4.91. Discovery and fix by "Mad Alex".
820 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
821 When redis_servers is set to a list of > 1 element, and the Redis servers
822 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
823 case of MOVED into a DEFER case instead, thus moving the query onto the
824 next server in the list. For a cluster of N elements, all N servers must
825 be defined in redis_servers.
827 GF/02 Catch and remove uninitialized value warning in exiqsumm
828 Check for existence of @ARGV before looking at $ARGV[0]
830 JH/01 Replace the store_release() internal interface with store_newblock(),
831 which internalises the check required to safely use the old one, plus
832 the allocate and data copy operations duplicated in both (!) of the
833 extant use locations.
835 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
836 modifier. This matches the restriction on the commandline.
838 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
839 Previously only the last row was returned.
841 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
842 we assumed that tags in the header were well-formed, and parsed the
843 element content after inspecting only the first char of the tag.
844 Assumptions at that stage could crash the receive process on malformed
847 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
848 While running the DKIM ACL we operate on the Permanent memory pool so that
849 variables created with "set" persist to the DATA ACL. Also (at any time)
850 DNS lookups that fail create cache records using the Permanent pool. But
851 expansions release any allocations made on the current pool - so a dnsdb
852 lookup expansion done in the DKIM ACL releases the memory used for the
853 DNS negative-cache, and bad things result. Solution is to switch to the
854 Main pool for expansions.
855 While we're in that code, add checks on the DNS cache during store_reset,
856 active in the testsuite.
857 Problem spotted, and debugging aided, by Wolfgang Breyha.
859 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
860 When none of the hosts presented to a transport match an already-open
861 connection, close it and proceed with the list. Previously we would
862 queue the message. Spotted by Lena with Yahoo, probably involving
865 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
866 Previously a spurious "250 OK id=" response was appended to the proper
869 JH/08 The "support for" informational output now, which built with Content
870 Scanning support, has a line for the malware scanner interfaces compiled
871 in. Interface can be individually included or not at build time.
873 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
874 by the template makefile "src/EDITME". The "STREAM" support for an older
875 ClamAV interface method is removed.
877 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
878 rows affected is given instead).
880 JH/11 The runtime Berkeley DB library version is now additionally output by
881 "exim -d -bV". Previously only the compile-time version was shown.
883 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
884 SMTP connection. Previously, when one had more receipients than the
885 first, an abortive onward connection was made. Move to full support for
886 multiple onward connections in sequence, handling cutthrough connection
887 for all multi-message initiating connections.
889 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
890 routers. Previously, a multi-recipient message would fail to match the
891 onward-connection opened for the first recipient, and cause its closure.
893 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
894 a timeout on read on a GnuTLS initiating connection, resulting in the
895 initiating connection being dropped. This mattered most when the callout
896 was marked defer_ok. Fix to keep the two timeout-detection methods
899 JH/15 Relax results from ACL control request to enable cutthrough, in
900 unsupported situations, from error to silently (except under debug)
901 ignoring. This covers use with PRDR, frozen messages, queue-only and
904 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
906 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
907 metadata, resulting in a crash in free().
909 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
910 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
911 Broken also in d185889f4, with init system revamp.
913 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
914 Previously we abruptly closed the connection after reading a malware-
915 found indication; now we go on to read the "scan ok" response line,
918 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
919 ACL. Previously, a crash would result.
921 JH/19 Speed up macro lookups during configuration file read, by skipping non-
922 macro text after a replacement (previously it was only once per line) and
923 by skipping builtin macros when searching for an uppercase lead character.
925 JH/20 DANE support moved from Experimental to mainline. The Makefile control
926 for the build is renamed.
928 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
929 was allocated for every new TLS startup, meaning one per message. Fix
930 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
932 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
933 reported the original. Fix to report (as far as possible) the ACL
934 result replacing the original.
936 JH/23 Fix memory leak during multi-message connections using STARTTLS under
937 OpenSSL. Certificate information is loaded for every new TLS startup,
938 and the resources needed to be freed.
940 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
942 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
945 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
946 DATA response info to the (existing) per-recipient response info for
947 the "C=" log element. It can have useful tracking info from the
948 destination system. Patch from Simon Arlott.
950 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
951 length value. Previously this would segfault.
953 HS/02 Support Avast multiline protoocol, this allows passing flags to
954 newer versions of the scanner.
956 JH/28 Ensure that variables possibly set during message acceptance are marked
957 dead before release of memory in the daemon loop. This stops complaints
958 about them when the debug_store option is enabled. Discovered specifically
959 for sender_rate_period, but applies to a whole set of variables.
960 Do the same for the queue-runner and queue-list loops, for variables set
961 from spool message files. Do the same for the SMTP per-message loop, for
962 certain variables indirectly set in ACL operations.
964 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
965 as a multi-recipient message from a mailinglist manager). The coding had
966 an arbitrary cutoff number of characters while checking for more input;
967 enforced by writing a NUL into the buffer. This corrupted long / fast
968 input. The problem was exposed more widely when more pipelineing of SMTP
969 responses was introduced, and one Exim system was feeding another.
970 The symptom is log complaints of SMTP syntax error (NUL chars) on the
971 receiving system, and refused recipients seen by the sending system
972 (propating to people being dropped from mailing lists).
973 Discovered and pinpointed by David Carter.
975 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
976 replaced by the ${authresults } expansion.
978 JH/31 Bug 2257: Fix pipe transport to not use a socket-only syscall.
980 HS/03 Set a handler for SIGTERM and call exit(3) if running as PID 1. This
981 allows proper process termination in container environments.
983 JH/32 Bug 2258: Fix spool_wireformat in combination with LMTP transport.
984 Previously the "final dot" had a newline after it; ensure it is CR,LF.
986 JH/33 SPF: remove support for the "spf" ACL condition outcome values "err_temp"
987 and "err_perm", deprecated since 4.83 when the RFC-defined words
988 "temperror" and "permerror" were introduced.
990 JH/34 Re-introduce enforcement of no cutthrough delivery on transports having
991 transport-filters or DKIM-signing. The restriction was lost in the
992 consolidation of verify-callout and delivery SMTP handling.
993 Extend the restriction to also cover ARC-signing.
995 JH/35 Cutthrough: for a final-dot response timeout (and nonunderstood responses)
996 in defer=pass mode supply a 450 to the initiator. Previously the message
999 PP/02 DANE: add dane_require_tls_ciphers SMTP Transport option; if unset,
1000 tls_require_ciphers is used as before.
1002 HS/03 Malware Avast: Better match the Avast multiline protocol. Add
1003 "pass_unscanned". Only tmpfails from the scanner are written to
1004 the paniclog, as they may require admin intervention (permission
1005 denied, license issues). Other scanner errors (like decompression
1006 bombs) do not cause a paniclog entry.
1008 JH/36 Fix reinitialisation of DKIM logging variable between messages.
1009 Previously it was possible to log spurious information in receive log
1012 JH/37 Bug 2255: Revert the disable of the OpenSSL session caching. This
1013 triggered odd behaviour from Outlook Express clients.
1015 PP/03 Add util/renew-opendmarc-tlds.sh script for safe renewal of public
1018 JH/38 DKIM: accept Ed25519 pubkeys in SubjectPublicKeyInfo-wrapped form,
1019 since the IETF WG has not yet settled on that versus the original
1020 "bare" representation.
1022 JH/39 Fix syslog logging for syslog_timestamp=no and log_selector +millisec.
1023 Previously the millisecond value corrupted the output.
1024 Fix also for syslog_pid=no and log_selector +pid, for which the pid
1025 corrupted the output.
1031 JH/01 Rework error string handling in TLS interface so that the caller in
1032 more cases is responsible for logging. This permits library-sourced
1033 string to be attached to addresses during delivery, and collapses
1034 pairs of long lines into single ones.
1036 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
1037 during configuration. Wildcards are allowed and expanded.
1039 JH/02 Rework error string handling in DKIM to pass more info back to callers.
1040 This permits better logging.
1042 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
1043 do not close it down and have the child transport start it up again on
1044 the passed-on TCP connection. Instead, proxy the child (and any
1045 subsequent ones) for TLS via a unix-domain socket channel. Logging is
1046 affected: the continued delivery log lines do not have any DNSSEC, TLS
1047 Certificate or OCSP information. TLS cipher information is still logged.
1049 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
1050 identical IP addresses on different listening ports. Will also affect
1053 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
1054 add noisy ifdef guards to special-case this sillyness.
1055 Patch from Bernd Kuhls.
1057 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
1058 than 255 are no longer allowed.
1060 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
1061 Disable the session-cache too, which might reduce our load. Since we
1062 currrectly use a new context for every connection, both as server and
1063 client, there is no benefit for these.
1064 GnuTLS appears to not support tickets server-side by default (we don't
1065 call gnutls_session_ticket_enable_server()) but client side is enabled
1066 by default on recent versions (3.1.3 +) unless the PFS priority string
1069 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
1070 <https://reproducible-builds.org/specs/source-date-epoch/>.
1072 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
1073 the check for any unsuccessful recipients did not notice the limit, and
1074 erroneously found still-pending ones.
1076 JH/08 Pipeline CHUNKING command and data together, on kernels that support
1077 MSG_MORE. Only in-clear (not on TLS connections).
1079 JH/09 Avoid using a temporary file during transport using dkim. Unless a
1080 transport-filter is involved we can buffer the headers in memory for
1081 creating the signature, and read the spool data file once for the
1082 signature and again for transmission.
1084 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
1085 4.77 as the kernel support then wasn't solid, having issues in 64bit
1086 mode. Now, it's been long enough. Add support for FreeBSD also.
1088 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
1089 case where the routing stage had gathered several addresses to send to
1090 a host before calling the transport for the first, we previously failed
1091 to close down TLS in the old transport process before passing the TCP
1092 connection to the new process. The new one sent a STARTTLS command
1093 which naturally failed, giving a failed delivery and bloating the retry
1094 database. Investigation and fix prototype from Wolfgang Breyha.
1096 JH/12 Fix check on SMTP command input synchronisation. Previously there were
1097 false-negatives in the check that the sender had not preempted a response
1098 or prompt from Exim (running as a server), due to that code's lack of
1099 awareness of the SMTP input buffering.
1101 PP/04 Add commandline_checks_require_admin option.
1102 Exim drops privileges sanely, various checks such as -be aren't a
1103 security problem, as long as you trust local users with access to their
1104 own account. When invoked by services which pass untrusted data to
1105 Exim, this might be an issue. Set this option in main configuration
1106 AND make fixes to the calling application, such as using `--` to stop
1109 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
1110 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
1111 responses to those, into a single TLS record each way (this usually means
1112 a single packet). As a side issue, smtp_enforce_sync now works on TLS
1115 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
1116 affects you only if you're dancing at the edge of the param size limits.
1117 If you are, and this message makes sense to you, then: raise the
1118 configured limit or use OpenSSL 1.1. Nothing we can do for older
1121 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
1122 cmdline element to get the documented default one. Previously it was
1125 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
1126 are unsuable for non-message-specific reasons - eg. connection timeout,
1127 banner-time rejection.
1129 JH/16 Fix logging of delivery remote port, when specified by router, under
1132 PP/06 Repair manualroute's ability to take options in any order, even if one
1133 is the name of a transport.
1136 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
1138 JH/17 Change the list-building routines interface to use the expanding-string
1139 triplet model, for better allocation and copying behaviour.
1141 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
1142 Previously it was constructed the first time a possibly-matching string
1143 was met in the configuration file input during startup; now it is done
1146 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
1147 compatible one, to avoid the (poorly documented) possibility of a config
1148 file in the working directory redirecting the DB files, possibly correpting
1149 some existing file. CVE-2017-10140 assigned for BDB.
1151 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
1152 cache-hot. Previously, although the result was properly cached, the
1153 initial verify call returned a defer.
1155 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
1156 the main verify for receipient in uncached-mode.
1158 JH/22 Retire historical build files to an "unsupported" subdir. These are
1159 defined as "ones for which we have no current evidence of testing".
1161 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
1162 if present. Previously it was ignored.
1164 JH/24 Start using specified-initialisers in C structure init coding. This is
1165 a C99 feature (it's 2017, so now considered safe).
1167 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
1168 if was a fixed-sized field and bitmask ops via macros; it is now more
1171 PP/07 GitHub PR 56: Apply MariaDB build fix.
1172 Patch provided by Jaroslav Škarvada.
1174 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
1175 during Coverity cleanups [4.87 JH/47]
1176 Diagnosis and fix provided by Michael Fischer v. Mollard.
1178 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
1179 the right size to place the terminating semicolon on its own folded
1180 line, the header hash was calculated to an incorrect value thanks to
1181 the (relaxed) space the fold became.
1183 HS/02 Fix Bug 2130: large writes from the transport subprocess were chunked
1184 and confused the parent.
1186 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
1187 which could crash as a result. This could lead to undeliverable messages.
1189 JH/28 Logging: "next input sent too soon" now shows where input was truncated
1192 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
1193 matters on fast-turnover and PID-randomising systems, which were getting
1194 out-of-order delivery.
1196 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
1197 a possibly-overlapping copy. The symptom was that "Remote host closed
1198 connection in response to HELO" was logged instead of the actual 4xx
1201 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
1202 Previously only that bufferd was discarded, resulting in SYMTP command
1205 JH/32 DKIM: when a message has multiple signatures matching an identity given
1206 in dkim_verify_signers, run the dkim acl once for each. Previously only
1207 one run was done. Bug 2189.
1209 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
1210 "panic the current process" to "deliberately defer". The panic log is
1211 still written with the problem list name; the mail and reject logs now
1212 get a temp-reject line for the message that was being handled, saying
1213 something like "domains check lookup or other defer". The SMTP 451
1214 message is still "Temporary local problem".
1216 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
1217 A crafted sequence of BDAT commands could result in in-use memory beeing
1218 freed. CVE-2017-16943.
1220 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
1221 from SMTP input. Previously it was always done; now only done for DATA
1222 and not BDAT commands. CVE-2017-16944.
1224 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
1225 to the message (such as an overlong header line). Previously this was
1226 not done and we did not exit BDAT mode. Followon from the previous item
1227 though a different problem.
1233 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
1234 than -2003 did; needs libidn2 in addition to libidn.
1236 JH/02 The path option on a pipe transport is now expanded before use.
1238 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
1239 Patch provided by "Björn", documentation fix added too.
1241 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
1242 missing a wire-to-host endian conversion.
1244 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
1245 close after a BDAT command line could be taken as a following command,
1246 giving a synch failure. Fix by only checking for synch immediately
1247 before acknowledging the chunk.
1249 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
1250 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
1251 macro. Patches provided by Josh Soref.
1253 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
1254 Previously we did not; the RFC seems ambiguous and VRFY is not listed
1255 by IANA as a service extension. However, John Klensin suggests that we
1258 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
1259 the dkim code may be unix-mode line endings rather than smtp wire-format
1260 CRLF, so prepend a CR to any bare LF.
1262 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
1263 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
1265 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
1266 the first were themselves being wrongly included in the feed into dkim
1267 processing; with most chunk sizes in use this resulted in an incorrect
1268 body hash calculated value.
1270 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
1271 DKIM signature block, for verification. Although advised against by
1272 standards it is specifically not ruled illegal.
1274 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
1276 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
1277 missing a body hash (the bh= tag).
1279 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
1280 It seems that HAProxy sends the Proxy Protocol information in clear and
1281 only then does a TLS startup, so do the same.
1283 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
1284 TCP connections (such as for Spamd) unless the daemon successfully set
1285 Fast Open mode on its listening sockets. This fixes breakage seen on
1286 too-old kernels or those not configured for Fast Open, at the cost of
1287 requiring both directions being enabled for TFO, and TFO never being used
1288 by non-daemon-related Exim processes.
1290 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
1291 endings, at least on the first header line. Try to canonify any that get
1292 past that check, despite the cost.
1294 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
1295 now limited to an arbitrary five deep, while parsing addresses with the
1296 strip_excess_angle_brackets option enabled.
1298 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
1299 instead leave the unprompted TLS handshake in socket buffer for the
1300 TLS library to consume.
1302 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
1304 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
1306 JH/16 Drop variables when they go out of scope. Memory management drops a whole
1307 region in one operation, for speed, and this leaves assigned pointers
1308 dangling. Add checks run only under the testsuite which checks all
1309 variables at a store-reset and panics on a dangling pointer; add code
1310 explicitly nulling out all the variables discovered. Fixes one known
1311 bug: a transport crash, where a dangling pointer for $sending_ip_address
1312 originally assigned in a verify callout, is re-used.
1314 PP/06 Drop '.' from @INC in various Perl scripts.
1316 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
1318 PP/08 Reduce a number of compilation warnings under clang; building with
1319 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
1320 should be warning-free.
1322 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
1324 HS/01 Fix portability problems introduced by PP/08 for platforms where
1325 realloc(NULL) is not equivalent to malloc() [SunOS et al].
1327 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
1328 chunk. This allows us to accept broken chunked messages. We need a more
1329 general solution here.
1331 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
1332 already-broken messages in the queue.
1334 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
1336 JH/19 Fix reference counting bug in routing-generated-address tracking.
1342 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
1343 supports it and a size is available (ie. the sending peer gave us one).
1345 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
1346 years of being deprecated). The replacements are the ACLs
1347 acl_smtp_mime and acl_not_smtp_mime.
1349 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
1350 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
1351 or even an in-clear connection were permitted. Now, if the host lookup
1352 was dnssec and dane was requested then the host is only used if the
1353 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
1354 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
1355 if one fails this test.
1356 This means that a poorly-configured remote DNS will make it incommunicado;
1357 but it protects against a DNS-interception attack on it.
1359 JH/04 Bug 1810: make continued-use of an open smtp transport connection
1360 non-noisy when a race steals the message being considered.
1362 JH/05 If main configuration option tls_certificate is unset, generate a
1363 self-signed certificate for inbound TLS connections.
1365 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
1366 in rewrites and routers.
1368 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
1369 and logged a warning sing 4.83; now they are a configuration file error.
1371 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
1372 (lacking @domain). Apply the same qualification processing as RCPT.
1374 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
1376 JH/10 Support ${sha256:} applied to a string (as well as the previous
1379 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
1380 a cutthrough deliver is pending, as we always want to make a connection.
1381 This also avoids re-routing the message when later placing the cutthrough
1382 connection after a verify cache hit.
1383 Do not update it with the verify result either.
1385 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
1386 when routing results in more than one destination address.
1388 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
1389 signing (which inhibits the cutthrough capability). Previously only
1390 the presence of an option was tested; now an expansion evaluating as
1391 empty is permissible (obviously it should depend only on data available
1392 when the cutthrough connection is made).
1394 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
1395 the relevant preceding SMTP command did not note the pipelining mode.
1397 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
1398 Previously they were not counted.
1400 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
1401 as one having no matching records. Previously we deferred the message
1402 that needed the lookup.
1404 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
1405 distinguished as "(=".
1407 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
1408 for missing MX records. Previously it only worked for missing A records.
1410 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
1412 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
1413 after the data-go-ahead and data-ack. Patch from Jason Betts.
1415 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
1416 even for a "none" policy. Patch from Tony Meyer.
1418 JH/22 Fix continued use of a connection for further deliveries. If a port was
1419 specified by a router, it must also match for the delivery to be
1422 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
1423 When one of the recipients of a message was unsuitable for the connection
1424 (has no matching addresses), we lost track of needing to mark it
1425 deferred. As a result mail would be lost.
1427 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
1429 JH/25 Decoding ACL controls is now done using a binary search; the source code
1430 takes up less space and should be simpler to maintain. Merge the ACL
1431 condition decode tables also, with similar effect.
1433 JH/26 Fix problem with one_time used on a redirect router which returned the
1434 parent address unchanged. A retry would see the parent address marked as
1435 delivered, so not attempt the (identical) child. As a result mail would
1438 JH/27 Fix a possible security hole, wherein a process operating with the Exim
1439 UID can gain a root shell. Credit to http://www.halfdog.net/ for
1440 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
1443 JH/28 Enable {spool,log} filesystem space and inode checks as default.
1444 Main config options check_{log,spool}_{inodes,space} are now
1445 100 inodes, 10MB unless set otherwise in the configuration.
1447 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
1448 Previously it only applied to the main-section connection policy
1451 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
1453 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
1454 by me. Added RFC7919 DH primes as an alternative.
1456 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
1457 are not in the system include path.
1459 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
1460 GnuTLS, when a session startup failed (eg because the client disconnected)
1461 Exim did stdio operations after fclose. This was exposed by a recent
1462 change which nulled out the file handle after the fclose.
1464 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
1465 signed directly by the cert-signing cert, rather than an intermediate
1466 OCSP-signing cert. This is the model used by LetsEncrypt.
1468 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
1470 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
1471 an incoming connection.
1473 HS/02 Bug 1802: Do not half-close the connection after sending a request
1476 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
1477 fallback to "prime256v1".
1479 JH/34 SECURITY: Use proper copy of DATA command in error message.
1480 Could leak key material. Remotely exploitable. CVE-2016-9963.
1486 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
1487 and 3.4.4 - once the server is enabled to respond to an OCSP request
1488 it does even when not requested, resulting in a stapling non-aware
1489 client dropping the TLS connection.
1491 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
1492 support variable-length bit vectors. No functional change.
1494 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
1495 The I= interface field on outgoing lines is now after the H= remote
1496 host field, same as incoming lines. There is a separate
1497 outgoing_interface log selector which allows you to disable the
1500 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
1501 If not running log_selector +smtp_connection the mainlog would be held
1502 open indefinitely after a "too many connections" event, including to a
1503 deleted file after a log rotate. Leave the per net connection logging
1504 leaving it open for efficiency as that will be quickly detected by the
1505 check on the next write.
1507 HS/01 Bug 1671: Fix post transport crash.
1508 Processing the wait-<transport> messages could crash the delivery
1509 process if the message IDs didn't exist for some reason. When
1510 using 'split_spool_directory=yes' the construction of the spool
1511 file name failed already, exposing the same netto behaviour.
1513 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
1514 mime_regex ACL conditions.
1516 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
1517 to DSN fail messages (bounces): remote IP, remote greeting, remote response
1518 to HELO, local diagnostic string.
1520 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
1521 log line to debug. Even when configured with a tls authenticator many
1522 client connections are expected to not authenticate in this way, so
1523 an authenticate fail is not an error.
1525 HS/02 Add the Exim version string to the process info. This way exiwhat
1526 gives some more detail about the running daemon.
1528 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
1529 matter for fast-change records such as DNSBLs.
1531 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
1532 retry record, even if constant. There may be multiple transports with
1533 different interface settings and the retry behaviour needs to be kept
1536 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
1538 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
1540 JH/10 Bug 840: fix log_defer_output option of pipe transport
1542 JH/11 Bug 830: use same host for all RCPTS of a message, even under
1543 hosts_randomize. This matters a lot when combined with mua_wrapper.
1545 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
1546 ${quote_pgsql:<string>} operator.
1548 JH/13 Bug 1708: avoid misaligned access in cached lookup.
1550 JH/14 Change header file name for freeradius-client. Relevant if compiling
1551 with Radius support; from the Gentoo tree and checked under Fedora.
1553 JH/15 Bug 1712: Introduce $prdr_requested flag variable
1555 JH/16 Bug 1714: Permit an empty string as expansion result for transport
1556 option transport_filter, meaning no filtering.
1558 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
1560 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
1561 defaults to "*" (all hosts). The variable is now available when not built
1562 with TLS, default unset, mainly to enable keeping the testsuite sane.
1563 If a server certificate is not supplied (via tls_certificate) an error is
1564 logged, and clients will find TLS connections fail on startup. Presumably
1565 they will retry in-clear.
1566 Packagers of Exim are strongly encouraged to create a server certificate
1567 at installation time.
1569 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
1570 with the $config_file variable.
1572 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
1573 in transport context, after the attempt, and per-recipient. The latter type
1574 is per host attempted. The event data is the error message, and the errno
1575 information encodes the lookup type (A vs. MX) used for the (first) host,
1576 and the trailing two digits of the smtp 4xx response.
1578 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
1579 to write to mainlog (or rejectlog, paniclog) in the window between file
1580 creation and permissions/ownership being changed. Particularly affects
1581 installations where exicyclog is run as root, rather than exim user;
1582 result is that the running daemon panics and dies.
1584 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
1586 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
1587 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
1588 "pri" and "weight". Note that the previous implicit priority given by the
1589 list order is no longer honoured.
1591 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
1592 for DKIM processing.
1594 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
1595 by defining SUPPORT_SOCKS.
1597 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
1598 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
1599 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
1600 variables are renamed to proxy_{local,external}_{address,port}.
1602 JH/27 Move Internationalisation support from Experimental to mainline, enabled
1603 for a build by defining SUPPORT_I18N
1605 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
1606 of the query string, and make ${quote_redis:} do that quoting.
1608 JH/29 Move Events support from Experimental to mainline, enabled by default
1609 and removable for a build by defining DISABLE_EVENT.
1611 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
1613 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
1614 cached by the daemon.
1616 JH/32 Move Redis support from Experimental to mainline, enabled for a build
1617 by defining LOOKUP_REDIS. The libhiredis library is required.
1619 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
1620 keys are given for lookup.
1622 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
1623 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
1624 only supported when built with TLS support. The PolarSSL SHA routines
1625 are still used when the TLS library is too old for convenient support.
1627 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
1628 openssl_options), for security. OpenSSL forces this from version 1.1.0
1629 server-side so match that on older versions.
1631 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
1632 allocation for $value could be released as the expansion processing
1633 concluded, but leaving the global pointer active for it.
1635 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
1636 and to use the domains and local_parts ACL conditions.
1638 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
1639 incorrectly not doubled on cutthrough transmission, hence seen as a
1640 body-termination at the receiving system - resulting in truncated mails.
1641 Commonly the sender saw a TCP-level error, and retransmitted the message
1642 via the normal store-and-forward channel. This could result in duplicates
1643 received - but deduplicating mailstores were liable to retain only the
1644 initial truncated version.
1646 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
1648 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
1650 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
1651 we're in there, support oversigning also; bug 1309.
1653 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
1655 HS/04 Add support for keep_environment and add_environment options.
1657 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
1658 either intentional arithmetic overflow during PRNG, or testing config-
1661 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
1662 delivery resulted in actual delivery. Cancel cutthrough before DATA
1665 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
1666 rejects a recipient: pass the reject to the originator.
1668 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
1669 Many were false-positives and ignorable, but it's worth fixing the
1672 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
1673 for the new environment-manipulation done at startup. Move the routines
1674 from being local to tls.c to being global via the os.c file.
1676 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
1677 an extract embedded as result-arg for a map, the first arg for extract
1678 is unavailable so we cannot tell if this is a numbered or keyed
1679 extraction. Accept either.
1685 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
1688 JH/02 The smtp transport option "multi_domain" is now expanded.
1690 JH/03 The smtp transport now requests PRDR by default, if the server offers
1693 JH/04 Certificate name checking on server certificates, when exim is a client,
1694 is now done by default. The transport option tls_verify_cert_hostnames
1695 can be used to disable this per-host. The build option
1696 EXPERIMENTAL_CERTNAMES is withdrawn.
1698 JH/05 The value of the tls_verify_certificates smtp transport and main options
1699 default to the word "system" to access the system default CA bundle.
1700 For GnuTLS, only version 3.0.20 or later.
1702 JH/06 Verification of the server certificate for a TLS connection is now tried
1703 (but not required) by default. The verification status is now logged by
1704 default, for both outbound TLS and client-certificate supplying inbound
1707 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
1710 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
1711 Status Notification (bounce) messages are now MIME format per RFC 3464.
1712 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
1713 under the control of the dsn_advertise_hosts option, and routers may
1714 have a dsn_lasthop option.
1716 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
1717 default, modifiable by a malware= option. The list separator for
1718 the options can now be changed in the usual way. Bug 68.
1720 JH/10 The smtp_receive_timeout main option is now expanded before use.
1722 JH/11 The incoming_interface log option now also enables logging of the
1723 local interface on delivery outgoing connections.
1725 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
1726 if the interface and destination host and port all match.
1728 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
1731 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
1732 Patch from Andrew Lewis.
1734 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
1735 now supports optional time-restrictions, weighting, and priority
1736 modifiers per server. Patch originally by <rommer@active.by>.
1738 JH/16 The spamd_address main option now supports a mixed list of local
1739 and remote servers. Remote servers can be IPv6 addresses, and
1740 specify a port-range.
1742 JH/17 Bug 68: The spamd_address main option now supports an optional
1743 timeout value per server.
1745 JH/18 Bug 1581: Router and transport options headers_add/remove can
1746 now have the list separator specified.
1748 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
1751 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
1754 JH/21 Support for the A6 type of dns record is withdrawn.
1756 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
1757 rather than the verbs used.
1759 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
1760 from 255 to 1024 chars.
1762 JH/24 Verification callouts now attempt to use TLS by default.
1764 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
1765 are generic router options now. The defaults didn't change.
1767 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
1768 Original patch from Alexander Shikoff, worked over by JH.
1770 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
1771 files if ZCAT_COMMAND is not executable.
1773 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
1775 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
1777 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
1778 Normally benign, it bites when the pair was led to by a CNAME;
1779 modern usage is to not canonicalize the domain to a CNAME target
1780 (and we were inconsistent anyway for A-only vs AAAA+A).
1782 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
1784 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
1785 when evaluating $sender_host_dnssec.
1787 JH/31 Check the HELO verification lookup for DNSSEC, adding new
1788 $sender_helo_dnssec variable.
1790 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
1792 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
1794 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
1796 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
1797 documented as working, but never had. Support all but $spam_report.
1799 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
1800 added for tls authenticator.
1802 HS/03 Add perl_taintmode main config option
1808 TL/01 When running the test suite, the README says that variables such as
1809 no_msglog_check are global and can be placed anywhere in a specific
1810 test's script, however it was observed that placement needed to be near
1811 the beginning for it to behave that way. Changed the runtest perl
1812 script to read through the entire script once to detect and set these
1813 variables, reset to the beginning of the script, and then run through
1814 the script parsing/test process like normal.
1816 TL/02 The BSD's have an arc4random API. One of the functions to induce
1817 adding randomness was arc4random_stir(), but it has been removed in
1818 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
1819 function when detected.
1821 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
1822 cause callback expansion.
1824 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
1825 syntax errors in an expansion can be treated as a string instead of
1826 logging or causing an error, due to the internal use of bool_lax
1827 instead of bool when processing it.
1829 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
1830 server certificates when making smtp deliveries.
1832 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
1834 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
1836 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
1838 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
1839 Merged patch from Sebastian Wiedenroth.
1841 JH/05 Fix results-pipe from transport process. Several recipients, combined
1842 with certificate use, exposed issues where response data items split
1843 over buffer boundaries were not parsed properly. This eventually
1844 resulted in duplicates being sent. This issue only became common enough
1845 to notice due to the introduction of connection certificate information,
1846 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1848 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1849 size buffer was used, resulting in syntax errors when an expansion
1852 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1853 version 3.3.6 or later.
1855 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1856 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1857 both become "event_action", the variables become $event_name, $event_data
1858 and $event_defer_errno. There is a new variable $verify_mode, usable in
1859 routers, transports and related events. The tls:cert event is now also
1860 raised for inbound connections, if the main configuration event_action
1863 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1864 early OCSP support, but no stapling (appears to be less than 1.0.0).
1866 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1867 server certificate names available under the smtp transport option
1868 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1871 JH/10 Time-related extraction expansions from certificates now use the main
1872 option "timezone" setting for output formatting, and are consistent
1873 between OpenSSL and GnuTLS compilations. Bug 1541.
1875 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1876 encoded parameter in the incoming message. Bug 1558.
1878 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1879 include certificate info, eximon was claiming there were spoolfile
1882 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1884 JH/14 Log delivery-related information more consistently, using the sequence
1885 "H=<name> [<ip>]" wherever possible.
1887 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1888 are problematic for Debian distribution, omit them from the release
1891 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1893 JH/16 Fix string representation of time values on 64bit time_t architectures.
1896 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1897 output list separator was used.
1902 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1903 checkers that were complaining about end of non-void function with no
1906 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1907 This was a regression introduced in 4.83 by another bugfix.
1909 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1911 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1912 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1918 TF/01 Correctly close the server side of TLS when forking for delivery.
1920 When a message was received over SMTP with TLS, Exim failed to clear up
1921 the incoming connection properly after forking off the child process to
1922 deliver the message. In some situations the subsequent outgoing
1923 delivery connection happened to have the same fd number as the incoming
1924 connection previously had. Exim would try to use TLS and fail, logging
1925 a "Bad file descriptor" error.
1927 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1928 utilities have not been installed.
1930 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1931 temporary space as the ACL may create new global variables.
1933 TL/01 LDAP support uses per connection or global context settings, depending
1934 upon the detected version of the libraries at build time.
1936 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1937 to extract and use the src ip:port in logging and expansions as if it
1938 were a direct connection from the outside internet. PPv2 support was
1939 updated based on HAProxy spec change in May 2014.
1941 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1943 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1944 Properly escape header and check for NULL return.
1946 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1949 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1951 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1952 characters in header names, implemented as a verify condition.
1953 Contributed by Michael Fischer v. Mollard.
1955 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1956 results permerror and temperror. Previous values are deprecated but
1957 still accepted. In a future release, err_perm and err_temp will be
1958 completely removed, which will be a backward incompatibility if the
1959 ACL tests for either of these two old results. Patch contributed by
1960 user bes-internal on the mailing list.
1962 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1964 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1965 selectors, in both main and reject logs.
1967 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1970 JH/07 Add malware type "sock" for talking to simple daemon.
1972 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1974 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1975 routers/transports under cutthrough routing.
1977 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1978 numbers. Touch up "bool" conditional to keep the same definition.
1980 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1982 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1984 JH/12 Expand items in router/transport headers_add or headers_remove lists
1985 individually rather than the list as a whole. Bug 1452.
1987 Required for reasonable handling of multiple headers_ options when
1988 they may be empty; requires that headers_remove items with embedded
1989 colons must have them doubled (or the list-separator changed).
1991 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1992 view the policy declared in the DMARC record. Currently, $dmarc_status
1993 is a combined value of both the record presence and the result of the
1996 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1998 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1999 dnslookup router and the smtp transport (applying to the forward
2002 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
2003 of ldap servers used for a specific lookup. Patch provided by Heiko
2006 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
2007 New variable $lookup_dnssec_authenticated for observability.
2009 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
2010 Patch submitted by Lars Timman.
2012 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
2014 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
2015 Requires trusted mode and valid format message id, aborts otherwise.
2016 Patch contributed by Heiko Schlichting.
2018 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
2019 certextract with support for various fields. Bug 1358.
2021 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
2022 is requested by default, modifiable by smtp transport option
2025 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
2026 operate on certificate variables to give certificate fingerprints
2027 Also new ${sha256:cert_variable}.
2029 JH/23 The PRDR feature is moved from being Experimental into the mainline.
2031 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
2032 Christian Aistleitner.
2034 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
2036 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
2037 file. Patch from Wolfgang Breyha.
2039 JH/25 Expand the coverage of the delivery $host and $host_address to
2040 client authenticators run in verify callout. Bug 1476.
2042 JH/26 Port service names are now accepted for tls_on_connect_ports, to
2043 align with daemon_smtp_ports. Bug 72.
2045 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
2046 support and error reporting did not work properly.
2048 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
2049 and is readable. Patch from Andrew Colin Kissa.
2051 TL/14 Enhance documentation of ${run expansion and how it parses the
2052 commandline after expansion, particularly in the case when an
2053 unquoted variable expansion results in an empty value.
2055 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
2057 PP/02 Fix internal collision of T_APL on systems which support RFC3123
2058 by renaming away from it. Addresses GH issue 15, reported by
2061 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
2063 TL/15 SECURITY: prevent double expansion in math comparison functions
2064 (can expand unsanitized data). Not remotely exploitable.
2071 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
2073 PP/02 Make -n do something, by making it not do something.
2074 When combined with -bP, the name of an option is not output.
2076 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
2079 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
2080 $sender_host_name and config options to manage this, and basic check
2083 PP/05 DSCP support for outbound connections and control modifier for inbound.
2085 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
2086 (Only plugin which currently uses this is kerberos4, which nobody should
2087 be using, but we should make it available and other future plugins might
2088 conceivably use it, even though it would break NAT; stuff *should* be
2089 using channel bindings instead).
2091 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
2092 name; added for Sendmail compatibility; requires admin caller.
2093 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
2094 just ignore it); requires trusted caller.
2095 Also parse but ignore: -Ac -Am -X<logfile>
2098 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
2100 TL/02 Add +smtp_confirmation as a default logging option.
2102 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
2103 Patch by Magnus Holmgren from 2007-02-20.
2105 TL/04 Bugzilla 1281 - Spec typo.
2106 Bugzilla 1283 - Spec typo.
2107 Bugzilla 1290 - Spec grammar fixes.
2109 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
2111 TL/06 Add Experimental DMARC support using libopendmarc libraries.
2113 TL/07 Fix an out of order global option causing a segfault. Reported to dev
2114 mailing list by by Dmitry Isaikin.
2116 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
2118 JH/02 Support "G" suffix to numbers in ${if comparisons.
2120 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
2122 NM/01 Bugzilla 1197 - Spec typo
2123 Bugzilla 1196 - Spec examples corrections
2125 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
2127 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
2128 gnutls_enable_pkcs11, but renamed to more accurately indicate its
2131 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
2132 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
2134 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
2135 "acl {{name}{arg}...}", and optional args on acl condition
2138 JH/05 Permit multiple router/transport headers_add/remove lines.
2140 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
2142 JH/07 Avoid using a waiting database for a single-message-only transport.
2143 Performance patch from Paul Fisher. Bugzilla 1262.
2145 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
2148 JH/09 Add $headers_added variable, with content from use of ACL modifier
2149 add_header (but not yet added to the message). Bugzilla 199.
2151 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
2152 Pulled from Bugzilla 817 by Wolfgang Breyha.
2154 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
2156 (nb: this is the same fix as in Exim 4.80.1)
2158 JH/11 Add A= logging on delivery lines, and a client_set_id option on
2161 JH/12 Add optional authenticated_sender logging to A= and a log_selector
2164 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
2166 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
2167 advertise SMTP AUTH mechanism to us, instead of a generic
2168 protocol violation error. Also, make Exim more robust to bad
2169 data from the Dovecot auth socket.
2171 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
2173 When a queue runner is handling a message, Exim first routes the
2174 recipient addresses, during which it prunes them based on the retry
2175 hints database. After that it attempts to deliver the message to
2176 any remaining recipients. It then updates the hints database using
2179 So if a recipient address works intermittently, it can get repeatedly
2180 deferred at routing time. The retry hints record remains fresh so the
2181 address never reaches the final cutoff time.
2183 This is a fairly common occurrence when a user is bumping up against
2184 their storage quota. Exim had some logic in its local delivery code
2185 to deal with this. However it did not apply to per-recipient defers
2186 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
2188 This change adds a proper retry rule check during routing so that the
2189 final cutoff time is checked against the message's age. We only do
2190 this check if there is an address retry record and there is not a
2191 domain retry record; this implies that previous attempts to handle
2192 the address had the retry_use_local_parts option turned on. We use
2193 this as an approximation for the destination being like a local
2194 delivery, as in LMTP.
2196 I suspect this new check makes the old local delivery cutoff check
2197 redundant, but I have not verified this so I left the code in place.
2199 TF/02 Correct gecos expansion when From: is a prefix of the username.
2201 Test 0254 submits a message to Exim with the header
2205 When I ran the test suite under the user fanf2, Exim expanded
2206 the header to contain my full name, whereas it should have added
2207 a Resent-Sender: header. It erroneously treats any prefix of the
2208 username as equal to the username.
2210 This change corrects that bug.
2212 GF/01 DCC debug and logging tidyup
2213 Error conditions log to paniclog rather than rejectlog.
2214 Debug lines prefixed by "DCC: " to remove any ambiguity.
2216 TF/03 Avoid unnecessary rebuilds of lookup-related code.
2218 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
2219 Bug spotted by Jeremy Harris; was flawed since initial commit.
2220 Would have resulted in OCSP responses post-SNI triggering an Exim
2221 NULL dereference and crash.
2223 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
2225 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
2226 Bug detection, analysis and fix by Samuel Thibault.
2227 Bugzilla 1331, Debian bug #698092.
2229 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
2231 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
2232 Server implementation by Todd Lyons, client by JH.
2233 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
2234 config variable "prdr_enable" controls whether the server
2235 advertises the facility. If the client requests PRDR a new
2236 acl_data_smtp_prdr ACL is called once for each recipient, after
2237 the body content is received and before the acl_smtp_data ACL.
2238 The client is controlled by both of: a hosts_try_prdr option
2239 on the smtp transport, and the server advertisement.
2240 Default client logging of deliveries and rejections involving
2241 PRDR are flagged with the string "PRDR".
2243 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
2244 fclose(). Diagnosis by Todd Lyons.
2246 PP/17 Update configure.default to handle IPv6 localhost better.
2247 Patch by Alain Williams (plus minor tweaks).
2250 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
2251 This is now consistent with GnuTLS, and is now documented: the
2252 previous undocumented portable approach to treating the option as
2253 unset was to force an expansion failure. That still works, and
2254 an empty string is now equivalent.
2256 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
2257 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
2258 not performing validation itself.
2260 PP/20 Added force_command boolean option to pipe transport.
2261 Patch from Nick Koston, of cPanel Inc.
2263 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
2266 TF/04 Added udpsend ACL modifier and hexquote expansion operator
2268 PP/21 Fix eximon continuous updating with timestamped log-files.
2269 Broken in a format-string cleanup in 4.80, missed when I repaired the
2270 other false fix of the same issue.
2271 Report and fix from Heiko Schlichting.
2274 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
2275 Report from Prashanth Katuri.
2277 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
2278 It's SecureTransport, so affects any MacOS clients which use the
2279 system-integrated TLS libraries, including email clients.
2281 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
2282 using a MIME ACL for non-SMTP local injection.
2283 Report and assistance in diagnosis by Warren Baker.
2285 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
2287 JH/16 Fix comparisons for 64b. Bugzilla 1385.
2289 TL/09 Add expansion variable $authenticated_fail_id to keep track of
2290 last id that failed so it may be referenced in subsequent ACL's.
2292 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
2295 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
2296 ldap library initialization, allowing self-signed CA's to be
2297 used. Also properly sets require_cert option later in code by
2298 using NULL (global ldap config) instead of ldap handle (per
2299 session). Bug diagnosis and testing by alxgomz.
2301 TL/12 Enhanced documentation in the ratelimit.pl script provided in
2302 the src/util/ subdirectory.
2304 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
2305 renamed to Transport Post Delivery Action by Jeremy Harris, as
2308 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
2309 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
2310 redis_servers = needs to be configured which will be used by the redis
2311 lookup. Patch from Warren Baker, of The Packet Hub.
2313 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
2315 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
2316 hostname or reverse DNS when processing a host list. Used suggestions
2317 from multiple comments on this bug.
2319 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
2321 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
2322 Missed a few lines, added it to make the runtest require no keyboard
2325 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
2326 contains upper case chars. Make router use caseful_local_part.
2328 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
2329 support when GnuTLS has been built with p11-kit.
2335 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
2337 This, or similar/improved, will also be change PP/11 of 4.82.
2343 PP/01 Handle short writes when writing local log-files.
2344 In practice, only affects FreeBSD (8 onwards).
2345 Bugzilla 1053, with thanks to Dmitry Isaikin.
2347 NM/01 Bugzilla 949 - Documentation tweak
2349 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
2352 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
2354 PP/02 Implemented gsasl authenticator.
2356 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
2358 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
2359 `pkg-config foo` for cflags/libs.
2361 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
2362 with rest of GSASL and with heimdal_gssapi.
2364 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
2365 `pkg-config foo` for cflags/libs for the TLS implementation.
2367 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
2368 properties get this fed in as external SSF. A number of robustness
2369 and debugging improvements to the cyrus_sasl authenticator.
2371 PP/08 cyrus_sasl server now expands the server_realm option.
2373 PP/09 Bugzilla 1214 - Log authentication information in reject log.
2374 Patch by Jeremy Harris.
2376 PP/10 Added dbmjz lookup type.
2378 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
2380 PP/12 MAIL args handles TAB as well as SP, for better interop with
2381 non-compliant senders.
2382 Analysis and variant patch by Todd Lyons.
2384 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
2385 Bug report from Lars Müller <lars@samba.org> (via SUSE),
2386 Patch from Dirk Mueller <dmueller@suse.com>
2388 PP/13 tls_peerdn now print-escaped for spool files.
2389 Observed some $tls_peerdn in wild which contained \n, which resulted
2390 in spool file corruption.
2392 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
2393 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
2394 or write after TLS renegotiation, which otherwise led to messages
2397 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
2398 as a tracking header (ie: a signed header comes before the signature).
2399 Patch from Wolfgang Breyha.
2401 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
2402 comma-sep list; embedded commas doubled.
2404 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
2406 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
2408 Report and patch from Dmitry Banschikov.
2410 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
2411 Removed SSL_clear() after SSL_new() which led to protocol negotiation
2412 failures. We appear to now support TLS1.1+ with Exim.
2414 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
2415 lets Exim select keys and certificates based upon TLS SNI from client.
2416 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
2417 before an outbound SMTP session. New log_selector, +tls_sni.
2419 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
2420 NULL dereference. Report and patch from Alun Jones.
2422 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
2423 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
2424 Not seeing resolver debug output on NetBSD, but suspect this is a
2425 resolver implementation change.
2427 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
2428 Left warnings. Added "eximon gdb" invocation mode.
2430 PP/21 Defaulting "accept_8bitmime" to true, not false.
2432 PP/22 Added -bw for inetd wait mode support.
2434 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
2435 locate the relevant includes and libraries. Made this the default.
2437 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
2438 Bugzilla 1246, report and most of solution from Tomasz Kusy.
2440 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
2441 This may cause build issues on older platforms.
2443 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
2444 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
2445 gnutls_require_mac & gnutls_require_protocols (no longer supported).
2446 Added SNI support via GnuTLS too.
2447 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
2449 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
2451 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
2452 Applied second patch from Janne, implementing suggestion to default
2453 multiple-strings-in-record handling to match SPF spec.
2455 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
2457 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
2458 read-only, out of scope).
2459 Patch from Wolfgang Breyha, report from Stuart Northfield.
2461 PP/29 Fix three issues highlighted by clang analyser static analysis.
2462 Only crash-plausible issue would require the Cambridge-specific
2463 iplookup router and a misconfiguration.
2464 Report from Marcin Mirosław.
2466 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
2468 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
2469 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
2470 As part of this, removing so much warning spew let me fix some minor
2471 real issues in debug logging.
2473 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
2474 assignment on my part. Fixed.
2476 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
2477 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
2478 Janne Snabb (who went above and beyond: thank you).
2480 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
2481 string otherwise requires a connection and a bunch more work and it's
2482 relatively easy to get wrong. Should also expose TLS library linkage
2485 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
2486 64-bit ${eval} (JH/03).
2488 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
2489 GNU libc to support some of the 64-bit stuff, should not lead to
2490 conflicts. Defined before os.h is pulled in, so if a given platform
2491 needs to override this, it can.
2493 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
2494 protection layer was required, which is not implemented.
2495 Bugzilla 1254, patch from Wolfgang Breyha.
2497 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
2498 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
2499 tls_dhparam take prime identifiers. Also unbreak combination of
2500 OpenSSL+DH_params+TLSSNI.
2502 PP/39 Disable SSLv2 by default in OpenSSL support.
2508 PP/01 Solaris build fix for Oracle's LDAP libraries.
2509 Bugzilla 1109, patch from Stephen Usher.
2511 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
2513 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
2516 TF/02 Fix a couple more cases where we did not log the error message
2517 when unlink() failed. See also change 4.74-TF/03.
2519 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
2520 lock up or crash if it happened to be inside a call to libc when it
2521 got a SIGUSR1 from exiwhat.
2523 The SIGUSR1 handler appends the current process status to the process
2524 log which is later printed by exiwhat. It used to use the general
2525 purpose logging code to do this, but several functions it calls are
2526 not safe for signals.
2528 The new output code in the SIGUSR1 handler is specific to the process
2529 log, and simple enough that it's easy to inspect for signal safety.
2530 Removing some special cases also simplifies the general logging code.
2531 Removing the spurious timestamps from the process log simplifies
2534 TF/04 Improved ratelimit ACL condition.
2536 The /noupdate option has been deprecated in favour of /readonly which
2537 has clearer semantics. The /leaky, /strict, and /readonly update modes
2538 are mutually exclusive. The update mode is no longer included in the
2539 database key; it just determines when the database is updated. (This
2540 means that when you upgrade Exim will forget old rate measurements.)
2542 Exim now checks that the per_* options are used with an update mode that
2543 makes sense for the current ACL. For example, when Exim is processing a
2544 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
2545 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
2546 must specify per_mail/readonly. If you omit the update mode it defaults to
2547 /leaky where that makes sense (as before) or /readonly where required.
2549 The /noupdate option is now undocumented but still supported for
2550 backwards compatibility. It is equivalent to /readonly except that in
2551 ACLs where /readonly is required you may specify /leaky/noupdate or
2552 /strict/noupdate which are treated the same as /readonly.
2554 A useful new feature is the /count= option. This is a generalization
2555 of the per_byte option, so that you can measure the throughput of other
2556 aggregate values. For example, the per_byte option is now equivalent
2557 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
2559 The per_rcpt option has been generalized using the /count= mechanism
2560 (though it's more complicated than the per_byte equivalence). When it is
2561 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
2562 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
2563 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
2564 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
2565 Note that using per_rcpt with a non-readonly update mode in more than
2566 one ACL will cause the recipients to be double-counted. (The per_mail
2567 and per_byte options don't have this problem.)
2569 The handling of very low rates has changed slightly. If the computed rate
2570 is less than the event's count (usually one) then this event is the first
2571 after a long gap. In this case the rate is set to the same as this event's
2572 count, so that the first message of a spam run is counted properly.
2574 The major new feature is a mechanism for counting the rate of unique
2575 events. The new per_addr option counts the number of different
2576 recipients that someone has sent messages to in the last time period. It
2577 behaves like per_rcpt if all the recipient addresses are different, but
2578 duplicate recipient addresses do not increase the measured rate. Like
2579 the /count= option this is a general mechanism, so the per_addr option
2580 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
2581 example, measure the rate that a client uses different sender addresses
2582 with the options per_mail/unique=$sender_address. There are further
2583 details in the main documentation.
2585 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
2587 TF/06 Removed a few PCRE remnants.
2589 TF/07 Automatically extract Exim's version number from tags in the git
2590 repository when doing development or release builds.
2592 PP/02 Raise smtp_cmd_buffer_size to 16kB.
2593 Bugzilla 879. Patch from Paul Fisher.
2595 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
2596 Heavily based on revision 40f9a89a from Simon Arlott's tree.
2599 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
2601 PP/05 Variable $av_failed, true if the AV scanner deferred.
2602 Bugzilla 1078. Patch from John Horne.
2604 PP/06 Stop make process more reliably on build failure.
2605 Bugzilla 1087. Patch from Heiko Schlittermann.
2607 PP/07 Make maildir_use_size_file an _expandable_ boolean.
2608 Bugzilla 1089. Patch from Heiko Schlittermann.
2610 PP/08 Handle ${run} returning more data than OS pipe buffer size.
2611 Bugzilla 1131. Patch from Holger Weiß.
2613 PP/09 Handle IPv6 addresses with SPF.
2614 Bugzilla 860. Patch from Wolfgang Breyha.
2616 PP/10 GnuTLS: support TLS 1.2 & 1.1.
2618 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
2621 PP/11 match_* no longer expand right-hand-side by default.
2622 New compile-time build option, EXPAND_LISTMATCH_RHS.
2623 New expansion conditions, "inlist", "inlisti".
2625 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
2627 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
2629 PP/14 fix log_write() format string regression from TF/03.
2630 Bugzilla 1152. Patch from Dmitry Isaikin.
2636 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
2638 PP/02 Harmonised TLS library version reporting; only show if debugging.
2639 Layout now matches that introduced for other libraries in 4.74 PP/03.
2641 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
2643 PP/04 New "dns_use_edns0" global option.
2645 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
2648 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
2649 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
2651 TK/01 Updated PolarSSL code to 0.14.2.
2652 Bugzilla 1097. Patch from Andreas Metzler.
2654 PP/07 Catch divide-by-zero in ${eval:...}.
2655 Fixes bugzilla 1102.
2657 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
2660 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
2661 format-string attack -- SECURITY: remote arbitrary code execution.
2663 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
2664 time unintentionally subject to list matching rules, letting the header
2665 cause arbitrary Exim lookups (of items which can occur in lists, *not*
2666 arbitrary string expansion). This allowed for information disclosure.
2668 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
2669 INT_MIN/-1 -- value coerced to INT_MAX.
2675 NM/01 Workaround for PCRE version dependency in version reporting
2678 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
2679 This fixes portability to compilers other than gcc, notably
2680 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
2682 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
2683 makefiles for portability to HP-UX and POSIX correctness.
2685 PP/01 Permit LOOKUP_foo enabling on the make command-line.
2686 Also via indented variable definition in the Makefile.
2687 (Debugging by Oliver Heesakkers).
2689 PP/02 Restore caching of spamd results with expanded spamd_address.
2690 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
2692 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
2693 Improves build reliability. Fix from: Frank Elsner
2695 NM/02 Fix wide character breakage in the rfc2047 coding
2696 Fixes bug 1064. Patch from Andrey N. Oktyabrski
2698 NM/03 Allow underscore in dnslist lookups
2699 Fixes bug 1026. Patch from Graeme Fowler
2701 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
2702 Code patches from Adam Ciarcinski of NetBSD.
2704 NM/04 Fixed exiqgrep to cope with mailq missing size issue
2707 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
2708 is logged, to avoid truncation. Patch from John Horne.
2710 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
2711 Patch from Jakob Hirsch.
2713 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
2714 SQL string expansion failure details.
2715 Patch from Andrey Oktyabrski.
2717 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
2718 Patch from Simon Arlott.
2720 PP/09 New lookups functionality failed to compile on old gcc which rejects
2721 extern declarations in function scope.
2722 Patch from Oliver Fleischmann
2724 PP/10 Use sig_atomic_t for flags set from signal handlers.
2725 Check getgroups() return and improve debugging.
2726 Fixed developed for diagnosis in bug 927 (which turned out to be
2729 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
2730 Patch from Mark Zealey.
2732 PP/12 Bugzilla 1056: Improved spamd server selection.
2733 Patch from Mark Zealey.
2735 PP/13 Bugzilla 1086: Deal with maildir quota file races.
2736 Based on patch from Heiko Schlittermann.
2738 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
2739 Patch from Uwe Doering, sign-off by Michael Haardt.
2741 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
2742 variable declaration deep within a block. Bug and patch from
2745 PP/15 lookups-Makefile IRIX compatibility coercion.
2747 PP/16 Make DISABLE_DKIM build knob functional.
2749 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
2750 Patch by Simon Arlott
2752 TF/03 Fix valgrind.h portability to C89 compilers that do not support
2753 variable argument macros. Our copy now differs from upstream.
2759 TF/01 Failure to get a lock on a hints database can have serious
2760 consequences so log it to the panic log.
2762 TF/02 Log LMTP confirmation messages in the same way as SMTP,
2763 controlled using the smtp_confirmation log selector.
2765 TF/03 Include the error message when we fail to unlink a spool file.
2767 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
2768 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
2769 for maintaining out-of-tree patches for some time.
2771 PP/01 Bugzilla 139: Documentation and portability issues.
2772 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
2773 Handle per-OS dynamic-module compilation flags.
2775 PP/02 Let /dev/null have normal permissions.
2776 The 4.73 fixes were a little too stringent and complained about the
2777 permissions on /dev/null. Exempt it from some checks.
2778 Reported by Andreas M. Kirchwitz.
2780 PP/03 Report version information for many libraries, including
2781 Exim version information for dynamically loaded libraries. Created
2782 version.h, now support a version extension string for distributors
2783 who patch heavily. Dynamic module ABI change.
2785 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
2786 privilege escalation vulnerability whereby the Exim run-time user
2787 can cause root to append content of the attacker's choosing to
2790 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
2793 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
2794 If dropping privileges for untrusted macros, we disabled normal logging
2795 on the basis that it would fail; for the Exim run-time user, this is not
2796 the case, and it resulted in successful deliveries going unlogged.
2797 Fixed. Reported by Andreas Metzler.
2803 PP/01 Date: & Message-Id: revert to normally being appended to a message,
2804 only prepend for the Resent-* case. Fixes regression introduced in
2805 Exim 4.70 by NM/22 for Bugzilla 607.
2807 PP/02 Include check_rfc2047_length in configure.default because we're seeing
2808 increasing numbers of administrators be bitten by this.
2810 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
2812 PP/03 Bugzilla 994: added openssl_options main configuration option.
2814 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
2816 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
2818 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
2820 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
2821 without a peer certificate, leading to a segfault because of an
2822 assumption that peers always have certificates. Be a little more
2823 paranoid. Problem reported by Martin Tscholak.
2825 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
2826 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
2827 NB: ClamAV planning to remove STREAM in "middle of 2010".
2828 CL also introduces -bmalware, various -d+acl logging additions and
2829 more caution in buffer sizes.
2831 PP/09 Implemented reverse_ip expansion operator.
2833 PP/10 Bugzilla 937: provide a "debug" ACL control.
2835 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
2837 PP/12 Bugzilla 973: Implement --version.
2839 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2841 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2843 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2845 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2846 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2847 ignore trailing whitespace.
2849 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2851 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2854 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2855 Notification from Dr Andrew Aitchison.
2857 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2858 ExtendedDetectionInfo response format.
2859 Notification from John Horne.
2861 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2864 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2865 XSL and documented dependency on system catalogs, with examples of how
2868 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2871 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2872 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2873 configuration file which is writeable by the Exim user or group.
2875 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2876 of configuration files to cover files specified with the -C option if
2877 they are going to be used with root privileges, not just the default
2880 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2881 option (effectively making it always true).
2883 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2884 files to be used while preserving root privileges.
2886 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2887 that rogue child processes cannot use them.
2889 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2890 run-time user, instead of root.
2892 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2893 Exim run-time user without dropping privileges.
2895 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2896 result string, instead of calling string_vformat() twice with the same
2899 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2900 for other users. Others should always drop root privileges if they use
2901 -C on the command line, even for a whitelisted configure file.
2903 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2905 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2911 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2912 $data_path, and $header_path variables; fixed documentation bugs and
2915 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2916 exipick to access non-standard spools, including the "frozen" queue
2919 NM/01 Bugzilla 965: Support mysql stored procedures.
2920 Patch from Alain Williams
2922 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2924 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2925 Patch from Andreas Metzler
2927 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2928 Patch from Kirill Miazine
2930 NM/05 Bugzilla 671: Added umask to procmail example.
2932 JJ/03 installed exipick 20100323.0, fixing doc bug
2934 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2935 directory. Notification and patch from Dan Rosenberg.
2937 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2939 TK/02 Improve log output when DKIM signing operation fails.
2941 MH/01 Treat the transport option dkim_domain as a colon separated
2942 list, not as a single string, and sign the message with each element,
2943 omitting multiple occurences of the same signer.
2945 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2946 Bugzilla 985, 986. Patch by Simon Arlott
2948 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2949 Patch by Simon Arlott
2951 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2952 MBX locking. Notification from Dan Rosenberg.
2958 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2960 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2962 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2964 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2966 NM/04 Bugzilla 915: Flag broken perl installation during build.
2972 TK/01 Added patch by Johannes Berg that expands the main option
2973 "spamd_address" if it starts with a dollar sign.
2975 TK/02 Write list of recipients to X-Envelope-Sender header when building
2976 the mbox-format spool file for content scanning (suggested by Jakob
2979 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2980 (http://www.dcc-servers.net/) support via dccifd. Activated by
2981 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2983 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2984 by Mark Daniel Reidel <mr@df.eu>.
2986 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2987 When building exim an external PCRE library is now needed -
2988 PCRE is a system library on the majority of modern systems.
2989 See entry on PCRE_LIBS in EDITME file.
2991 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2992 conversation. Added nologin parameter to request.
2993 Patch contributed by Kirill Miazine.
2995 TF/01 Do not log submission mode rewrites if they do not change the address.
2997 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2999 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
3000 log files in place. Contributed by Roberto Lima.
3002 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
3004 TF/03 Bugzilla 615: When checking the local_parts router precondition
3005 after a local_part_suffix or local_part_prefix option, Exim now
3006 does not use the address's named list lookup cache, since this
3007 contains cached lookups for the whole local part.
3009 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
3010 Robert Millan. Documentation is in experimental-spec.txt.
3012 TF/04 Bugzilla 668: Fix parallel build (make -j).
3014 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
3016 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
3017 Patch provided by Jan Srzednicki.
3019 TF/05 Leading white space used to be stripped from $spam_report which
3020 wrecked the formatting. Now it is preserved.
3022 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
3023 that they are available at delivery time.
3025 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
3027 TF/08 TLS error reporting now respects the incoming_interface and
3028 incoming_port log selectors.
3030 TF/09 Produce a more useful error message if an SMTP transport's hosts
3031 setting expands to an empty string.
3033 NM/06 Bugzilla 744: EXPN did not work under TLS.
3034 Patch provided by Phil Pennock.
3036 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
3037 Patch provided by Richard Godbee.
3039 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
3040 acl_smtp_notquit, added index entry.
3042 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
3043 Patch provided by Eugene Bujak.
3045 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
3046 accept(). Patch provided by Maxim Dounin.
3048 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
3049 Patch provided by Phil Pennock.
3051 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
3053 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
3054 Patch provided by Brad "anomie" Jorsch.
3056 NM/14 Bugzilla 622: Added timeout setting to transport filter.
3057 Patch provided by Dean Brooks.
3059 TK/05 Add native DKIM support (does not depend on external libraries).
3061 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
3062 Patch provided by Graeme Fowler.
3064 NM/16 Bugzilla 851: Documentation example syntax fix.
3066 NM/17 Changed NOTICE file to remove references to embedded PCRE.
3068 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
3071 NM/19 Bugzilla 745: TLS version reporting.
3072 Patch provided by Phil Pennock.
3074 NM/20 Bugzilla 167: bool: condition support.
3075 Patch provided by Phil Pennock.
3077 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
3078 clients. Patch provided by Phil Pennock.
3080 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
3081 Patch provided by Brad "anomie" Jorsch.
3083 NM/23 Bugzilla 687: Fix misparses in eximstats.
3084 Patch provided by Heiko Schlittermann.
3086 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
3087 Patch provided by Heiko Schlittermann.
3089 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
3090 plus update to original patch.
3092 NM/26 Bugzilla 799: Documentation correction for ratelimit.
3094 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
3095 Patch provided by David Brownlee.
3097 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
3099 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
3101 NM/30 Bugzilla 888: TLS documentation bugfixes.
3103 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
3105 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
3106 Unlike the original bugzilla I have changed all shell scripts in src tree.
3108 NM/33 Bugzilla 898: Transport filter timeout fix.
3109 Patch by Todd Rinaldo.
3111 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
3112 Patch by Serge Demonchaux.
3114 NM/35 Bugzilla 39: Base64 decode bug fixes.
3115 Patch by Jakob Hirsch.
3117 NM/36 Bugzilla 909: Correct connect() call in dcc code.
3119 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
3121 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
3123 NM/39 Bugzilla 911: Fixed MakeLinks build script.
3129 TK/01 Add preliminary DKIM support. Currently requires a forked version of
3130 ALT-N's libdkim that I have put here:
3131 http://duncanthrax.net/exim-experimental/
3133 Note to Michael Haardt: I had to rename some vars in sieve.c. They
3134 were called 'true' and it seems that C99 defines that as a reserved
3135 keyword to be used with 'bool' variable types. That means you could
3136 not include C99-style headers which use bools without triggering
3137 build errors in sieve.c.
3139 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
3140 as mailq or other aliases. Changed the --help handling significantly
3141 to do whats expected. exim_usage() emits usage/help information.
3143 SC/01 Added the -bylocaldomain option to eximstats.
3145 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
3147 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
3149 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
3155 PH/01 Another patch from the Sieve maintainer.
3157 PH/02 When an IPv6 address is converted to a string for single-key lookup
3158 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
3159 dots are used instead of colons so that keys in lsearch files need not
3160 contain colons. This was done some time before quoting was made available
3161 in lsearch files. However, iplsearch files do require colons in IPv6 keys
3162 (notated using the quote facility) so as to distinguish them from IPv4
3163 keys. This meant that lookups for IP addresses in host lists did not work
3164 for iplsearch lookups.
3166 This has been fixed by arranging for IPv6 addresses to be expressed with
3167 colons if the lookup type is iplsearch. This is not incompatible, because
3168 previously such lookups could never work.
3170 The situation is now rather anomalous, since one *can* have colons in
3171 ordinary lsearch keys. However, making the change in all cases is
3172 incompatible and would probably break a number of configurations.
3174 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
3177 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
3178 conversion specification without a maximum field width, thereby enabling
3179 a rogue spamd server to cause a buffer overflow. While nobody in their
3180 right mind would setup Exim to query an untrusted spamd server, an
3181 attacker that gains access to a server running spamd could potentially
3182 exploit this vulnerability to run arbitrary code as the Exim user.
3184 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
3185 $primary_hostname instead of what libspf2 thinks the hosts name is.
3187 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
3188 a directory entry by the name of the lookup key. Previously, if a
3189 symlink pointed to a non-existing file or a file in a directory that
3190 Exim lacked permissions to read, a lookup for a key matching that
3191 symlink would fail. Now it is enough that a matching directory entry
3192 exists, symlink or not. (Bugzilla 503.)
3194 PH/03 The body_linecount and body_zerocount variables are now exported in the
3197 PH/04 Added the $dnslist_matched variable.
3199 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
3200 This means they are set thereafter only if the connection becomes
3203 PH/06 Added the client_condition to authenticators so that some can be skipped
3204 by clients under certain conditions.
3206 PH/07 The error message for a badly-placed control=no_multiline_responses left
3207 "_responses" off the end of the name.
3209 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
3211 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
3212 (without spaces) instead of just copying the configuration text.
3214 PH/10 Added the /noupdate option to the ratelimit ACL condition.
3216 PH/11 Added $max_received_linelength.
3218 PH/12 Added +ignore_defer and +include_defer to host lists.
3220 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
3221 way in which PCRE > 7.0 is built.
3223 PH/14 Implemented queue_only_load_latch.
3225 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
3226 MAIL command. The effect was to mangle the value on 64-bit systems.
3228 PH/16 Another patch from the Sieve maintainer.
3230 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
3232 PH/18 If a system quota error occurred while trying to create the file for
3233 a maildir delivery, the message "Mailbox is full" was not appended to the
3234 bounce if the delivery eventually timed out. Change 4.67/27 below applied
3235 only to a quota excession during the actual writing of the file.
3237 PH/19 It seems that peer DN values may contain newlines (and other non-printing
3238 characters?) which causes problems in log lines. The DN values are now
3239 passed through string_printing() before being added to log lines.
3241 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
3242 and InterBase are left for another time.)
3244 PH/21 Added message_body_newlines option.
3246 PH/22 Guard against possible overflow in moan_check_errorcopy().
3248 PH/23 POSIX allows open() to be a macro; guard against that.
3250 PH/24 If the recipient of an error message contained an @ in the local part
3251 (suitably quoted, of course), incorrect values were put in $domain and
3252 $local_part during the evaluation of errors_copy.
3258 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
3259 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
3262 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
3263 issue a MAIL command.
3265 PH/02 In an ACL statement such as
3267 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
3269 if a client was not listed at all, or was listed with a value other than
3270 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
3271 the condition was not true (as it should be), so access was not denied.
3272 The bug was that the ! inversion was incorrectly passed on to the second
3273 item. This has been fixed.
3275 PH/03 Added additional dnslists conditions == and =& which are different from
3276 = and & when the dns lookup returns more than one IP address.
3278 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
3279 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
3281 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
3282 FSYNC, which compiles an option called disable_fsync that allows for
3283 bypassing fsync(). The documentation is heavily laced with warnings.
3285 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
3287 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
3288 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
3289 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
3290 including adding "make clean"; (3) Added -fPIC when compiling the test
3291 dynamically loaded module, to get rid of a warning.
3293 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
3294 message fails, move_frozen_messages = true and ignore_bounce_errors_after
3295 = 0s. The bug is otherwise harmless.
3297 PH/07 There was a bug in the dovecot authenticator such that the value of
3298 $auth1 could be overwritten, and so not correctly preserved, after a
3299 successful authentication. This usually meant that the value preserved by
3300 the server_setid option was incorrect.
3302 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
3304 PH/09 Installed PCRE release 7.0.
3306 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
3307 run for batched SMTP input. It is now run at the start of every message
3308 in the batch. While fixing this I discovered that the process information
3309 (output by running exiwhat) was not always getting set for -bs and -bS
3310 input. This is fixed, and it now also says "batched" for BSMTP.
3312 PH/11 Added control=no_pipelining.
3314 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
3315 patch, slightly modified), and move the expansion of helo_data till after
3316 the connection is made in the smtp transport (so it can use these
3319 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
3321 PH/14 Added log_selector = +pid.
3323 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
3325 PH/16 Add ${if forany and ${if forall.
3327 PH/17 Added dsn_from option to vary the From: line in DSNs.
3329 PH/18 Flush SMTP output before performing a callout, unless control =
3330 no_callout_flush is set.
3332 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
3333 was true (the default) a successful delivery failed to delete the retry
3334 item, thus causing premature timeout of the address. The bug is now
3337 PH/20 Added hosts_avoid_pipelining to the smtp transport.
3339 PH/21 Long custom messages for fakedefer and fakereject are now split up
3340 into multiline responses in the same way that messages for "deny" and
3341 other ACL rejections are.
3343 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
3344 with slight modification.
3346 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
3347 draft, changing the syntax and factoring some duplicate code".
3349 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
3350 for deliveries of the second and subsequent messages over the same SMTP
3353 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
3354 ${reduce, with only minor "tidies".
3356 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
3358 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
3359 expansion side effects.
3361 PH/27 When a message times out after an over-quota error from an Exim-imposed
3362 quota, the bounce message says "mailbox is full". This message was not
3363 being given when it was a system quota that was exceeded. It now should
3366 MH/03 Made $recipients available in local_scan(). local_scan() already has
3367 better access to the recipient list through recipients_list[], but
3368 $recipients can be useful in postmaster-provided expansion strings.
3370 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
3371 in the case of a MAIL command with additional options following the
3372 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
3373 were accidentally chopped off.
3375 PH/29 SMTP synchronization checks are implemented when a command is read -
3376 there is a check that no more input is waiting when there shouldn't be
3377 any. However, for some commands, a delay in an ACL can mean that it is
3378 some time before the response is written. In this time, more input might
3379 arrive, invalidly. So now there are extra checks after an ACL has run for
3380 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
3381 pipelining has not been advertised.
3383 PH/30 MH's patch to allow iscntrl() characters to be list separators.
3385 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
3386 returned in the SMTP response when smtp_return_error_details was false.
3387 This has been fixed.
3389 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
3390 instead of the C I/O that was originally supplied, because problems were
3391 reported on Solaris.
3393 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
3394 Exim which did not show up earlier: it was assuming that a call to
3395 SSL_CTX_set_info_callback() might give an error value. In fact, there is
3396 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
3397 was a macro that became an assignment, so it seemed to work. This has
3398 changed to a proper function call with a void return, hence the compile
3399 error. Exim's code has been fixed.
3401 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
3404 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
3406 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
3408 PH/37 If a message is not accepted after it has had an id assigned (e.g.
3409 because it turns out to be too big or there is a timeout) there is no
3410 "Completed" line in the log. When some messages of this type were
3411 selected by exigrep, they were listed as "not completed". Others were
3412 picked up by some special patterns. I have improved the selection
3413 criteria to be more general.
3415 PH/38 The host_find_failed option in the manualroute router can now be set
3416 to "ignore", to completely ignore a host whose IP address cannot be
3417 found. If all hosts are ignored, the behaviour is controlled by the new
3418 host_all_ignored option.
3420 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
3421 homing or because of multiple MX records with /mx) generated more than
3422 one IP address, and the following item turned out to be the local host,
3423 all the secondary addresses of the first item were incorrectly removed
3424 from the list, along with the local host and any following hosts (which
3425 is what is supposed to happen).
3427 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
3428 whoever called Exim into the -H file. In the case of the daemon it was
3429 behaving confusingly. When first started, it used values for whoever
3430 started the daemon, but after a SIGHUP it used the Exim user (because it
3431 calls itself on a restart). I have changed the code so that it now always
3434 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
3435 message are rejected with the same error (e.g. no authentication or bad
3436 sender address), and a DATA command is nevertheless sent (as can happen
3437 with PIPELINING or a stupid MUA), the error message that was given to the
3438 RCPT commands is included in the rejection of the DATA command. This is
3439 intended to be helpful for MUAs that show only the final error to their
3442 PH/42 Another patch from the Sieve maintainer.
3444 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
3445 Eximstats - Fixed some broken HTML links and added missing column headers
3447 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
3448 columns for Rejects, Temp Rejects, Ham, and Spam rows.
3450 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
3452 PH/43 Yet another patch from the Sieve maintainer.
3454 PH/44 I found a way to check for a TCP/IP connection going away before sending
3455 the response to the final '.' that terminates a message, but only in the
3456 case where the client has not sent further data following the '.'
3457 (unfortunately, this is allowed). However, in many cases there won't be
3458 any further data because there won't be any more messages to send. A call
3459 to select() can be used: if it shows that the input is "ready", there is
3460 either input waiting, or the socket has been closed. An attempt to read
3461 the next input character can distinguish the two cases. Previously, Exim
3462 would have sent an OK response which the client would never have see.
3463 This could lead to message repetition. This fix should cure that, at
3464 least in a lot of common cases.
3466 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
3467 advertised in response to EHLO.
3473 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
3474 fixed by 4.65/MH/01 (is this a record?) are fixed:
3476 (i) An empty string was always treated as zero by the numeric comparison
3477 operators. This behaviour has been restored.
3479 (ii) It is documented that the numeric comparison operators always treat
3480 their arguments as decimal numbers. This was broken in that numbers
3481 starting with 0 were being interpreted as octal.
3483 While fixing these problems I realized that there was another issue that
3484 hadn't been noticed. Values of message_size_limit (both the global option
3485 and the transport option) were treated as octal if they started with 0.
3486 The documentation was vague. These values are now always treated as
3487 decimal, and I will make that clear in the documentation.
3493 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
3494 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
3497 MH/01 Don't check that the operands of numeric comparison operators are
3498 integers when their expansion is in "skipping" mode (fixes bug
3499 introduced by 4.64-PH/07).
3501 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
3502 child addresses, Exim now panics and dies. Previously, because the count
3503 is held in a short int, deliveries were likely to be lost. As such a
3504 large number of recipients for a single message is ridiculous
3505 (performance will be very, very poor), I have chosen to impose a limit
3506 rather than extend the field.
3512 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
3513 leftover -K file (the existence of which was triggered by #402).
3514 While we were at it, introduced process PID as part of the -K
3515 filename. This should rule out race conditions when creating
3518 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
3519 processing considerably. Previous code took too long for large mails,
3520 triggering a timeout which in turn triggers #401.
3522 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
3523 in the DK code in transports.c. sendfile() is not really portable,
3524 hence the _LINUX specificness.
3526 TF/01 In the add_headers option to the mail command in an Exim filter,
3527 there was a bug that Exim would claim a syntax error in any
3528 header after the first one which had an odd number of characters
3531 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
3532 callout verification, Exim cached a "reject" for the entire domain. This
3533 is correct for most verifications, but it is not correct for a recipient
3534 verification with use_sender or use_postmaster set, because in that case
3535 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
3536 case of MAIL FROM:<> rejection from other early rejections (e.g.
3537 rejection of HELO). When verifying a recipient using a non-null MAIL
3538 address, the cache is ignored if it shows MAIL FROM:<> rejection.
3539 Whatever the result of the callout, the value of the domain cache is
3540 left unchanged (for any other kind of callout, getting as far as trying
3541 RCPT means that the domain itself is ok).
3543 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
3546 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
3547 manifest itself as EPIPE rather than ECONNECT. When tidying away a
3548 session, the daemon ignores ECONNECT errors and logs others; it now
3549 ignores EPIPE as well.
3551 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
3552 (quoted-printable decoding).
3554 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
3555 later the small subsequent patch to fix an introduced bug.
3557 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
3559 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
3561 PH/08 An error is now given if message_size_limit is specified negative.
3563 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
3564 to be given (somewhat) arbitrary names.
3566 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
3569 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
3570 miscellaneous code fixes
3572 PH/10 Added the log_reject_target ACL modifier to specify where to log
3575 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
3576 hostname. This is wrong, because it relates to the incoming message (and
3577 probably the interface on which it is arriving) and not to the outgoing
3578 callout (which could be using a different interface). This has been
3579 changed to use the value of the helo_data option from the smtp transport
3580 instead - this is what is used when a message is actually being sent. If
3581 there is no remote transport (possible with a router that sets up host
3582 addresses), $smtp_active_hostname is used.
3584 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
3585 tweaks were necessary in order to get it to work (see also 21 below):
3586 (a) The code assumed that strncpy() returns a negative number on buffer
3587 overflow, which isn't the case. Replaced with Exim's string_format()
3589 (b) There were several signed/unsigned issues. I just did the minimum
3590 hacking in of casts. There is scope for a larger refactoring.
3591 (c) The code used strcasecmp() which is not a standard C function.
3592 Replaced with Exim's strcmpic() function.
3593 (d) The code set only $1; it now sets $auth1 as well.
3594 (e) A simple test gave the error "authentication client didn't specify
3595 service in request". It would seem that Dovecot has changed its
3596 interface. Fortunately there's a specification; I followed it and
3597 changed what the client sends and it appears to be working now.
3599 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
3602 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
3603 address A is aliased to B and C, where B exists and C does not. Without
3604 -v the output is "A verified" because verification stops after a
3605 successful redirection if more than one address is generated. However,
3606 with -v the child addresses are also verified. Exim was outputting "A
3607 failed to verify" and then showing the successful verification for C,
3608 with its parentage. It now outputs "B failed to verify", showing B's
3609 parentage before showing the successful verification of C.
3611 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
3612 look up a TXT record in a specific list after matching in a combined
3615 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
3616 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
3617 they consult the DNS. I had assumed they would set it the way they
3618 wanted; and indeed my experiments on Linux seem to show that in some
3619 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
3620 To be on the safe side, however, I have now made the interface to
3621 host_find_byname() similar to host_find_bydns(), with an argument
3622 containing the DNS resolver options. The host_find_byname() function now
3623 sets these options at its start, just as host_find_bydns() does. The smtp
3624 transport options dns_qualify_single and dns_search_parents are passed to
3625 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
3626 of host_find_byname() use the default settings of RES_DEFNAMES
3627 (qualify_single) but not RES_DNSRCH (search_parents).
3629 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
3630 spool_read_header() do less string testing, by means of a preliminary
3631 switch on the second character of optional "-foo" lines. (This is
3632 overdue, caused by the large number of possibilities that now exist.
3633 Originally there were few.) While I was there, I also converted the
3634 str(n)cmp tests so they don't re-test the leading "-" and the first
3635 character, in the hope this might squeeze out yet more improvement.
3637 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
3638 flag allowing group syntax was set by the header_syntax check but not
3639 turned off, possible causing trouble later; (2) The flag was not being
3640 set at all for the header_verify test, causing "group"-style headers to
3641 be rejected. I have now set it in this case, and also caused header_
3642 verify to ignore an empty address taken from a group. While doing this, I
3643 came across some other cases where the code for allowing group syntax
3644 while scanning a header line wasn't quite right (mostly, not resetting
3645 the flag correctly in the right place). These bugs could have caused
3646 trouble for malformed header lines. I hope it is now all correct.
3648 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
3649 with the "reply" argument non-NULL. The code, however (which originally
3650 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
3651 but it didn't always do it. This confused somebody who was copying the
3652 code for some other use. I have removed all the tests.
3654 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
3655 feature that was used to support insecure browsers during the U.S. crypto
3656 embargo. It requires special client support, and Exim is probably the
3657 only MTA that supported it -- and would never use it because real RSA is
3658 always available. This code has been removed, because it had the bad
3659 effect of slowing Exim down by computing (never used) parameters for the
3660 RSA_EXPORT functionality.
3662 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
3663 authenticator to fail if there's a tab character in the incoming data
3664 (there should never be unless someone is messing about, as it's supposed
3665 to be base64-encoded). Also added, on Timo's advice, the "secured" option
3666 if the connection is using TLS or if the remote IP is the same as the
3667 local IP, and the "valid-client-cert option" if a client certificate has
3670 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
3671 authenticators. This can be used for authorization after authentication
3672 succeeds. (In the case of plaintext, it servers for both authentication
3675 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
3676 if any retry times were supplied.
3678 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
3679 connection, where there is no client IP address to check. In this
3680 situation, the verify now always succeeds.
3682 PH/25 Applied John Jetmore's -Mset patch.
3684 PH/26 Added -bem to be like -Mset, but loading a message from a file.
3686 PH/27 In a string expansion for a processed (not raw) header when multiple
3687 headers of the same name were present, leading whitespace was being
3688 removed from all of them, but trailing whitespace was being removed only
3689 from the last one. Now trailing whitespace is removed from each header
3690 before concatenation. Completely empty headers in a concatenation (as
3691 before) are ignored.
3693 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
3694 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
3696 PH/29 [Removed. This was a change that I later backed out, and forgot to
3697 correct the ChangeLog entry (that I had efficiently created) before
3698 committing the later change.]
3700 PH/30 Exim was sometimes attempting to deliver messages that had suffered
3701 address errors (4xx response to RCPT) over the same connection as other
3702 messages routed to the same hosts. Such deliveries are always "forced",
3703 so retry times are not inspected. This resulted in far too many retries
3704 for the affected addresses. The effect occurred only when there were more
3705 hosts than the hosts_max_try setting in the smtp transport when it had
3706 the 4xx errors. Those hosts that it had tried were not added to the list
3707 of hosts for which the message was waiting, so if all were tried, there
3708 was no problem. Two fixes have been applied:
3710 (i) If there are any address or message errors in an SMTP delivery, none
3711 of the hosts (tried or untried) are now added to the list of hosts
3712 for which the message is waiting, so the message should not be a
3713 candidate for sending over the same connection that was used for a
3714 successful delivery of some other message. This seems entirely
3715 reasonable: after all the message is NOT "waiting for some host".
3716 This is so "obvious" that I'm not sure why it wasn't done
3717 previously. Hope I haven't missed anything, but it can't do any
3718 harm, as the worst effect is to miss an optimization.
3720 (ii) If, despite (i), such a delivery is accidentally attempted, the
3721 routing retry time is respected, so at least it doesn't keep
3722 hammering the server.
3724 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
3725 in ${readsocket because some servers need this prod.
3727 PH/32 Added some extra debug output when updating a wait-xxx database.
3729 PH/33 The hint "could be header name not terminated by colon", which has been
3730 given for certain expansion errors for a long time, was not being given
3731 for the ${if def:h_colon_omitted{... case.
3733 PH/34 The spec says: "With one important exception, whenever a domain list is
3734 being scanned, $domain contains the subject domain." There was at least
3735 one case where this was not true.
3737 PH/35 The error "getsockname() failed: connection reset by peer" was being
3738 written to the panic log as well as the main log, but it isn't really
3739 panic-worthy as it just means the connection died rather early on. I have
3740 removed the panic log writing for the ECONNRESET error when getsockname()
3743 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
3744 runs only) independently of the message's sender address. This meant
3745 that, if the 4xx error was in fact related to the sender, a different
3746 message to the same recipient with a different sender could confuse
3747 things. In particular, this can happen when sending to a greylisting
3748 server, but other circumstances could also provoke similar problems.
3749 I have changed the default so that the retry time for these errors is now
3750 based a combination of the sender and recipient addresses. This change
3751 can be overridden by setting address_retry_include_sender=false in the
3754 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
3755 remote server are returned as part of bounce messages. This was not
3756 happening for LMTP over a pipe (the lmtp transport), but now it is the
3757 same for both kinds of LMTP.
3759 PH/38 Despite being documented as not happening, Exim was rewriting addresses
3760 in header lines that were in fact CNAMEs. This is no longer the case.
3762 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
3763 and queue runs started by the daemon processed all messages. This has
3764 been fixed so that -R and -S can now usefully be given with -q<time>.
3766 PH/40 Import PCRE release 6.7 (fixes some bugs).
3768 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
3770 PH/42 Give an error if -q is specified more than once.
3772 PH/43 Renamed the variables $interface_address and $interface_port as
3773 $received_ip_address and $received_port, to make it clear that these
3774 values apply to message reception, and not to the outgoing interface when
3775 a message is delivered. (The old names remain recognized, of course.)
3777 PH/44 There was no timeout on the connect() call when using a Unix domain
3778 socket in the ${readsocket expansion. There now is.
3780 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
3781 be meaningful with "accept".
3783 SC/01 Eximstats V1.43
3784 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
3786 SC/02 Eximstats V1.44
3787 Use a glob alias rather than an array ref in the generated
3788 parser. This improves both readability and performance.
3790 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
3791 Collect SpamAssassin and rejection statistics.
3792 Don't display local sender or destination tables unless
3793 there is data to show.
3794 Added average volumes into the top table text output.
3796 SC/04 Eximstats V1.46
3797 Collect data on the number of addresses (recipients)
3798 as well as the number of messages.
3800 SC/05 Eximstats V1.47
3801 Added 'Message too big' to the list of mail rejection
3802 reasons (thanks to Marco Gaiarin).
3804 SC/06 Eximstats V1.48
3805 Mainlog lines which have GMT offsets and are too short to
3806 have a flag are now skipped.
3808 SC/07 Eximstats V1.49 (Alain Williams)
3809 Added the -emptyok flag.
3811 SC/08 Eximstats V1.50
3812 Fixes for obtaining the IP address from reject messages.
3814 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
3815 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
3816 whitespace changes from 4.64-PH/27
3818 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
3821 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
3822 are found, allow negative numbers in numeric criteria)
3824 JJ/06 exipick.20061117.2, added new $message_body_missing variable
3826 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
3827 to match changes made in 4.64-PH/43
3829 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
3831 PH/47 Put in an explicit test for a DNS lookup of an address record where the
3832 "domain" is actually an IP address, and force a failure. This locks out
3833 those revolvers/nameservers that support "A-for-A" lookups, in
3834 contravention of the specifications.
3836 PH/48 When a host name was looked up from an IP address, and the subsequent
3837 forward lookup of the name timed out, the host name was left in
3838 $sender_host_name, contrary to the specification.
3840 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3841 restricted to single-key lookups, Exim was not diagnosing an error if
3842 * or *@ was used with a query-style lookup.
3844 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3846 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3847 long ago, but noone interested enough thought of it. Let's just say that
3848 the "1.1" means that there are some new functions that weren't there at
3849 some point in the past.
3851 PH/51 Error processing for expansion failure of helo_data from an smtp
3852 transport during callout processing was broken.
3854 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3855 tested/used via the -bh/-bhc/-bs options.
3857 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3858 bug, fixed in subsequent PCRE releases).
3860 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3861 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3863 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3869 SC/01 Use a glob alias rather than an array ref in eximstats generated
3870 parser. This improves both readability and performance.
3872 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3873 Don't display local sender or destination tables in eximstats unless
3874 there is data to show.
3875 Added average volumes into the eximstats top table text output.
3877 SC/03 Collect data on the number of addresses (recipients) as well
3878 as the number of messages in eximstats.
3880 TF/01 Correct an error in the documentation for the redirect router. Exim
3881 does (usually) call initgroups() when daemonizing.
3883 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3884 with consistent privilege compared to when running as a daemon.
3886 TF/03 Note in the spec that $authenticated_id is not set for local
3887 submissions from trusted users.
3889 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3890 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3892 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3893 by adding some example configuration directives to the default
3894 configuration file. A little bit of work is required to uncomment the
3895 directives and define how usernames and passwords are checked, but
3896 there is now a framework to start from.
3898 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3899 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3900 without this. I don't know how relevant this is to other LDAP libraries.
3902 PH/02 Add the verb name to the "unknown ACL verb" error.
3904 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3906 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3908 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3909 directory not expanded when it should be if an expanded home directory
3910 was set for the address (which is overridden by the transport).
3912 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3915 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3916 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3917 because it is too late at that time, and has no effect.
3919 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3920 security issue with \' (bugzilla #107). I could not use the
3921 PQescapeStringConn() function, because it needs a PGconn value as one of
3924 PH/08 When testing addresses using -bt, indicate those final addresses that
3925 are duplicates that would not cause an additional delivery. At least one
3926 person was confused, thinking that -bt output corresponded to deliveries.
3927 (Suppressing duplicates isn't a good idea as you lose the information
3928 about possibly different redirections that led to the duplicates.)
3930 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3931 systems where poll() doesn't work, in particular OS X.
3933 PH/10 Added more information to debugging output for retry time not reached.
3935 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3936 operations in malware.c.
3938 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3941 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3942 syslog_duplication set false, log lines that would normally be written
3943 both the the main log and to the reject log were not written to syslog at
3946 PH/14 In the default configuration, change the use of "message" in ACL warn
3947 statements to "add_header".
3949 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3950 not followed by a command (e.g. "seen endif").
3952 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3953 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3956 PH/17 Added extra conditions to the default value of delay_warning_condition
3960 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3961 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3962 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3965 The Auto-Submitted: and various List- headers are standardised, whereas I
3966 don't think Precedence: ever was.
3968 PH/18 Refactored debugging code in route_finduser() to show more information,
3969 in particular, the error code if getpwnam() issues one.
3971 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3972 This is apparently needed in addition to the PH/07 change above to avoid
3973 any possible encoding problems.
3975 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3976 but not after initializing Perl.
3978 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3979 output them only if debugging. By default they are written stderr,
3980 apparently, which is not desirable.
3982 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3985 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3988 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3990 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3991 authenticated or an ident call has been made. Suppress the default
3992 values for $authenticated_id and $authenticated_sender (but permit -oMai
3993 and -oMas) when testing with -bh.
3995 PH/24 Re-jigged the order of the tests in the default configuration so that the
3996 tests for valid domains and recipients precede the DNS black list and CSA
3997 tests, on the grounds that those ones are more expensive.
3999 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
4000 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
4001 This bug exists in every version of Exim that I still have, right back to
4004 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
4005 However, an attempt to turn on case-sensitivity in a regex key by
4006 including (?-i) didn't work because the subject string was already
4007 lowercased, and the effects were non-intuitive. It turns out that a
4008 one-line patch can be used to allow (?-i) to work as expected.
4014 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
4015 other effects) broke the use of negated acl sub-conditions.
4017 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
4020 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
4021 "Deny" causes Exim to reject the incoming connection with a 554 error.
4022 Unfortunately, if there is a major crisis, such as a disk failure,
4023 tcp-wrappers gives "deny", whereas what one would like would be some
4024 kind of temporary error. A kludge has been added to help with this.
4025 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
4026 554 error is used if errno is still zero or contains ENOENT (which occurs
4027 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
4030 PH/03 Add -lutil to the default FreeBSD LIBS setting.
4032 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
4033 errors. Otherwise a message that provokes a temporary error (when other
4034 messages do not) can cause a whole host to time out.
4036 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
4037 addresses were routed directly to files or pipes from a redirect router.
4038 File deliveries just didn't batch; pipe deliveries might have suffered
4041 PH/06 A failure to get a lock for a hints database would erroneously always say
4042 "Failed to get write lock", even when it was really a read lock.
4044 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
4045 of 0600. This has been changed to use the value of the lockfile_mode
4046 option (which defaults to 0600).
4048 PH/08 Applied small patch from the Sieve maintainer.
4050 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
4051 folder from quota calculations, a direct delivery into this folder messed
4052 up the contents of the maildirsize file. This was because the regex was
4053 used only to exclude .Trash (or whatever) when the size of the mailbox
4054 was calculated. There was no check that a delivery was happening into an
4055 excluded directory. This bug has been fixed by ignoring all quota
4056 processing for deliveries into excluded directories.
4058 PH/10 Added the maildirfolder_create_regex option to appendfile.
4064 PH/01 The code for finding all the local interface addresses on a FreeBSD
4065 system running IPv6 was broken. This may well have applied to all BSD
4066 systems, as well as to others that have similar system calls. The broken
4067 code found IPv4 interfaces correctly, but gave incorrect values for the
4068 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
4069 that it would not match correctly against @[] and not recognize the IPv6
4072 PH/02 The ipliteral router was not recognizing addresses of the form user@
4073 [ipv6:....] because it didn't know about the "ipv6:" prefix.
4075 PH/03 Added disable_ipv6.
4077 PH/04 Changed $reply_address to use the raw form of the headers instead of the
4078 decoded form, because it is most often used to construct To: headers
4079 lines in autoreplies, and the decoded form may well be syntactically
4080 invalid. However, $reply_address has leading white space removed, and all
4081 newlines turned into spaces so that the autoreply transport does not
4084 PH/05 If group was specified without a user on a router, and no group or user
4085 was specified on a transport, the group from the router was ignored.
4087 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
4088 for visible compile-time settings that can be used to change these
4089 numbers, for those that want even more. Backwards compatibility with old
4090 spool files has been maintained. However, going back to a previous Exim
4091 release will lost any variables that are in spool files.
4093 PH/07 Two small changes when running in the test harness: increase delay when
4094 passing a TCP/IP connection to a new process, in case the original
4095 process has to generate a bounce, and remove special handling of
4096 127.0.0.2 (sic), which is no longer necessary.
4098 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
4099 be the same on different OS.
4101 PH/09 Moved a debug statement in filter processing to avoid a race problem when
4104 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
4105 whether --show-vars was specified or not
4107 JJ/02 exipick: Added support for new ACL variable spool format introduced
4110 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
4111 syntactically invalid From: or Reply-to: line, and a filter used this to
4112 generate an autoreply, and therefore failed to obtain an address for the
4113 autoreply, Exim could try to deliver to a non-existent relative file
4114 name, causing unrelated and misleading errors. What now happens is that
4115 it logs this as a hard delivery error, but does not attempt to create a
4118 PH/11 The exinext utility has a -C option for testing purposes, but although
4119 the given file was scanned by exinext itself; it wasn't being passed on
4120 when Exim was called.
4122 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
4123 an end-of-file indication when reading a command response.
4125 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
4126 compiled. In many other places in Exim, IPv6 addresses are always
4127 recognized, so I have changed this. It also means that IPv4 domain
4128 literals of the form [IPV4:n.n.n.n] are now always recognized.
4130 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
4131 used if the router is not running as root, for example, when verifying at
4132 ACL time, or when using -bh. The debugging output from this situation was
4133 non-existent - all you got was a failure to exec. I have made two
4136 (a) Failures to set uid/gid, the current directory, or a process leader
4137 in a subprocess such as that created by queryprogram now generate
4138 suitable debugging output when -d is set.
4140 (b) The queryprogram router detects when it is not running as root,
4141 outputs suitable debugging information if -d is set, and then runs
4142 the subprocess without attempting to change uid/gid.
4144 PH/15 Minor change to Makefile for building test_host (undocumented testing
4147 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
4148 additional section of a DNS packet that returns MX or SRV records.
4149 Instead, it always explicitly searches for A/AAAA records. This avoids
4150 major problems that occur when a DNS server includes only records of one
4151 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
4152 fixed another bug: if SRV records were looked up and the corresponding
4153 address records were *not* found in the additional section, the port
4154 values from the SRV records were lost.
4156 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
4157 using the correct key (the original address) when searching the retry
4158 rules in order to find which one to use for generating the retry hint.
4160 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
4161 adding the default one. Similarly, if it contains a Reply-To: header, the
4162 errors_reply_to option, if set, is not used.
4164 PH/19 When calculating a retry time, Exim used to measure the "time since
4165 failure" by looking at the "first failed" field in the retry record. Now
4166 it does not use this if it is later than than the arrival time of the
4167 message. Instead it uses the arrival time. This makes for better
4168 behaviour in cases where some deliveries succeed, thus re-setting the
4169 "first failed" field. An example is a quota failure for a huge message
4170 when small messages continue to be delivered. Without this change, the
4171 "time since failure" will always be short, possible causing more frequent
4172 delivery attempts for the huge message than are intended.
4173 [Note: This change was subsequently modified - see PH/04 for 4.62.]
4175 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
4176 $1, $2, $3) because the numerical variables can be reset during some
4177 expansion items (e.g. "match"), thereby losing the authentication data.
4179 PH/21 Make -bV show the size of off_t variables so that the test suite can
4180 decide whether to run tests for quotas > 2G.
4182 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
4183 mailbox_size, and mailbox_filecount in the appendfile transport. If a
4184 filecount value is greater than 2G or if a quota value is greater than 2G
4185 on a system where the size of off_t is not greater than 4, a panic error
4188 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
4189 never match. The debug and -bh output now contains an explicit error
4190 message indicating a malformed IPv4 address or mask.
4192 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
4193 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
4194 PH/23 above applies.
4196 PH/25 Do not write to syslog when running in the test harness. The only
4197 occasion when this arises is a failure to open the main or panic logs
4198 (for which there is an explicit test).
4200 PH/26 Added the /no_tell option to "control=freeze".
4202 PH/27 If a host name lookup failed very early in a connection, for example, if
4203 the IP address matched host_lookup and the reverse lookup yielded a name
4204 that did not have a forward lookup, an error message of the form "no IP
4205 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
4206 could be logged. Now it outputs the IP address instead of "NULL".
4208 PH/28 An enabling patch from MH: add new function child_open_exim2() which
4209 allows the sender and the authenticated sender to be set when
4210 submitting a message from within Exim. Since child_open_exim() is
4211 documented for local_scan(), the new function should be too.
4213 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
4214 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
4215 results in an empty string is now treated as unset.
4217 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
4219 PH/31 Added sender_verify_fail logging option.
4221 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
4222 needed by RFC 822 but not by RFC 2822 was commented out. I have now
4223 tidied the source and removed it altogether.
4225 PH/33 When a queue run was abandoned because the load average was too high, a
4226 log line was always written; now it is written only if the queue_run log
4227 selector is set. In addition, the log line for abandonment now contains
4228 information about the queue run such as the pid. This is always present
4229 in "start" and "stop" lines but was omitted from the "abandon" line.
4231 PH/34 Omit spaces between a header name and the colon in the error message that
4232 is given when verify = headers_syntax fails (if there are lots of them,
4233 the message gets confusing).
4235 PH/35 Change the default for dns_check_names_pattern to allow slashes within
4236 names, as there are now some PTR records that contain slashes. This check
4237 is only to protect against broken name servers that fall over on strange
4238 characters, so the fact that it applies to all lookups doesn't matter.
4240 PH/36 Now that the new test suite is complete, we can remove some of the
4241 special code in Exim that was needed for the old test suite. For example,
4242 sorting DNS records because real resolvers return them in an arbitrary
4243 order. The new test suite's fake resolver always returns records in the
4246 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
4247 bounces) except when queue_only is set, to avoid logging races between
4248 the different processes.
4250 PH/38 Panic-die if .include specifies a non-absolute path.
4252 PH/39 A tweak to the "H" retry rule from its user.
4254 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
4255 a label. They prevented compilation on older perls.
4257 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
4258 a warning to be raised on newish perls.
4260 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
4261 on queue. Changes to match documented behaviour of showing count of
4262 messages matching specified criteria.
4264 PH/40 Changed the default ident timeout from 30s to 5s.
4266 PH/41 Added support for the use of login_cap features, on those BSD systems
4267 that have them, for controlling the resources used by pipe deliveries.
4269 PH/42 The content-scanning code uses fopen() to create files in which to put
4270 message data. Previously it was not paying any attention to the mode of
4271 the files. Exim runs with umask(0) because the rest of the code creates
4272 files with open(), and sets the required mode explicitly. Thus, these
4273 files were ending up world-writeable. This was not a big issue, because,
4274 being within the spool directory, they were not world-accessible. I have
4275 created a function called modefopen, which takes an additional mode
4276 argument. It sets umask(777), creates the file, chmods it to the required
4277 mode, then resets the umask. All the relevant calls to fopen() in the
4278 content scanning code have been changed to use this function.
4280 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
4281 to 24 hours. This avoids potential overflow problems when processing G
4282 and H retry rules. I suspect nobody ever tinkers with this value.
4284 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
4286 PH/45 When the plaintext authenticator is running as a client, the server's
4287 challenges are checked to ensure they are valid base64 strings. By
4288 default, the authentication attempt is cancelled if an invalid string is
4289 received. Setting client_ignore_invalid_base64 true ignores these errors.
4290 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
4291 they are received. Thus, the responses can be made to depend on the
4292 challenges. If an invalid string is ignored, an empty string is placed in
4295 PH/46 Messages that are created by the autoreply transport now contains a
4296 References: header, in accordance with RFCs 2822 and 3834.
4298 PH/47 Added authenticated_sender_force to the smtp transport.
4300 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
4302 PH/49 Installed latest patch from the Sieve maintainer.
4304 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
4305 was also set, the appendfile transport was unnecessarily scanning a
4306 directory of message files (e.g. for maildir delivery) to find the count
4307 of files (along with the size), even though it did not need this
4308 information. It now does the scan only if it needs to find either the
4309 size of the count of files.
4311 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
4313 PH/52 Two bugs concerned with error handling when the smtp transport is
4316 (i) Exim was not creating retry information for temporary errors given
4317 for individual recipients after the DATA command when the smtp transport
4318 was used in LMTP mode. This meant that they could be retried too
4319 frequently, and not timed out correctly.
4321 (ii) Exim was setting the flag that allows error details to be returned
4322 for LMTP errors on RCPT commands, but not for LMTP errors for individual
4323 recipients that were returned after the DATA command.
4325 PH/53 This is related to PH/52, but is more general: for any failing address,
4326 when detailed error information was permitted to be returned to the
4327 sender, but the error was temporary, then after the final timeout, only
4328 "retry timeout exceeded" was returned. Now it returns the full error as
4329 well as "retry timeout exceeded".
4331 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
4332 do this, and (what is worse) MTAs that accept it.
4334 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
4335 will now be deprecated.
4337 PH/56 New os.c-cygwin from the Cygwin maintainer.
4339 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
4340 formats (previously only available in exim formats via -bpr, -bpru,
4341 and -bpra. Now also available in native and exiqgrep formats)
4343 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
4344 with very large, slow to parse queues
4346 JJ/08 exipick: added ! as generic prefix to negate any criteria format
4348 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
4350 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
4351 responses to authentication challenges, though it was showing the
4352 challenges; (ii) I've removed the CR characters from the debug output for
4355 PH/58 Allow for the insertion of a newline as well as a space when a string
4356 is turned into more than one encoded-word during RFC 2047 encoding. The
4357 Sieve code now uses this.
4359 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
4360 data_4xx, lost_connection, tls_required.
4362 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
4363 message was being sent as an SMTP response.
4365 PH/61 Add -l and -k options to exicyclog.
4367 PH/62 When verifying, if an address was redirected to one new address, so that
4368 verification continued, and the new address failed or deferred after
4369 having set something in $address_data, the value of $address_data was not
4370 passed back to the ACL. This was different to the case when no
4371 redirection occurred. The value is now passed back in both cases.
4373 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
4374 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
4375 use login_cap.h, so on its own it isn't the distinguishing feature. The
4376 new name refers directly to the setclassresources() function.
4378 PH/65 Added configuration files for NetBSD3.
4380 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
4382 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
4383 is preferred over IPv4.
4385 PH/68 The bounce_return_message and bounce_return_body options were not being
4386 honoured for bounces generated during the reception of non-SMTP messages.
4387 In particular, this applied to messages rejected by the ACL. This bug has
4388 been fixed. However, if bounce_return_message is true and bounce_return_
4389 body is false, the headers that are returned for a non-SMTP message
4390 include only those that have been read before the error was detected.
4391 (In the case of an ACL rejection, they have all been read.)
4393 PH/69 The HTML version of the specification is now built in a directory called
4394 spec_html instead of spec.html, because the latter looks like a path with
4395 a MIME-type, and this confuses some software.
4397 PH/70 Catch two compiler warnings in sieve.c.
4399 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
4400 function verify_get_ident() calls ip_connect() to connect a socket, but
4401 if the "connect()" function timed out, ip_connect() used to close the
4402 socket. However, verify_get_ident() also closes the socket later, and in
4403 between Exim writes to the log, which may get opened at this point. When
4404 the socket was closed in ip_connect(), the log could get the same file
4405 descriptor number as the socket. This naturally causes chaos. The fix is
4406 not to close the socket in ip_connect(); the socket should be closed by
4407 the function that creates it. There was only one place in the code where
4408 this was missing, in the iplookup router, which I don't think anybody now
4409 uses, but I've fixed it anyway.
4411 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
4412 well as to direct DNS lookups. Otherwise the handling of names in host
4413 lists is inconsistent and therefore confusing.
4419 PH/01 Two changes to the default runtime configuration:
4421 (1) Move the checks for relay_from_hosts and authenticated clients from
4422 after to before the (commented out) DNS black list checks.
4424 (2) Add control=submission to the relay_from_hosts and authenticated
4425 clients checks, on the grounds that messages accepted by these
4426 statements are most likely to be submissions.
4428 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
4430 (1) Generate an error if the third argument for the ${prvs expansion is
4433 (2) Treat a missing third argument of ${prvscheck as if it were an empty
4436 (3) Reset the variables that are obtained from the first argument of
4437 ${prvscheck and used in the second argument before leaving the code,
4438 because their memory is reclaimed, so using them afterwards may do
4441 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
4442 one (it's much easier than Tom thought :-).
4444 (5) Because of (4), we can now allow for the use of $prvscheck_result
4445 inside the third argument.
4447 PH/03 For some reason, the default setting of PATH when running a command from
4448 a pipe transport was just "/usr/bin". I have changed it to
4451 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
4452 anything to be listed in the output from -bV.
4454 PH/05 When a filter generated an autoreply, the entire To: header line was
4455 quoted in the delivery log line, like this:
4457 => >A.N.Other <ano@some.domain> <original@ddress> ...
4459 This has been changed so that it extracts the operative address. There
4460 may be more than one such address. If so, they are comma-separated, like
4463 => >ano@some.domain,ona@other.domain <original@ddress> ...
4465 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
4466 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
4467 not being looked up in the rDNS to get a host name, Exim was showing the
4468 IP address twice in Received: lines, even though the IP addresses were
4469 identical. For example:
4471 Received: from [1.2.3.4] (helo=[1.2.3.4])
4473 However, if the real host name was known, it was omitting the HELO data
4474 if it matched the actual IP address. This has been tidied up so that it
4475 doesn't show the same IP address twice.
4477 PH/07 When both +timestamp and +memory debugging was on, the value given by
4478 $tod_xxx expansions could be wrong, because the tod_stamp() function was
4479 called by the debug printing, thereby overwriting the timestamp buffer.
4480 Debugging no longer uses the tod_stamp() function when +timestamp is set.
4482 PH/08 When the original message was included in an autoreply transport, it
4483 always said "this is a copy of the message, including all the headers",
4484 even if body_only or headers_only was set. It now gives an appropriate
4487 PH/09 Applied a patch from the Sieve maintainer which:
4489 o fixes some comments
4490 o adds the (disabled) notify extension core
4491 o adds some debug output for the result of if/elsif tests
4492 o points to the current vacation draft in the documentation
4493 and documents the missing references header update
4497 o fixes a bug in processing the envelope test (when testing
4498 multiple envelope elements, the last element determined the
4501 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
4502 Electronic Mail") by including:
4504 Auto-submitted: auto-generated
4506 in the messages that it generates (bounce messages and others, such as
4507 warnings). In the case of bounce messages for non-SMTP messages, there was
4508 also a typo: it was using "Auto_submitted" (underscore instead of
4509 hyphen). Since every message generated by Exim is necessarily in response
4510 to another message, thes have all been changed to:
4512 Auto-Submitted: auto-replied
4514 in accordance with these statements in the RFC:
4516 The auto-replied keyword:
4518 - SHOULD be used on messages sent in direct response to another
4519 message by an automatic process,
4521 - MUST NOT be used on manually-generated messages,
4523 - MAY be used on Delivery Status Notifications (DSNs) and Message
4524 Disposition Notifications (MDNs),
4526 - MUST NOT be used on messages generated by automatic or periodic
4527 processes, except for messages which are automatic responses to
4530 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
4531 to the default Received: header definition.
4533 PH/12 Added log selector acl_warn_skipped (default on).
4535 PH/13 After a successful wildlsearch lookup, discard the values of numeric
4536 variables because (a) they are in the wrong storage pool and (b) even if
4537 they were copied, it wouldn't work properly because of the caching.
4539 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
4540 checking when decoding. Apparently there are clients that generate
4541 overlong encoded strings. Why am I not surprised?
4543 PH/15 If the first argument of "${if match_address" was not empty, but did not
4544 contain an "@" character, Exim crashed. Now it writes a panic log message
4545 and treats the condition as false.
4547 PH/16 In autoreply, treat an empty string for "once" the same as unset.
4549 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
4550 extension "envelope-auth". The code is finished and in agreement with
4551 other implementations, but there is no documentation so far and in fact,
4552 nobody wrote the draft yet. This extension is currently #undef'ed, thus
4553 not changing the active code.
4555 Print executed "if" and "elsif" statements when debugging is used. This
4556 helps a great deal to understand what a filter does.
4558 Document more things not specified clearly in RFC3028. I had all this
4559 sorted out, when out of a sudden new issues came to my mind. Oops."
4561 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
4564 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
4565 canonical form (as documented). However, after a host name lookup from
4566 the IP address, check_host() was doing a simple string comparison with
4567 addresses acquired from the DNS when checking that the found name did
4568 have the original IP as one of its addresses. Since any found IPv6
4569 addresses are likely to be in abbreviated form, the comparison could
4570 fail. Luckily, there already exists a function for doing the comparison
4571 by converting both addresses to binary, so now that is used instead of
4572 the text comparison.
4574 PH/20 There was another similar case to PH/19, when a complete host name was
4575 given in a host list; looking up its IP address could give an abbreviated
4576 form, whereas the current host's name might or might not be abbreviated.
4577 The same fix has been applied.
4583 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
4584 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
4587 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
4588 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
4590 PH/03 Typo: missing ".o" in src/pcre/Makefile.
4592 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
4593 header line, restrict the check to what is listed in RFCs 2369 and 2929.
4594 Also, for "Auto-Submitted", treat anything other than "no" as
4595 non-personal, in accordance with RFC 3834. (Previously it treated
4596 anything starting "auto-" as non-personal.)
4598 TF/01 The control=submission/name=... option had a problem with syntax
4599 errors if the name included a slash character. The /name= option
4600 now slurps the rest of the string, so it can include any characters
4601 but it must come last in the list of options (after /sender_retain
4604 PH/05 Some modifications to the interface to the fake nameserver for the new
4612 TK/01 Added the "success_on_redirect" address verification option. See
4613 NewStuff for rationale and an example.
4615 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
4617 PH/02 Patch to exigrep to allow it to work on syslog lines.
4619 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
4620 fread() to skip over the body file's header line, because in Cygwin the
4621 header line is locked and is inaccessible.
4623 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
4624 co-exist for some time) to make it clear that it is the Exim ID that is
4625 referenced, not the Message-ID: header line.
4627 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
4628 string_format() function, because snprintf() does not exist on all
4631 PH/06 The use of forbid_filter_existstest now also locks out the use of the
4632 ${stat: expansion item.
4634 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
4635 protocol synchronization error", to keep the pedants happy.
4637 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
4638 well as for IRIX systems, when gcc is being used. See the host.c source
4641 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
4643 PH/10 Named domain lists were not working if used in a queue_smtp_domains
4646 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
4647 transport and to the smtp transport in LMTP mode.
4649 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
4651 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
4652 run a filter in a subprocess. This could lead to confusion in subsequent
4653 lookups in the parent process. There should also be a search_tidyup() at
4654 the end of the subprocess.
4656 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
4657 only if the host matched helo_try_verify_hosts, which caused the
4658 verification to occur when the EHLO/HELO command was issued. The ACL just
4659 tested the remembered result. Now, if a previous verification attempt has
4660 not happened, "verify = helo" does it there and then.
4662 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
4664 TK/03 Fix log output including CR from clamd.
4666 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
4667 exist provoked a memory error which could cause a segfault.
4669 PH/15 Installed PCRE 6.2
4671 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
4673 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
4674 of the problem. Specifically, suggested +O2 rather than +O1 for the
4677 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
4679 PH/20 If a delivery was routed to a non-standard port by means of an SRV
4680 record, the port was not correctly logged when the outgoing_port log
4681 selector was set (it logged the transort's default port).
4683 PH/21 Added support for host-specific ports to manualroute, queryprogram,
4684 fallback_hosts, and "hosts" in the smtp transport.
4686 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
4687 host errors such as "Connection refused".
4689 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
4690 authentication with radiusclient 0.4.9:
4692 - Error returned from rc_read_config was caught wrongly
4693 - Username/password not passed on to radius server due to wrong length.
4695 The presumption is that some radiusclient API changes for 4.51/PH/17
4696 were not taken care of correctly. The code is still untested by me (my
4697 Linux distribution still has 0.3.2 of radiusclient), but it was
4698 contributed by a Radius user.
4700 PH/24 When doing a callout, the value of $domain wasn't set correctly when
4701 expanding the "port" option of the smtp transport.
4703 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
4704 while reading a MIME header. Thanks to Tom Hughes for a patch.
4706 PH/24 Include config.h inside local_scan.h so that configuration settings are
4709 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
4710 that in an ACL for RCPT (for example), you can examine exactly what was
4713 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
4714 commands, but it was not correctly comparing the address with the actual
4715 client host address. Thus, it would show the EHLO address in Received:
4716 header lines when this was not necessary.
4718 PH/27 Added the % operator to ${eval:}.
4720 PH/28 Exim tries to create and chdir to its spool directory when it starts;
4721 it should be ignoring failures (because with -C, for example, it has lost
4722 privilege). It wasn't ignoring creation failures other than "already
4725 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
4728 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
4729 because an input line was too long, either on its own, or by virtue of
4730 too many continuations, the temporary file was not being removed, and the
4731 return code was incorrect.
4733 PH/31 Missing "BOOL" in function definition in filtertest.c.
4735 PH/32 Applied Sieve patches from the maintainer.
4737 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
4739 PH/33 Added "verify = not_blind".
4741 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
4742 Local/Makefile (with some defaults set). These are used in built scripts
4743 such as exicyclog, but they have never been used in the exim_install
4744 script (though there are many overriding facilities there). I have
4745 arranged that the exim_install script now takes note of these two
4748 PH/35 Installed configuration files for Dragonfly.
4750 PH/36 When a locally submitted message by a trusted user did not contain a
4751 From: header, and the sender address was obtained from -f or from an SMTP
4752 MAIL command, and the trusted user did not use -F to supply a sender
4753 name, $originator_name was incorrectly used when constructing a From:
4754 header. Furthermore, $originator_name was used for submission mode
4755 messages from external hosts without From: headers in a similar way,
4756 which is clearly wrong.
4758 PH/37 Added control=suppress_local_fixups.
4760 PH/38 When log_selector = +received_sender was set, and the addition of the
4761 sender made the log line's construction buffer exactly full, or one byte
4762 less than full, an overflow happened when the terminating "\n" was
4765 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
4766 when the result of a list match is failure because a DNS lookup failed.
4768 PH/40 RM_COMMAND is now used in the building process.
4770 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
4771 the "build-* directories that it finds.
4773 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
4774 address in a domain literal was a prefix of an interface address.
4776 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
4777 when verifying a sender address, unless rewrite_headers is false.
4779 PH/44 Wrote a long comment about why errors_to addresses are verified as
4780 recipients, not senders.
4782 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
4783 the ratelimit ACL was added.
4785 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
4787 PH/46 Added extra information about PostgreSQL errors to the error string.
4789 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
4790 avoiding the need to install special zones in a real server. This is
4791 backwards compatible; if it can't find the fake resolver, it drops back.
4792 Thus, both old and new test suites can be run.
4794 TF/02 Added util/ratelimit.pl
4796 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
4797 clock is set back in time.
4799 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
4800 Candler <B.Candler@pobox.com>.
4802 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
4803 OK for addresses that are the result of redirections.
4805 PH/48 A number of further additions for the benefit of the new test suite,
4806 including a fake gethostbyname() that interfaces to the fake DNS resolver
4809 TF/06 The fix for widen_domains has also been applied to qualify_single and
4810 search_parents which are the other dnslookup options that can cause
4813 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
4816 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
4818 TF/07 Exim produced the error message "an SRV record indicated no SMTP
4819 service" if it encountered an MX record with an empty target hostname.
4820 The message is now "an MX or SRV record indicated no SMTP service".
4822 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
4823 if the DNS of the sending site is misconfigured. This is quite a
4824 common situation. This change restores the behaviour of treating a
4825 helo verification defer as a failure.
4827 PH/51 If self=fail was set on a router, the bounce message did not include the
4828 actual error message.
4834 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
4836 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
4837 command itself ended in error, the underlying message about the transport
4838 filter timeout was being overwritten with the pipe command error. Now the
4839 underlying error message should be appended to the second error message.
4841 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4843 PH/02 Reduce the amount of output that "make" produces by default. Full output
4844 can still be requested.
4846 PH/03 The warning log line about a condition test deferring for a "warn" verb
4847 was being output only once per connection, rather than after each
4848 occurrence (because it was using the same function as for successful
4849 "warn" verbs). This seems wrong, so I have changed it.
4851 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4852 it should not have, which might have caused a crash in the right
4853 circumstances, but probably never did.
4855 PH/04 Installed a modified version of Tony Finch's patch to make submission
4856 mode fix the return path as well as the Sender: header line, and to
4857 add a /name= option so that you can make the user's friendly name appear
4860 TF/03 Added the control = fakedefer ACL modifier.
4862 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4863 Mark Lowes for thorough testing.
4865 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4867 TK/03 Merged latest SRS patch from Miles Wilton.
4869 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4870 with the definition in sysexits.h (which is #included earlier).
4871 Fortunately, Exim does not actually use EX_OK. The code used to try to
4872 preserve the sysexits.h value, by assuming that macro definitions were
4873 scanned for macro replacements. I have been disabused of this notion,
4874 so now the code just undefines EX_OK before #including unistd.h.
4876 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4877 in the smtp transport. When a block could not be written in a single
4878 write() function, the timeout was being re-applied to each part-write.
4879 This seems wrong - if the receiver was accepting one byte at a time it
4880 would take for ever. The timeout is now adjusted when this happens. It
4881 doesn't have to be particularly precise.
4883 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4884 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4886 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4887 without a domain if the check to <postmaster@domain> fails.
4889 SC/01 Eximstats: added -xls and the ability to specify output files
4890 (patch written by Frank Heydlauf).
4892 SC/02 Eximstats: use FileHandles for outputting results.
4894 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4896 SC/04 Eximstats: fixed display of large numbers with -nvr option
4898 SC/05 Eximstats: fixed merging of reports with empty tables.
4900 SC/06 Eximstats: added the -include_original_destination flag
4902 SC/07 Eximstats: removed tabs and trailing whitespace.
4904 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4906 TK/06 MBOX spool code: Add real "From " MBOX separator line
4907 so the .eml file is really in mbox format (even though
4908 most programs do not really care). Patch from Alex Miller.
4910 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4911 The latter is generated from $received_to and is only set if the
4912 message has one envelope recipient. SA can use these headers,
4913 obviously out-of-the-box. Patch from Alex Miller.
4915 PH/08 The ${def test on a variable was returning false if the variable's
4916 value was "0", contrary to what the specification has always said!
4917 The result should be true unless the variable is empty.
4919 PH/09 The syntax error of a character other than { following "${if
4920 def:variable_name" (after optional whitespace) was not being diagnosed.
4921 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4922 accidental colon was present, for example, could give incorrect results.
4924 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4925 result is used (not including appendfile, where other changes are about
4928 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4929 This involved changing a lot of size variables from int to off_t. It
4930 should work with maildirs and everything.
4932 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4933 spamd dying while we are connected to it.
4935 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4938 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4939 function for BDB 4.3.
4941 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4943 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4944 the sources for PCRE have been re-arranged and more files are now
4947 PH/15 The code I had for printing potentially long long variables in PH/11
4948 above was not the best (it lost precision). The length of off_t variables
4949 is now inspected at build time, and an appropriate printing format (%ld
4950 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4951 to be "long long int" or "long int". This is needed for the internal
4952 formatting function string_vformat().
4954 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4955 the configuration file to be ":syslog", then the script "guesses" where
4956 the logs files are, rather than using the compiled in default. In our
4957 case the guess is not the same as the compiled default, so the script
4958 suddenly stopped working when I started to use syslog. The patch checks
4959 to see if log_file_path is "". If so, it attempts to read it from exim
4960 with no configuration file to get the compiled in version, before it
4961 falls back to the previous guessing code."
4963 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4964 implementing BATV in an Exim configuration. See NewStuff for the gory
4967 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4968 Makefile that are specific to HP-UX.
4970 PH/18 If the "use_postmaster" option was set for a recipient callout together
4971 with the "random" option, the postmaster address was used as the MAIL
4972 FROM address for the random test, but not for the subsequent recipient
4973 test. It is now used for both.
4975 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4976 patch removes a few documentation additions to RFC 3028, because the
4977 latest draft now contains them. It adds the new en;ascii-case comparator
4978 and a new error check for 8bit text in MIME parts. Comparator and
4979 require names are now matched exactly. I enabled the subaddress
4980 extension, but it is not well tested yet (read: it works for me)."
4982 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4983 rework some of the code of TK/09 above to avoid the hardwired use of
4984 "%lld" and "long long". Replaced the call to snprintf() with a call to
4987 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4988 records point to non-existent hosts", "retry timeout exceeded", and
4989 "retry time not reached for any host after a long failure period".
4991 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4992 experimental DomainKeys support:
4994 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4995 (2) On an error such as an illegally used "control", the wrong name for
4996 the control was given.
4998 These problems did NOT occur unless DomainKeys support was compiled.
5000 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
5002 PH/24 Added ${if match_ip condition.
5004 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
5005 databases so that it will be absolutely obvious if a crash occurs in the
5006 DB library. This is a regular occurrence (often caused by mis-matched
5009 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
5010 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
5011 picked up on a user's system that detects such things. There doesn't seem
5012 to be a gcc warning option for this - only an attribute that has to be
5013 put on the function's prototype. It seems that in Fedora Core 4 they have
5014 set this on a number of new functions. No doubt there will be more in due
5017 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
5018 specify a transport. However, if an address that was verified by such a
5019 router was the subject of a callout, Exim crashed because it tried to
5020 read the rcpt_include_affixes from the non-existent transport. Now it
5021 just assumes that the setting of that option is false. This bug was
5022 introduced by 4.51/PH/31.
5024 PH/28 Changed -d+all to exclude +memory, because that information is very
5025 rarely of interest, but it makes the output a lot bigger. People tend to
5026 do -d+all out of habit.
5028 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
5029 code in os-type was giving problems when libc.so lives in lib64, like on
5032 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
5033 aren't the modern standard, and it seems that some systems' include files
5034 don't always have them. Exim was already checking for some of the newer
5035 ones like T_AAAA, and defining it itself. I've added checks for all the
5036 record types that Exim uses.
5038 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
5039 not automatically generating a new one, as it is supposed to. This
5040 prevented TLS from working. If the file did exist, but contained invalid
5041 data, a new version was generated, as expected. It was only the case of a
5042 non-existent file that was broken.
5044 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
5045 with a change in libdomainkeys > 0.64.
5047 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
5048 from DNS. If the selector record carries the flag, it now has
5049 precedence over the domain-wide flag.
5051 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
5053 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
5054 the use of a port name that isn't defined in /etc/services) occurred, the
5055 message was deferred as in a normal delivery, and thus remained on the
5056 spool, instead of being failed because of the mua_wrapper setting. This
5057 is now fixed, and I tidied up some of the mua_wrapper messages at the
5060 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
5061 the messages in a hash of arrays rather than using individual hashes.
5062 This is a bit cleaner and results in dramatic memory savings, albeit
5063 at a slight CPU cost.
5065 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
5066 as requested by Marc Sherman.
5068 SC/10 Eximstats: added histograms for user specified patterns as requested
5071 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
5073 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
5074 fopen() in the content-scanning modules that did not already have it.
5080 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
5081 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
5083 TK/02 Fix ACL "control" statement not being available in MIME ACL.
5085 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
5087 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
5088 to test Sieve filters that use "vacation".
5090 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
5091 that changes the way the GnuTLS parameters are stored in the cache file.
5092 The new format can be generated externally. For backward compatibility,
5093 if the data in the cache doesn't make sense, Exim assumes it has read an
5094 old-format file, and it generates new data and writes a new file. This
5095 means that you can't go back to an older release without removing the
5098 PH/03 A redirect router that has both "unseen" and "one_time" set does not
5099 work if there are any delivery delays because "one_time" forces the
5100 parent to be marked "delivered", so its unseen clone is never tried
5101 again. For this reason, Exim now forbids the simultaneous setting of
5104 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
5105 redirected to themselves ("homonym" addresses). Read the long ChangeLog
5106 entry if you want to know the details. The fix, however, neglected to
5107 consider the case when local delivery batching is involved. The test for
5108 "previously delivered" was not happening when checking to see if an
5109 address could be batched with a previous (undelivered) one; under
5110 certain circumstances this could lead to multiple deliveries to the same
5113 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
5114 in its include files, and this causes problems building Exim.
5116 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
5117 header_syntax) but Exim was just ignoring anything given after a slash.
5118 In particular, this caused confusion with an attempt to use "verify =
5119 reverse_host_lookup/defer_ok". An error is now given when options are
5120 supplied for verify items that do not have them. (Maybe reverse_host_
5121 lookup should have a defer_ok option, but that's a different point.)
5123 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
5124 defined by RFC 821) to 2048, because there were problems with some AUTH
5125 commands, and RFC 1869 says the size should be increased for extended
5126 SMTP commands that take arguments.
5128 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
5131 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
5132 "unknown" error; now it says that the functionality isn't in the binary.
5134 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
5135 an address' error message when a string expansion fails (syntax or
5136 whatever). Otherwise the password may appear in the log. Following change
5137 PH/42 below, there is no longer a chance of it appearing in a bounce
5140 PH/11 Installed exipick version 20050225.0 from John Jetmore.
5142 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
5143 of its addresses was ever tried. (Bugzilla bug #2.)
5145 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
5146 the result incorrectly in the debug output. (It correctly added a newline
5147 to what was transported.)
5149 TF/01 Added $received_time.
5151 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
5152 commented out examples of how to interface to a virus scanner and to
5153 SpamAssassin. Also added commented examples of av_scanner and
5154 spamd_address settings.
5156 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
5157 and controls are allowed in which ACLs. There were a couple of minor
5158 errors. Some of the entries in the conditions table (which is a table of
5159 where they are NOT allowed) were getting very unwieldy; rewrote them as a
5160 negation of where the condition IS allowed.
5162 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
5164 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
5165 header file does not have a version number, so I've had to invent a new
5166 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
5167 API. The code is untested by me (my Linux distribution still has 0.3.2 of
5168 radiusclient), but it was contributed by a Radius user.
5170 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
5171 files or directories, for OpenSSL.
5173 PH/19 When an Exim process that is running as root has to create an Exim log
5174 file, it does so in a subprocess that runs as exim:exim so as to get the
5175 ownership right at creation (otherwise, other Exim processes might see
5176 the file with the wrong ownership). There was no test for failure of this
5177 fork() call, which would lead to the process getting stuck as it waited
5178 for a non-existent subprocess. Forks do occasionally fail when resources
5179 run out. I reviewed all the other calls to fork(); they all seem to check
5182 PH/20 When checking for unexpected SMTP input at connect time (before writing
5183 the banner), Exim was not dealing correctly with a non-positive return
5184 from the read() function. If the client had disconnected by this time,
5185 the result was a log entry for a synchronization error with an empty
5186 string after "input=" when read() returned zero. If read() returned -1
5187 (an event I could not check), uninitialized data bytes were printed.
5188 There were reports of junk text (parts of files, etc) appearing after
5191 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
5193 PH/22 Added support for macro redefinition, and (re)definition in between
5194 driver and ACL definitions.
5196 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
5197 forgetting to use the resulting value; it was using the unexpanded value.
5199 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
5200 hadn't been configured. The fix is from Juergen Kreileder, who
5201 understands it better than I do:
5203 "Here's what I see happening with three configured cyrus_sasl
5204 authenticators configured (plain, login, cram-md5):
5206 On startup auth_cyrus_sasl_init() gets called for each of these.
5207 This means three calls to sasl_listmech() without a specified mech_list.
5208 => SASL tests which mechs of all available mechs actually work
5209 => three warnings about OTP not working
5210 => the returned list contains: plain, login, cram-md5, digest-md5, ...
5212 With the patch, sasl_listmech() also gets called three times. But now
5213 SASL's mech_list option is set to the server_mech specified in the the
5214 authenticator. Or in other words, the answer from sasl_listmech()
5215 gets limited to just the mech you're testing for (which is different
5217 => the return list contains just 'plain' or 'login', 'cram-md5' or
5218 nothing depending on the value of ob->server_mech.
5220 I've just tested the patch: Authentication still works fine,
5221 unavailable mechs specified in the exim configuration are still
5222 caught, and the auth.log warnings about OTP are gone."
5224 PH/25 When debugging is enabled, the contents of the command line are added
5225 to the debugging output, even when log_selector=+arguments is not
5228 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
5229 answer is "GNU", and only if the return is "GNU/something" is the answer
5232 PH/27 $acl_verify_message is now set immediately after the failure of a
5233 verification in an ACL, and so is available in subsequent modifiers. In
5234 particular, the message can be preserved by coding like this:
5236 warn !verify = sender
5237 set acl_m0 = $acl_verify_message
5239 Previously, $acl_verify_message was set only while expanding "message"
5240 and "log_message" when a very denied access.
5242 PH/28 Modified OS/os.c-Linux with
5244 -#ifndef OS_LOAD_AVERAGE
5245 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
5247 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
5248 nomenclature these days.)
5250 PH/29 Installed patch from the Sieve maintainer that adds the options
5251 sieve_useraddress and sieve_subaddress to the redirect router.
5253 PH/30 In these circumstances:
5254 . Two addresses routed to the same list of hosts;
5255 . First host does not offer TLS;
5256 . First host accepts first address;
5257 . First host gives temporary error to second address;
5258 . Second host offers TLS and a TLS session is established;
5259 . Second host accepts second address.
5260 Exim incorrectly logged both deliveries with the TLS parameters (cipher
5261 and peerdn, if requested) that were in fact used only for the second
5264 PH/31 When doing a callout as part of verifying an address, Exim was not paying
5265 attention to any local part prefix or suffix that was matched by the
5266 router that accepted the address. It now behaves in the same way as it
5267 does for delivery: the affixes are removed from the local part unless
5268 rcpt_include_affixes is set on the transport.
5270 PH/32 Add the sender address, as F=<...>, to the log line when logging a
5271 timeout during the DATA phase of an incoming message.
5273 PH/33 Sieve envelope tests were broken for match types other than :is. I have
5274 applied a patch sanctioned by the Sieve maintainer.
5276 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
5277 the uid or gid is negative. A case of a negative gid caused this to be
5278 noticed. The fix allows for either to be negative.
5280 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
5281 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
5284 PH/36 The change PH/12 above was broken. Fixed it.
5286 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
5287 the grounds that routing the same address twice would always produce the
5288 same answer. This might have been true once, but it is certainly no
5289 longer true now. Routing a child address may depend on the previous
5290 routing that produced that child. Some complicated redirection strategies
5291 went wrong when messages had multiple recipients, and made Exim's
5292 behaviour dependent on the order in which the addresses were given.
5294 I have moved the duplicate checking until after the routing is complete.
5295 Exim scans the addresses that are assigned to local and remote
5296 transports, and removes any duplicates. This means that more work will be
5297 done, as duplicates will always all be routed, but duplicates are
5298 presumably rare, so I don't expect this is of any significance.
5300 For deliveries to pipes, files, and autoreplies, the duplicate checking
5301 still happens during the routing process, since they are not going to be
5304 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
5305 It corrects a timeout issue with spamd. This is Ian's comment: "The
5306 background is that sometimes spamd either never reads data from a
5307 connection it has accepted, or it never writes response data. The exiscan
5308 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
5309 blindly assumes that writes won't block so it may never time out."
5311 PH/39 Allow G after quota size as well as K and M.
5313 PH/40 The value set for $authenticated_id in an authenticator may not contain
5314 binary zeroes or newlines because the value is written to log lines and
5315 to spool files. There was no check on this. Now the value is run through
5316 the string_printing() function so that such characters are converted to
5317 printable escape sequences.
5319 PH/41 $message_linecount is a new variable that contains the total number of
5320 lines in the message. Compare $body_linecount, which is the count for the
5323 PH/42 Exim no longer gives details of delivery errors for specific addresses in
5324 bounce and delay warning messages, except in certain special cases, which
5327 (a) An SMTP error message from a remote host;
5328 (b) A message specified in a :fail: redirection;
5329 (c) A message specified in a "fail" command in a system filter;
5330 (d) A message specified in a FAIL return from the queryprogram router;
5331 (e) A message specified by the cannot_route_message router option.
5333 In these cases only, Exim does include the error details in bounce and
5334 warning messages. There are also a few cases where bland messages such
5335 as "unrouteable address" or "local delivery error" are given.
5337 PH/43 $value is now also set for the "else" part of a ${run expansion.
5339 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
5340 being worked on, but at least Exim now implements the latest version to
5343 PH/45 In a pipe transport, although a timeout while waiting for the pipe
5344 process to complete was treated as a delivery failure, a timeout while
5345 writing the message to the pipe was logged, but erroneously treated as a
5346 successful delivery. Such timeouts include transport filter timeouts. For
5347 consistency with the overall process timeout, these timeouts are now
5348 treated as errors, giving rise to delivery failures by default. However,
5349 there is now a new Boolean option for the pipe transport called
5350 timeout_defer, which, if set TRUE, converts the failures into defers for
5351 both kinds of timeout. A transport filter timeout is now identified in
5354 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
5355 systems where "make" and "gmake" are different, calling "gmake" at top
5356 level broke things. I've arranged for the value of $(MAKE) to be passed
5357 from the Makefile to this script so that it can call the same version of
5361 A note about Exim versions 4.44 and 4.50
5362 ----------------------------------------
5364 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
5365 changes of various kinds. As a consequence, a big documentation update was
5366 needed. This delayed the release for rather longer than seemed good, especially
5367 in the light of a couple of (minor) security issues. Therefore, the changes
5368 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
5369 release. So 4.44 and 4.50 are in effect two different branches that both start
5372 I have left the 4.50 change log unchanged; it contains all the changes since
5373 4.43. The change log for 4.44 is below; many of its items are identical to
5374 those for 4.50. This seems to be the most sensible way to preserve the
5375 historical information.
5381 1. Minor wording change to the doc/README.SIEVE file.
5383 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
5384 computation of the current number of files was incorrect.
5386 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5387 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5390 4. Give more explanation in the error message when the command for a transport
5391 filter fails to execute.
5393 5. There are several places where Exim runs a non-Exim command in a
5394 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5395 was being done only for the command run by the queryprogram router. It is
5396 now done for all such subprocesses. The other cases are: ${run, transport
5397 filters, and the commands run by the lmtp and pipe transports.
5399 6. Added CONFIGURE_GROUP build-time option.
5401 7. Some older OS have a limit of 256 on the maximum number of file
5402 descriptors. Exim was using setrlimit() to set 1000 as a large value
5403 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5404 systems. I've change it so that if it can't get 1000, it tries for 256.
5406 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5407 was an oversight, and furthermore, ever since the addition of extra
5408 controls (e.g. 4.43/32), the checks on when to allow different forms of
5409 "control" were broken. There should now be diagnostics for all cases when a
5410 control that does not make sense is encountered.
5412 9. Added the /retain_sender option to "control=submission".
5414 10. $recipients is now available in the predata ACL (oversight).
5416 11. Tidy the search cache before the fork to do a delivery from a message
5417 received from the command line. Otherwise the child will trigger a lookup
5418 failure and thereby defer the delivery if it tries to use (for example) a
5419 cached ldap connection that the parent has called unbind on.
5421 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5422 of $address_data from the recipient verification was clobbered by the
5423 sender verification.
5425 13. The value of address_data from a sender verification is now available in
5426 $sender_address_data in subsequent conditions in the ACL statement.
5428 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
5430 15. Added a new option "connect=<time>" to callout options, to set a different
5433 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5434 was its contents. (It was OK if the option was not defined at all.)
5436 17. A "Completed" log line is now written for messages that are removed from
5437 the spool by the -Mrm option.
5439 18. New variables $sender_verify_failure and $recipient_verify_failure contain
5440 information about exactly what failed.
5442 19. Added -dd to debug only the daemon process.
5444 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
5445 handles timeouts, both on the server side and network timeouts. Renamed the
5446 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
5448 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
5449 It is now set to "smtps".
5451 22. $host_address is now set to the target address during the checking of
5452 ignore_target_hosts.
5454 23. When checking ignore_target_hosts for an ipliteral router, no host name was
5455 being passed; this would have caused $sender_host_name to have been used if
5456 matching the list had actually called for a host name (not very likely,
5457 since this list is usually IP addresses). A host name is now passed as
5460 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5461 code that specifies a non-restarting handler (typically sigaction() in
5462 modern systems) in an attempt to fix a rare and obscure crash bug.
5464 25. Narrowed the window for a race in the daemon that could cause it to ignore
5465 SIGCHLD signals. This is not a major problem, because they are used only to
5466 wake it up if nothing else does.
5468 26. A malformed maildirsize file could cause Exim to calculate negative values
5469 for the mailbox size or file count. Odd effects could occur as a result.
5470 The maildirsize information is now recalculated if the size or filecount
5473 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5474 support for a long time. Removed HAVE_SYS_VFS_H.
5476 28. Installed the latest version of exipick from John Jetmore.
5478 29. In an address list, if the pattern was not a regular expression, an empty
5479 subject address (from a bounce message) matched only if the pattern was an
5480 empty string. Non-empty patterns were not even tested. This was the wrong
5481 because it is perfectly reasonable to use an empty address as part of a
5482 database query. An empty address is now tested by patterns that are
5483 lookups. However, all the other forms of pattern expect the subject to
5484 contain a local part and a domain, and therefore, for them, an empty
5485 address still always fails if the pattern is not itself empty.
5487 30. Exim went into a mad DNS loop when attempting to do a callout where the
5488 host was specified on an smtp transport, and looking it up yielded more
5489 than one IP address.
5491 31. Re-factored the code for checking spool and log partition space into a
5492 function that finds that data and another that does the check. The former
5493 is then used to implement four new variables: $spool_space, $log_space,
5494 $spool_inodes, and $log_inodes.
5496 32. The RFC2047 encoding function was originally intended for short strings
5497 such as real names; it was not keeping to the 75-character limit for
5498 encoded words that the RFC imposes. It now respects the limit, and
5499 generates multiple encoded words if necessary. To be on the safe side, I
5500 have increased the buffer size for the ${rfc2047: expansion operator from
5503 33. It is now permitted to omit both strings after an "if" condition; if the
5504 condition is true, the result is "true". As before, when the second string
5505 is omitted, a false condition yields an empty string. This makes it less
5506 cumbersome to write custom ACL and router conditions.
5508 34. Failure to deliver a bounce message always caused it to be frozen, even if
5509 there was an errors_to setting on the router. The errors_to setting is now
5512 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5513 canonical form (fully expanded) before being placed in
5514 $sender_host_address.
5516 36. The table in the code that translates DNS record types into text (T_A to
5517 "A" for instance) was missing entries for NS and CNAME. It is just possible
5518 that this could have caused confusion if both these types were looked up
5519 for the same domain, because the text type is used as part of Exim's
5520 per-process caching. But the chance of anyone hitting this buglet seems
5523 37. The dnsdb lookup has been extended in a number of ways.
5525 (1) There is a new type, "zns", which walks up the domain tree until it
5526 finds some nameserver records. It should be used with care.
5528 (2) There is a new type, "mxh", which is like "mx" except that it returns
5529 just the host names, not the priorities.
5531 (3) It is now possible to give a list of domains (or IP addresses) to be
5532 looked up. The behaviour when one of the lookups defers can be
5533 controlled by a keyword.
5535 (4) It is now possible to specify the separator character for use when
5536 multiple records are returned.
5538 38. The dnslists ACL condition has been extended: it is now possible to supply
5539 a list of IP addresses and/or domains to be looked up in a particular DNS
5542 39. Added log_selector=+queue_time_overall.
5544 40. When running the queue in the test harness, wait just a tad after forking a
5545 delivery process, to get repeatability of debugging output.
5547 41. Include certificate and key file names in error message when GnuTLS fails
5548 to set them up, because the GnuTLS error message doesn't include the name
5549 of the failing file when there is a problem reading it.
5551 42. Allow both -bf and -bF in the same test run.
5553 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
5555 44. The "Exiscan patch" is now merged into the mainline Exim source.
5557 45. Sometimes the final signoff response after QUIT could fail to get
5558 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5559 < 0 before doing a fflush(). This bug looks as though it goes back to the
5560 introduction of TLS in release 3.20, but "sometimes" must have been rare
5561 because the tests only now provoked it.
5563 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
5564 (this can affect the format of dates).
5566 47. exim_tidydb, when checking for the continued existence of a message for
5567 which it has found a message-specific retry record, was not finding
5568 messages that were in split spool directories. Consequently, it was
5569 deleting retry records that should have stayed in existence.
5571 48. Steve fixed some bugs in eximstats.
5573 49. The SPA authentication driver was not abandoning authentication and moving
5574 on to the next authenticator when an expansion was forced to fail,
5575 contradicting the general specification for all authenticators. Instead it
5576 was generating a temporary error. It now behaves as specified.
5578 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
5579 (the order specifies the preference for clients). The order is now AES256,
5580 AES128, 3DES, ARCFOUR128.
5582 51. Small patch to Sieve code - explicitly set From: when generating an
5585 52. Exim crashed if a remote delivery caused a very long error message to be
5586 recorded - for instance if somebody sent an entire SpamAssassin report back
5587 as a large number of 550 error lines. This bug was coincidentally fixed by
5588 increasing the size of one of Exim's internal buffers (big_buffer) that
5589 happened as part of the Exiscan merge. However, to be on the safe side, I
5590 have made the code more robust (and fixed the comments that describe what
5593 53. Now that there can be additional text after "Completed" in log lines (if
5594 the queue_time_overall log selector is set), a one-byte patch to exigrep
5595 was needed to allow it to recognize "Completed" as not the last thing in
5598 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
5599 patch that reportedly fixes this has been added. I am not expert enough to
5600 create a test for it. This is what the patch creator wrote:
5602 "I found a little strange behaviour of ldap code when working with
5603 Windows 2003 AD Domain, where users was placed in more than one
5604 Organization Units. When I tried to give exim partial DN, the exit code
5605 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
5606 But simultaneously result of request was absolutely normal ldap result,
5607 so I produce this patch..."
5609 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
5610 REFERENCE, so I have modified the code to exclude the patch when that macro
5613 55. Some experimental protocols are using DNS PTR records for new purposes. The
5614 keys for these records are domain names, not reversed IP addresses. The
5615 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5616 leaves it alone. Component reversal etc. now happens only for IP addresses.
5619 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
5621 57. Double the size of the debug message buffer (to 2048) so that more of very
5622 long debug lines gets shown.
5624 58. The exicyclog utility now does better if the number of log files to keep
5625 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
5627 59. Two changes related to the smtp_active_hostname option:
5629 (1) $smtp_active_hostname is now available as a variable.
5630 (2) The default for smtp_banner uses $smtp_active_hostname instead
5631 of $primary_hostname.
5633 60. The host_aton() function is supposed to be passed a string that is known
5634 to be a valid IP address. However, in the case of IPv6 addresses, it was
5635 not checking this. This is a hostage to fortune. Exim now panics and dies
5636 if the condition is not met. A case was found where this could be provoked
5637 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5638 components; fortuitously, this particular loophole had already been fixed
5639 by change 4.50/55 above.
5641 If there are any other similar loopholes, the new check in host_aton()
5642 itself should stop them being exploited. The report I received stated that
5643 data on the command line could provoke the exploit when Exim was running as
5644 exim, but did not say which command line option was involved. All I could
5645 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5646 running as the user.
5649 61. There was a buffer overflow vulnerability in the SPA authentication code
5650 (which came originally from the Samba project). I have added a test to the
5651 spa_base64_to_bits() function which I hope fixes it.
5654 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
5655 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
5657 63. The daemon start-up calls getloadavg() while still root for those OS that
5658 need the first call to be done as root, but it missed one case: when
5659 deliver_queue_load_max is set with deliver_drop_privilege. This is
5660 necessary for the benefit of the queue runner, because there is no re-exec
5661 when deliver_drop_privilege is set.
5663 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
5664 This has been fixed.
5666 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5667 was in use, was not putting the data itself into the right store pool;
5668 consequently, it could be overwritten for a subsequent message in the same
5669 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5672 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
5674 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
5675 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
5676 the function were treating the return as a boolean value, which happened to
5677 work because 0=false and not-0=true, but is not correct code.
5679 68. The host_aton() function was not handling scoped IPv6 addresses (those
5680 with, for example, "%eth0" on the end) correctly.
5682 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
5683 negated items (that is, ~something) in unsigned ints. Some compilers
5684 apparently mutter when there is no cast.
5686 70. If an address verification called from an ACL failed, and did not produce a
5687 user-specific message (i.e. there was only a "system" message), nothing was
5688 put in $acl_verify_message. In this situation, it now puts the system
5691 71. Change 4.23/11 added synchronization checking at the start of an SMTP
5692 session; change 4.31/43 added the unwanted input to the log line - except
5693 that it did not do this in the start of session case. It now does.
5695 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
5696 This is wrong and can cause the other end to generate a synchronization
5697 error if it is another Exim or anything else that does the synchronization
5698 check. A QUIT command is no longer sent after a timeout.
5700 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
5701 during host lookups.
5703 74. The defer_ok option of callout verification was not working if it was used
5704 when verifying addresses in header lines, that is, for this case:
5706 verify = header_sender/callout=defer_ok
5708 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
5709 those file descriptors could be used for SMTP connections. If anything
5710 wrote to stderr (the example that came up was "warn" in embedded Perl), it
5711 could be sent to the SMTP client, causing chaos. The daemon now opens
5712 stdin, stdout, and stderr to /dev/null when it puts itself into the
5715 76. Arrange for output from Perl's "warn" command to be written to Exim's main
5716 log by default. The user can override this with suitable Perl magic.
5718 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
5719 the log message when discard triggers, was not working for the DATA ACL or
5720 for the non-SMTP ACL.
5722 78. Error message wording change in sieve.c.
5724 79. If smtp_accept_max_per_host was set, the number of connections could be
5725 restricted to fewer than expected, because the daemon was trying to set up
5726 a new connection before checking whether the processes handling previous
5727 connections had finished. The check for completed processes is now done
5728 earlier. On busy systems, this bug wouldn't be noticed because something
5729 else would have woken the daemon, and it would have reaped the completed
5732 80. If a message was submitted locally by a user whose login name contained one
5733 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
5734 It caused a spool format error. I have fixed the spool reading code. A
5735 related problem was that the "from" clause in the Received: line became
5736 illegal because of the space(s). It is now covered by ${quote_local_part.
5738 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
5741 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
5743 83. Patch from Sieve maintainer to fix unterminated string problem in
5744 "vacation" handling.
5746 84. Some minor changes to the Linux configuration files to help with other
5747 OS variants using glibc.
5749 85. One more patch for Sieve to update vacation handling to latest spec.
5752 ----------------------------------------------------
5753 See the note above about the 4.44 and 4.50 releases.
5754 ----------------------------------------------------
5760 1. Change 4.43/35 introduced a bug that caused file counts to be
5761 incorrectly computed when quota_filecount was set in an appendfile
5764 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5765 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5768 3. Give more explanation in the error message when the command for a transport
5769 filter fails to execute.
5771 4. There are several places where Exim runs a non-Exim command in a
5772 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5773 was being done only for the command run by the queryprogram router. It is
5774 now done for all such subprocesses. The other cases are: ${run, transport
5775 filters, and the commands run by the lmtp and pipe transports.
5777 5. Some older OS have a limit of 256 on the maximum number of file
5778 descriptors. Exim was using setrlimit() to set 1000 as a large value
5779 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5780 systems. I've change it so that if it can't get 1000, it tries for 256.
5782 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5783 was an oversight, and furthermore, ever since the addition of extra
5784 controls (e.g. 4.43/32), the checks on when to allow different forms of
5785 "control" were broken. There should now be diagnostics for all cases when a
5786 control that does not make sense is encountered.
5788 7. $recipients is now available in the predata ACL (oversight).
5790 8. Tidy the search cache before the fork to do a delivery from a message
5791 received from the command line. Otherwise the child will trigger a lookup
5792 failure and thereby defer the delivery if it tries to use (for example) a
5793 cached ldap connection that the parent has called unbind on.
5795 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5796 of $address_data from the recipient verification was clobbered by the
5797 sender verification.
5799 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5800 was its contents. (It was OK if the option was not defined at all.)
5802 11. A "Completed" log line is now written for messages that are removed from
5803 the spool by the -Mrm option.
5805 12. $host_address is now set to the target address during the checking of
5806 ignore_target_hosts.
5808 13. When checking ignore_target_hosts for an ipliteral router, no host name was
5809 being passed; this would have caused $sender_host_name to have been used if
5810 matching the list had actually called for a host name (not very likely,
5811 since this list is usually IP addresses). A host name is now passed as
5814 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5815 code that specifies a non-restarting handler (typically sigaction() in
5816 modern systems) in an attempt to fix a rare and obscure crash bug.
5818 15. Narrowed the window for a race in the daemon that could cause it to ignore
5819 SIGCHLD signals. This is not a major problem, because they are used only to
5820 wake it up if nothing else does.
5822 16. A malformed maildirsize file could cause Exim to calculate negative values
5823 for the mailbox size or file count. Odd effects could occur as a result.
5824 The maildirsize information is now recalculated if the size or filecount
5827 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5828 support for a long time. Removed HAVE_SYS_VFS_H.
5830 18. Updated exipick to current release from John Jetmore.
5832 19. Allow an empty sender to be matched against a lookup in an address list.
5833 Previously the only cases considered were a regular expression, or an
5836 20. Exim went into a mad DNS lookup loop when doing a callout where the
5837 host was specified on the transport, if the DNS lookup yielded more than
5840 21. The RFC2047 encoding function was originally intended for short strings
5841 such as real names; it was not keeping to the 75-character limit for
5842 encoded words that the RFC imposes. It now respects the limit, and
5843 generates multiple encoded words if necessary. To be on the safe side, I
5844 have increased the buffer size for the ${rfc2047: expansion operator from
5847 22. Failure to deliver a bounce message always caused it to be frozen, even if
5848 there was an errors_to setting on the router. The errors_to setting is now
5851 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5852 canonical form (fully expanded) before being placed in
5853 $sender_host_address.
5855 24. Updated eximstats to version 1.33
5857 25. Include certificate and key file names in error message when GnuTLS fails
5858 to set them up, because the GnuTLS error message doesn't include the name
5859 of the failing file when there is a problem reading it.
5861 26. Expand error message when OpenSSL has problems setting up cert/key files.
5864 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5865 (this can affect the format of dates).
5867 28. exim_tidydb, when checking for the continued existence of a message for
5868 which it has found a message-specific retry record, was not finding
5869 messages that were in split spool directories. Consequently, it was
5870 deleting retry records that should have stayed in existence.
5872 29. eximstats updated to version 1.35
5873 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5874 1.35 - bugfix such that pie charts by volume are generated correctly
5876 30. The SPA authentication driver was not abandoning authentication and moving
5877 on to the next authenticator when an expansion was forced to fail,
5878 contradicting the general specification for all authenticators. Instead it
5879 was generating a temporary error. It now behaves as specified.
5881 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5882 (the order specifies the preference for clients). The order is now AES256,
5883 AES128, 3DES, ARCFOUR128.
5885 31. Small patch to Sieve code - explicitly set From: when generating an
5888 32. Exim crashed if a remote delivery caused a very long error message to be
5889 recorded - for instance if somebody sent an entire SpamAssassin report back
5890 as a large number of 550 error lines. This bug was coincidentally fixed by
5891 increasing the size of one of Exim's internal buffers (big_buffer) that
5892 happened as part of the Exiscan merge. However, to be on the safe side, I
5893 have made the code more robust (and fixed the comments that describe what
5896 33. Some experimental protocols are using DNS PTR records for new purposes. The
5897 keys for these records are domain names, not reversed IP addresses. The
5898 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5899 leaves it alone. Component reversal etc. now happens only for IP addresses.
5902 34. The host_aton() function is supposed to be passed a string that is known
5903 to be a valid IP address. However, in the case of IPv6 addresses, it was
5904 not checking this. This is a hostage to fortune. Exim now panics and dies
5905 if the condition is not met. A case was found where this could be provoked
5906 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5907 components; fortuitously, this particular loophole had already been fixed
5908 by change 4.50/55 or 4.44/33 above.
5910 If there are any other similar loopholes, the new check in host_aton()
5911 itself should stop them being exploited. The report I received stated that
5912 data on the command line could provoke the exploit when Exim was running as
5913 exim, but did not say which command line option was involved. All I could
5914 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5915 running as the user.
5918 35. There was a buffer overflow vulnerability in the SPA authentication code
5919 (which came originally from the Samba project). I have added a test to the
5920 spa_base64_to_bits() function which I hope fixes it.
5923 36. The daemon start-up calls getloadavg() while still root for those OS that
5924 need the first call to be done as root, but it missed one case: when
5925 deliver_queue_load_max is set with deliver_drop_privilege. This is
5926 necessary for the benefit of the queue runner, because there is no re-exec
5927 when deliver_drop_privilege is set.
5929 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5930 was in use, was not putting the data itself into the right store pool;
5931 consequently, it could be overwritten for a subsequent message in the same
5932 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5935 38. Sometimes the final signoff response after QUIT could fail to get
5936 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5937 < 0 before doing a fflush(). This bug looks as though it goes back to the
5938 introduction of TLS in release 3.20, but "sometimes" must have been rare
5939 because the tests only now provoked it.
5945 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5946 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5947 Now it returns an integer. A number of calls were still expecting a T/F
5948 return. Fortuitously, in all cases, the tests worked in OK situations,
5949 which is the norm. However, things would have gone wrong on any write
5950 failures on the smtp file descriptor. This function is used when sending
5951 messages over SMTP and also when doing verify callouts.
5953 2. When Exim is called to do synchronous delivery of a locally submitted
5954 message (the -odf or -odi options), it no longer closes stderr before doing
5957 3. Implemented the mua_wrapper option.
5959 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5961 5. Implemented the functions header_remove(), header_testname(),
5962 header_add_at_position(), and receive_remove_recipient(), and exported them
5965 6. If an ACL "warn" statement specified the addition of headers, Exim already
5966 inserted X-ACL-Warn: at the start if there was no header name. However, it
5967 was not making this test for the second and subsequent header lines if
5968 there were newlines in the string. This meant that an invalid header could
5969 be inserted if Exim was badly configured.
5971 7. Allow an ACL "warn" statement to add header lines at the start or after all
5972 the Received: headers, as well as at the end.
5974 8. Added the rcpt_4xx retry error code.
5976 9. Added postmaster_mailfrom=xxx to callout verification option.
5978 10. Added mailfrom=xxxx to the callout verification option, for verify=
5981 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5982 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5984 12. Inserted some casts to stop certain compilers warning when using pointer
5985 differences as field lengths or precisions in printf-type calls (mostly
5986 affecting debugging statements).
5988 13. Added optional readline() support for -be (dynamically loaded).
5990 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5991 same clock tick as a message's arrival, so that its received time was the
5992 same as the "first fail" time on the retry record, and that message
5993 remained on the queue past the ultimate address timeout, every queue runner
5994 would try a delivery (because it was past the ultimate address timeout) but
5995 after another failure, the ultimate address timeout, which should have then
5996 bounced the address, did not kick in. This was a "< instead of <=" error;
5997 in most cases the first failure would have been in the next clock tick
5998 after the received time, and all would be well.
6000 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
6001 being recognized when the domain list was tested by the match_domain
6002 condition in an expansion string.
6004 16. Added the ${str2b64: operator.
6006 17. Exim was always calling setrlimit() to set a large limit for the number of
6007 processes, without checking whether the existing limit was already
6008 adequate. (It did check for the limit on file descriptors.) Furthermore,
6009 errors from getrlimit() and setrlimit() were being ignored. Now they are
6010 logged to the main and panic logs, but Exim does carry on, to try to do its
6011 job under whatever limits there are.
6013 18. Imported PCRE 5.0.
6015 19. Trivial typo in log message " temporarily refused connection" (the leading
6018 20. If the log selector return_path_on_delivery was set and an address was
6019 redirected to /dev/null, the delivery process crashed because it assumed
6020 that a return path would always be set for a "successful" delivery. In this
6021 case, the whole delivery is bypassed as an optimization, and therefore no
6024 21. Internal re-arrangement: the function for sending a challenge and reading
6025 a response while authentication was assuming a zero-terminated challenge
6026 string. It's now changed to take a pointer and a length, to allow for
6027 binary data in such strings.
6029 22. Added the cyrus_sasl authenticator (code supplied by MBM).
6031 23. Exim was not respecting finduser_retries when seeking the login of the
6032 uid under which it was called; it was always trying 10 times. (The default
6033 setting of finduser_retries is zero.) Also, it was sleeping after the final
6034 failure, which is pointless.
6036 24. Implemented tls_on_connect_ports.
6038 25. Implemented acl_smtp_predata.
6040 26. If the domain in control=submission is set empty, Exim assumes that the
6041 authenticated id is a complete email address when it generates From: or
6042 Sender: header lines.
6044 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
6045 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
6046 chown and chgrp in /bin and hostname in /usr/bin.
6048 28. Exim was keeping the "process log" file open after each use, just as it
6049 does for the main log. This opens the possibility of it remaining open for
6050 long periods when the USR1 signal hits a daemon. Occasional processlog
6051 errors were reported, that could have been caused by this. Anyway, it seems
6052 much more sensible not to leave this file open at all, so that is what now
6055 29. The long-running daemon process does not normally write to the log once it
6056 has entered its main loop, and it closes the log before doing so. This is
6057 so that log files can straightforwardly be renamed and moved. However,
6058 there are a couple of unusual error situations where the daemon does write
6059 log entries, and I had neglected to close the log afterwards.
6061 30. The text of an SMTP error response that was received during a remote
6062 delivery was being truncated at 512 bytes. This is too short for some of
6063 the long messages that one sometimes sees. I've increased the limit to
6066 31. It is now possible to make retry rules that apply only when a message has a
6067 specific sender, in particular, an empty sender.
6069 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
6070 it possible to be selective about when SMTP synchronization is enforced.
6072 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
6074 32. Implemented hosts_connection_nolog.
6076 33. Added an ACL for QUIT.
6078 34. Setting "delay_warning=" to disable warnings was not working; it gave a
6081 35. Added mailbox_size and mailbox_filecount to appendfile.
6083 36. Added control = no_multiline_responses to ACLs.
6085 37. There was a bug in the logic of the code that waits for the clock to tick
6086 in the case where the clock went backwards by a substantial amount such
6087 that the microsecond fraction of "now" was more than the microsecond
6088 fraction of "then" (but the whole seconds number was less).
6090 38. Added support for the libradius Radius client library this is found on
6091 FreeBSD (previously only the radiusclient library was supported).
6097 1. When certain lookups returned multiple values in the form name=value, the
6098 quoting of the values was not always being done properly. Specifically:
6099 (a) If the value started with a double quote, but contained no whitespace,
6101 (b) If the value contained whitespace other than a space character (i.e.
6102 tabs or newlines or carriage returns) it was not quoted.
6103 This fix has been applied to the mysql and pgsql lookups by writing a
6104 separate quoting function and calling it from the lookup code. The fix
6105 should probably also be applied to nisplus, ibase and oracle lookups, but
6106 since I cannot test any of those, I have not disturbed their existing code.
6108 2. A hit in the callout cache for a specific address caused a log line with no
6109 reason for rejecting RCPT. Now it says "Previous (cached) callout
6110 verification failure".
6112 3. There was an off-by-one bug in the queryprogram router. An over-long
6113 return line was truncated at 256 instead of 255 characters, thereby
6114 overflowing its buffer with the terminating zero. As well as fixing this, I
6115 have increased the buffer size to 1024 (and made a note to document this).
6117 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
6118 when Exim is waiting for an SMTP response from a remote server, Exim
6119 restarts its select() call on the socket, thereby resetting its timeout.
6120 This is not a problem when such interrupts are rare. Somebody set up a cron
6121 job to run exiwhat every 2 minutes, which is less than the normal select()
6122 timeout (5 or 10 minutes). This meant that the select() timeout never
6123 kicked in because it was always reset. I have fixed this by comparing the
6124 time when an interrupt arrives with the time at the start of the first call
6125 to select(). If more time than the timeout has elapsed, the interrupt is
6126 treated as a timeout.
6128 5. Some internal re-factoring in preparation for the addition of Sieve
6129 extensions (by MH). In particular, the "personal" test is moved to a
6130 separate function, and given an option for scanning Cc: and Bcc: (which is
6131 not set for Exim filters).
6133 6. When Exim created an email address using the login of the caller as the
6134 local part (e.g. when creating a From: or Sender: header line), it was not
6135 quoting the local part when it contained special characters such as @.
6137 7. Installed new OpenBSD configuration files.
6139 8. Reworded some messages for syntax errors in "and" and "or" conditions to
6140 try to make them clearer.
6142 9. Callout options, other than the timeout value, were being ignored when
6143 verifying sender addresses in header lines. For example, when using
6145 verify = header_sender/callout=no_cache
6147 the cache was (incorrectly) being used.
6149 10. Added a missing instance of ${EXE} to the exim_install script; this affects
6150 only the Cygwin environment.
6152 11. When return_path_on_delivery was set as a log selector, if different remote
6153 addresses in the same message used different return paths and parallel
6154 remote delivery occurred, the wrong values would sometimes be logged.
6155 (Whenever a remote delivery process finished, the return path value from
6156 the most recently started remote delivery process was logged.)
6158 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
6159 lines when AUTH and/or TLS are in use. This is the "received protocol"
6160 field. Exim used to use "asmtp" for authenticated SMTP, without any
6161 indication (in the protocol name) for TLS use. Now it follows the RFC and
6162 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
6163 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
6164 names appear in log lines as well as in Received: header lines.
6166 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
6167 extensions, and comparison tests, and to fix some bugs.
6169 14. Changes to the "personal" filter test:
6171 (1) The test was buggy in that it was just doing the equivalent of
6172 "contains" tests on header lines. For example, if a user's address was
6173 anne@some.where, the "personal" test would incorrectly be true for
6175 To: susanne@some.where
6177 This test is now done by extracting each address from the header in turn,
6178 and checking the entire address. Other tests that are part of "personal"
6179 are now done using regular expressions (for example, to check local parts
6180 of addresses in From: header lines).
6182 (2) The list of non-personal local parts in From: addresses has been
6183 extended to include "listserv", "majordomo", "*-request", and "owner-*",
6184 taken from the Sieve specification recommendations.
6186 (3) If the message contains any header line starting with "List-" it is
6187 treated as non-personal.
6189 (4) The test for "circular" in the Subject: header line has been removed
6190 because it now seems ill-conceived.
6192 15. Minor typos in src/EDITME comments corrected.
6194 16. Installed latest exipick from John Jetmore.
6196 17. If headers_add on a router specified a text string that was too long for
6197 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
6198 of string_sprintf() is now avoided.
6200 18. $message_body_size was not set (it was always zero) when running the DATA
6201 ACL and the local_scan() function.
6203 19. For the "mail" command in an Exim filter, no default was being set for
6204 the once_repeat time, causing a random time value to be used if "once" was
6205 specified. (If the value happened to be <= 0, no repeat happened.) The
6206 default is now 0s, meaning "never repeat". The "vacation" command was OK
6207 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
6208 (I found it when inspecting the code).
6210 20. There is now an overall timeout for performing a callout verification. It
6211 defaults to 4 times the callout timeout, which applies to individual SMTP
6212 commands during the callout. The overall timeout applies when there is more
6213 than one host that can be tried. The timeout is checked before trying the
6214 next host. This prevents very long delays if there are a large number of
6215 hosts and all are timing out (e.g. when the network connections are timing
6216 out). The value of the overall timeout can be changed by specifying an
6217 additional sub-option for "callout", called "maxwait". For example:
6219 verify = sender/callout=5s,maxwait=20s
6221 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
6222 to the end before writing, but this should make it even safer).
6224 22. Exim was forgetting that it had advertised PIPELINING for the second and
6225 subsequent messages on an SMTP connection. It was also not resetting its
6226 memory on STARTTLS and an internal HELO.
6228 23. When Exim logs an SMTP synchronization error within a session, it now
6229 records whether PIPELINING has been advertised or not.
6231 24. Added 3 instances of "(long int)" casts to time_t variables that were being
6232 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
6233 rather than long int.
6235 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
6237 26. Added the never_mail option to autoreply.
6243 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
6244 crash if the getsockname() call failed; this can happen if a connection is
6245 closed very soon after it is established. The problem was simply in the
6246 order in which certain operations were done, causing Exim to try to write
6247 to the SMTP stream before it had set up the file descriptor. The bug has
6248 been fixed by making things happen in the correct order.
6254 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
6255 before the connection was closed, thus losing the rejection response.
6257 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
6258 some early Solaris releases, but causes trouble in current releases where
6259 socklen_t is defined.
6261 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
6264 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
6267 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
6268 that releasing the top of it at the end releases what was used for sub-
6269 expansions (unless the block got too big). However, discard this block if
6270 the first thing is a variable or header, so that we can use its block when
6271 it is dynamic (useful for very large $message_headers, for example).
6273 6. Lookups now cache *every* query, not just the most recent. A new, separate
6274 store pool is used for this. It can be recovered when all lookup caches are
6275 flushed. Lookups now release memory at the end of their result strings.
6276 This has involved some general refactoring of the lookup sources.
6278 7. Some code has been added to the store_xxx() functions to reduce the amount
6279 of flapping under certain conditions.
6281 8. log_incoming_interface used to affect only the <= reception log lines. Now
6282 it causes the local interface and port to be added to several more SMTP log
6283 lines, for example "SMTP connection from", and rejection lines.
6285 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
6287 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
6289 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
6290 could be overwritten at the end of the current message (or the start of a
6291 new message if it was set in a HELO ACL). The value is now preserved for
6292 the duration of the SMTP connection.
6294 12. If a transport had a headers_rewrite setting, and a matching header line
6295 contained an unqualified address, that address was qualified, even if it
6296 did not match any rewriting rules. The underlying bug was that the values
6297 of the flags that permit the existence of unqualified sender and recipient
6298 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
6299 non-local messages, and by -bnq for local messages) were not being
6300 preserved with the message after it was received.
6302 13. When Exim was logging an SMTP synchronization error, it could sometimes log
6303 "next input=" as part of the text comprising the host identity instead of
6304 the correct text. The code was using the same buffer for two different
6305 strings. However, depending on which order the printing function evaluated
6306 its arguments, the bug did not always show up. Under Linux, for example, my
6307 test suite worked just fine.
6309 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
6310 doesn't work with some older versions of Perl. It has been changed to "my",
6311 which in any case is probably the better facility to use.
6313 15. A really picky compiler found some instances of statements for creating
6314 error messages that either had too many or two few arguments for the format
6317 16. The size of the buffer for calls to the DNS resolver has been increased
6318 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
6319 for addresses that have a lot of PTR records. This alleviates a problem; it
6320 does not fully solve it.
6322 17. A dnsdb lookup for PTR records that receives more data than will fit in the
6323 buffer now truncates the list and logs the incident, which is the same
6324 action as happens when Exim is looking up a host name and its aliases.
6325 Previously in this situation something unpredictable would happen;
6326 sometimes it was "internal error: store_reset failed".
6328 18. If a server dropped the connection unexpectedly when an Exim client was
6329 using GnuTLS and trying to read a response, the client delivery process
6330 crashed while trying to generate an error log message.
6332 19. If a "warn" verb in an ACL added multiple headers to a message in a single
6333 string, for example:
6335 warn message = H1: something\nH2: something
6337 the text was added as a single header line from Exim's point of view
6338 though it ended up OK in the delivered message. However, searching for the
6339 second and subsequent header lines using $h_h2: did not work. This has been
6340 fixed. Similarly, if a system filter added multiple headers in this way,
6341 the routers could not see them.
6343 20. Expanded the error message when iplsearch is called with an invalid key to
6344 suggest using net-iplsearch in a host list.
6346 21. When running tests using -bh, any delays imposed by "delay" modifiers in
6347 ACLs are no longer actually imposed (and a message to that effect is
6350 22. If a "gecos" field in a passwd entry contained escaped characters, in
6351 particular, if it contained a \" sequence, Exim got it wrong when building
6352 a From: or a Sender: header from that name. A second bug also caused
6353 incorrect handling when an unquoted " was present following a character
6354 that needed quoting.
6356 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
6357 was not being matched caselessly.
6359 24. Arranged for all hyphens in the exim.8 source to be escaped with
6362 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
6363 properly. Recipient callout cache records were still being keyed to include
6364 the sender, even when use_sender was set false. This led to far more
6365 callouts that were necessary. The sender is no longer included in the key
6366 when use_sender is false.
6368 26. Added "control = submission" modifier to ACLs.
6370 27. Added the ${base62d: operator to decode base 62 numbers.
6372 28. dnsdb lookups can now access SRV records.
6374 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
6375 the configuration file.
6377 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
6378 (-v) mode. This makes the output for a verbose queue run more intelligible.
6380 31. Added a use_postmaster feature to recipient callouts.
6382 32. Added the $body_zerocount variable, containing the number of binary zero
6383 bytes in the message body.
6385 33. The time of last modification of the "new" subdirectory is now used as the
6386 "mailbox time last read" when there is a quota error for a maildir
6389 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
6391 35. Added +ignore_unknown as a special item in host lists.
6393 36. Code for decoding IPv6 addresses in host lists is now included, even if
6394 IPv6 support is not being compiled. This fixes a bug in which an IPv6
6395 address was recognized as an IP address, but was then not correctly decoded
6396 into binary, causing unexpected and incorrect effects when compared with
6403 1. Very minor rewording of debugging text in manualroute to say "list of
6404 hosts" instead of "hostlist".
6406 2. If verify=header_syntax was set, and a header line with an unqualified
6407 address (no domain) and a large number of spaces between the end of the
6408 name and the colon was received, the reception process suffered a buffer
6409 overflow, and (when I tested it) crashed. This was caused by some obsolete
6410 code that should have been removed. The fix is to remove it!
6412 3. When running in the test harness, delay a bit after writing a bounce
6413 message to get a bit more predictability in the log output.
6415 4. Added a call to search_tidyup() just before forking a reception process. In
6416 theory, someone could use a lookup in the expansion of smtp_accept_max_
6417 per_host which, without the tidyup, could leave open a database connection.
6419 5. Added the variables $recipient_data and $sender_data which get set from a
6420 lookup success in an ACL "recipients" or "senders" condition, or a router
6421 "senders" option, similar to $domain_data and $local_part_data.
6423 6. Moved the writing of debug_print from before to after the "senders" test
6426 7. Change 4.31/66 (moving the time when the Received: is generated) caused
6427 problems for message scanning, either using a data ACL, or using
6428 local_scan() because the Received: header was not generated till after they
6429 were called (in order to set the time as the time of reception completion).
6430 I have revised the way this works. The header is now generated after the
6431 body is received, but before the ACL or local_scan() are called. After they
6432 are run, the timestamp in the header is updated.
6438 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
6439 before starting a queue runner without re-exec. This happened only when
6440 deliver_drop_privilege was set or when the Exim user was set to root. The
6441 effect of the bug was that timeouts during subsequent deliveries caused
6442 crashes instead of being properly handled. The handler is now left at its
6443 default (and expected) setting.
6445 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
6446 message, again when deliver_drop_privilege is set or Exim is run as root.
6447 The bug described in (1) was not present in this case, but the tidying up
6448 of the other signals was missing. I have made the two cases consistent.
6450 3. The ignore_target_hosts setting on a manualroute router was being ignored
6451 for hosts that were looked up using the /MX notation.
6453 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
6456 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
6457 operated on the sender address. After changing the $sender_address to <>
6458 for the sender address verify, Exim was re-instated it as the original
6459 (before rewriting) address, but remembering that it had rewritten it, so it
6460 wasn't rewriting it again. This bug also had the effect of breaking the
6461 sender address verification caching when the sender address was rewritten.
6463 6. The ignore_target_hosts option was being ignored by the ipliteral router.
6464 This has been changed so that if the ip literal address matches
6465 ignore_target_hosts, the router declines.
6467 7. Added expansion conditions match_domain, match_address, and match_local_
6468 part (NOT match_host).
6470 8. The placeholder for the Received: header didn't have a length field set.
6472 9. Added code to Exim itself and to exim_lock to test for a specific race
6473 condition that could lead to file corruption when using MBX delivery. The
6474 issue is with the lockfile that is created in /tmp. If this file is removed
6475 after a process has opened it but before that process has acquired a lock,
6476 there is the potential for a second process to recreate the file and also
6477 acquire a lock. This could lead to two Exim processes writing to the file
6478 at the same time. The added code performs the same test as UW imapd; it
6479 checks after acquiring the lock that its file descriptor still refers to
6480 the same named file.
6482 10. The buffer for building added header lines was of fixed size, 8192 bytes.
6483 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
6486 11. Added the smtp_active_hostname option. If used, this will typically be made
6487 to depend on the incoming interface address. Because $interface_address is
6488 not set up until the daemon has forked a reception process, error responses
6489 that can happen earlier (such as "too many connections") no longer contain
6492 12. If an expansion in a condition on a "warn" statement fails because a lookup
6493 defers, the "warn" statement is abandoned, and the next ACL statement is
6494 processed. Previously this caused the whole ACL to be aborted.
6496 13. Added the iplsearch lookup type.
6498 14. Added ident_timeout as a log selector.
6500 15. Added tls_certificate_verified as a log selector.
6502 16. Added a global option tls_require_ciphers (compare the smtp transport
6503 option of the same name). This controls incoming TLS connections.
6505 17. I finally figured out how to make tls_require_ciphers do a similar thing
6506 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
6507 before starting the TLS session.
6509 18. Tabs are now shown as \t in -bP output.
6511 19. If the log selector return_path_on_delivery was set, Exim crashed when
6512 bouncing a message because it had too many Received: header lines.
6514 20. If two routers both had headers_remove settings, and the first one included
6515 a superfluous trailing colon, the final name in the first list and the
6516 first name in the second list were incorrectly joined into one item (with a
6517 colon in the middle).
6523 1. Added -C and -D options to the exinext utility, mainly to make it easier
6524 to include in the automated testing, but these could be helpful when
6525 multiple configurations are in use.
6527 2. The exinext utility was not formatting the output nicely when there was
6528 an alternate port involved in the retry record key, nor when there was a
6529 message id as well (for retries that were specific to a specific message
6530 and a specific host). It was also confused by IPv6 addresses, because of
6531 the additional colons they contain. I have fixed the IPv4 problem, and
6532 patched it up to do a reasonable job for IPv6.
6534 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
6535 delivery, the log line now contains "pipelined" if PIPELINING was used.
6537 4. An SMTP transport process used to panic and die if the bind() call to set
6538 an explicit outgoing interface failed. This has been changed; it is now
6539 treated in the same way as a connect() failure.
6541 5. A reference to $sender_host_name in the part of a conditional expansion
6542 that was being skipped was still causing a DNS lookup. This no longer
6545 6. The def: expansion condition was not recognizing references to header lines
6546 that used bh_ and bheader_.
6548 7. Added the _cache feature to named lists.
6550 8. The code for checking quota_filecount in the appendfile transport was
6551 allowing one more file than it should have been.
6553 9. For compatibility with Sendmail, the command line option
6561 and sets the incoming protocol and host name (for trusted callers). The
6562 host name and its colon can be omitted when only the protocol is to be set.
6563 Note the Exim already has two private options, -pd and -ps, that refer to
6564 embedded Perl. It is therefore impossible to set a protocol value of "d" or
6565 "s", but I don't think that's a major issue.
6567 10. A number of refactoring changes to the code, none of which should affect
6570 (a) The number of logging options was getting close to filling up the
6571 32-bit word that was used as a bit map. I have split them into two classes:
6572 those that are passed in the argument to log_write(), and those that are
6573 only ever tested independently outside of that function. These are now in
6574 separate 32-bit words, so there is plenty of room for expansion again.
6575 There is no change in the user interface or the logging behaviour.
6577 (b) When building, for example, log lines, the code previously used a
6578 macro that called string_cat() twice, in order to add two strings. This is
6579 not really sufficiently general. Furthermore, there was one instance where
6580 it was actually wrong because one of the argument was used twice, and in
6581 one call a function was used. (As it happened, calling the function twice
6582 did not affect the overall behaviour.) The macro has been replaced by a
6583 function that can join an arbitrary number of extra strings onto a growing
6586 (c) The code for expansion conditions now uses a table and a binary chop
6587 instead of a serial search (which was left over from when there were very
6588 few conditions). Also, it now recognizes conditions like "pam" even when
6589 the relevant support is not compiled in: a suitably worded error message is
6590 given if an attempt is made to use such a condition.
6592 11. Added ${time_interval:xxxxx}.
6594 12. A bug was causing one of the ddress fields not to be passed back correctly
6595 from remote delivery subprocesses. The field in question was not being
6596 subsequently used, so this caused to problems in practice.
6598 13. Added new log selectors queue_time and deliver_time.
6600 14. Might have fixed a bug in maildirsizefile handling that threw up
6601 "unexpected character" debug warnings, and recalculated the data
6602 unnecessarily. In any case, I expanded the warning message to give more
6605 15. Added the message "Restricted characters in address" to the statements in
6606 the default ACL that block characters like @ and % in local parts.
6608 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
6609 Three changes have been made:
6611 (a) There was a serious bug; a negative response to MAIL caused the whole
6612 recipient domain to be cached as invalid, thereby blocking all messages
6613 to all local parts at the same domain, from all senders. This bug has
6614 been fixed. The domain is no longer cached after a negative response to
6615 MAIL if the sender used is not empty.
6617 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
6620 (c) A new callout option, "use_sender" has been added for people who want
6621 the modified behaviour.
6627 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
6630 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
6631 indeed breaks things for older releases.
6633 3. Added additional logging to the case where there is a problem reading data
6634 from a filter that is running in a subprocess using a pipe, in order to
6635 try to track down a specific problem.
6637 4. Testing facility fudge: when running in the test harness and attempting
6638 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
6639 getting "No route to host". Convert this to a timeout.
6641 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
6644 6. Some OS don't have socklen_t but use size_t instead. This affects the
6645 fifth argument of getsockopt() amongst other things. This is now
6646 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
6647 can be set for individual OS. I have set it for SunOS5, OSF1, and
6648 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
6649 some earlier ones do not.
6651 7. Change 4.30/15 was not doing the test caselessly.
6653 8. The standard form for an IPv6 address literal was being rejected by address
6654 parsing in, for example, MAIL and RCPT commands. An example of this kind of
6655 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6656 this, as well as the form without the "IPv6" on the front (but only when
6657 address literals are enabled, of course).
6659 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
6661 10. Exim crashed if a message with an empty sender address specified by -f
6662 encountered a router with an errors_to setting. This could be provoked only
6663 by a command such as
6667 where an empty string was supplied; "<>" did not hit this bug.
6669 11. Installed PCRE release 4.5.
6671 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
6672 remained set. It is now erased.
6674 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
6675 times from message ids (which are base 36 rather than the normal 62).
6677 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
6678 were being counted as actual protocol errors, and logged if the log
6679 selector +smtp_protocol_error was set. One cannot be perfect in this test,
6680 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
6681 and DATA following a set of rejected RCPTs do not count as protocol errors.
6682 In other words, Exim assumes they were pipelined, though this may not
6683 actually be the case. Of course, in all cases the client gets an
6684 appropriate error code.
6686 15. If a lookup fails in an ACL condition, a message about the failure may
6687 be available; it is used if testing the ACL cannot continue, because most
6688 such messages specify what the cause of the deferral is. However, some
6689 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
6690 that caused an old message to be retained and used if a later statement
6691 caused a defer, replacing the real cause of the deferral.
6693 16. If an IP address had so many PTR records that the DNS lookup buffer
6694 was not large enough to hold them, Exim could crash while trying to process
6695 the truncated data. It now detects and logs this case.
6697 17. Further to 4.21/58, another change has been made: if (and only if) the
6698 first line of a message (the first header line) ends with CRLF, a bare LF
6699 in a subsequent header line has a space inserted after it, so as not to
6700 terminate the header.
6702 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
6703 unnecessarily, used atoi() instead of strtol(), and didn't check the
6704 termination when getting file sizes from file names by regex.
6706 19. Completely re-implemented the support for maildirsize files, in the light
6707 of a number of problems with the previous contributed implementation
6708 (4.30/29). In particular:
6710 . If the quota is zero, the maildirsize file is maintained, but no quota is
6713 . If the maildir directory does not exist, it is created before any attempt
6714 to write a maildirsize file.
6716 . The quota value in the file is just a cache; if the quota is changed in
6717 the transport, the new value overrides.
6719 . A regular expression is available for excluding directories from the
6722 20. The autoreply transport checks the characters in options that define the
6723 message's headers; it allows continued headers, but it was checking with
6724 isspace() after an embedded newline instead of explicitly looking for a
6727 21. If all the "regular" hosts to which an address was routed had passed their
6728 expiry times, and had not reached their retry times, the address was
6729 bounced, even if fallback hosts were defined. Now Exim should go on to try
6732 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
6733 equivalent code in the SMTP transport. Some hosts send humungous responses
6734 to HELO/EHLO, more than 1024 it seems.
6736 23. Refactoring: code in filter.c used (void *) for "any old type" but this
6737 gives compiler warnings in some environments. I've now done it "properly",
6740 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
6741 (because of problems with the built-in one) was declared to return uschar *
6742 instead of char *, causing compiler failure.
6744 25. Fixed a file descriptor leak when processing alias/forward files.
6746 26. Fixed a minor format string issue in dbfn.c.
6748 27. Typo in exim.c: ("dmbnz" for "dbmnz").
6750 28. If a filter file refered to $h_xxx or $message_headers, and the headers
6751 contained RFC 2047 "words", Exim's memory could, under certain conditions,
6754 29. When a sender address is verified, it is cached, to save repeating the test
6755 when there is more than one recipient in a message. However, when the
6756 verification involves a callout, it is possible for different callout
6757 options to be set for different recipients. It is too complicated to keep
6758 track of this in the cache, so now Exim always runs a verification when a
6759 callout is required, relying on the callout cache for the optimization.
6760 The overhead is duplication of the address routing, but this should not be
6763 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
6764 to be verified with callout=postmaster, and the main callout worked but the
6765 postmaster check failed, the verification correctly failed. However, if a
6766 subsequent RCPT command asked for sender verification *without* the
6767 postmaster check, incorrect caching caused this verification also to fail,
6770 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
6771 it was not caching the DNS options (qualify_single, search_parents) that
6772 were used when the lookup failed. A subsequent lookup with different
6773 options therefore always gave the same answer, though there were cases
6774 where it should not have. (Example: a "domains = !$mx_any" option on a
6775 dnslookup router: the "domains" option is always processed without any
6776 widening, but the router might have qualify_single set.) Now Exim uses the
6777 cached value only when the same options are set.
6779 32. Added John Jetmore's "exipick" utility to the distribution.
6781 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
6782 than a timeout (e.g. a certificate is required, and is not provided), an
6783 Exim server now closes the connection immediately. Previously it waited for
6784 the client to close - but if the client is SSL, it seems that they each
6785 wait for each other, leading to a delay before one of them times out.
6787 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
6788 maintained 0.8.x compatibility because I don't think many are using it, and
6789 it is clearly obsolete.
6791 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
6794 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
6795 client certificate was expired. A simple patch fixes this, though I don't
6796 understand the full logic of why the verify callback is called multiple
6799 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
6800 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
6801 which causes problems with some clients (such as the Certicom SSL Plus
6802 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
6803 disables the coutermeasure allowing Eudora to connect."
6805 38. Exim was not checking that a write() to a log file succeeded. This could
6806 lead to Bad Things if a log got too big, in particular if it hit a file
6807 size limit. Exim now panics and dies if it cannot write to a log file, just
6808 as it does if it cannot open a log file.
6810 39. Modified OS/Makefile-Linux so that it now contains
6812 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
6814 The two -D definitions ensure that Exim is compiled with large file
6815 support, which makes it possible to handle log files that are bigger than
6818 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
6819 instance) a domain was checked against a named list that involved a lookup,
6820 causing $domain_data to be set, then another domain was checked against the
6821 same list, then the first domain was re-checked, the value of $domain_data
6822 after the final check could be wrong. In particular, if the second check
6823 failed, it could be set empty. This bug probably also applied to
6826 41. The strip_trailing_dot option was not being applied to the address given
6827 with the -f command-line option.
6829 42. The code for reading a message's header from the spool was incrementing
6830 $received_count, but never initializing it. This meant that the value was
6831 incorrect (doubled) while delivering a message in the same process in which
6832 it was received. In the most common configuration of Exim, this never
6833 happens - a fresh exec is done - but it can happen when
6834 deliver_drop_privilege is set.
6836 43. When Exim logs an SMTP synchronization error - client data sent too soon -
6837 it now includes up to 150 characters of the unexpected data in the log
6840 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6841 and building data strings. The size of both of these buffers was 10 000
6842 bytes - far larger than anybody would *ever* want, thought I. Needless to
6843 say, somebody hit the limit. I have increased the maximum line length to
6844 20 000 and the maximum data length of concatenated lines to 100 000. I have
6845 also fixed two bugs, because there was no checking on these buffers. Tsk,
6846 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6847 buffer is too small.
6849 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6850 lsearch lookups. Now it does.
6852 46. When parsing a route_list item in a manualroute router, a fixed-length
6853 buffer was used for the list of hosts. I made this 1024 bytes long,
6854 thinking that nobody would ever have a list of hosts that long. Wrong.
6855 Somebody had a whole pile of complicated expansion conditions, and the
6856 string was silently truncated, leading to an expansion error. It turns out
6857 that it is easier to change to an unlimited length (owing to other changes
6858 that have happened since this code was originally written) than to build
6859 structure for giving a limitation error. The length of the item that
6860 expands into the list of hosts is now unlimited.
6862 47. The lsearch lookup could not handle data where the length of text line was
6863 more than 4095 characters. Such lines were truncated, leading to shortened
6864 data being returned. It should now handle lines of any length.
6866 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6867 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6870 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6871 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6872 no longer happens in comment lines. A list of replacements is now placed
6873 at the head of all of the source files, except those whose only change is
6874 to replace PERL_COMMAND in the very first #! line.
6876 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6877 messages on the queue, with a bottom-up merge sort, using code contributed
6878 by Michael Haardt. This should make operations like -bp somewhat faster on
6879 large queues. It won't affect queue runners, except when queue_run_in_order
6882 51. Installed eximstats 1.31 in the distribution.
6884 52. Added support for SRV lookups to the dnslookup router.
6886 53. If an ACL referred to $message_body or $message_body_end, the value was not
6887 reset for any messages that followed in the same SMTP session.
6889 54. The store-handling optimization for building very long strings was not
6890 differentiating between the different store pools. I don't think this
6891 actually made any difference in practice, but I've tidied it.
6893 55. While running the routers to verify a sender address, $sender_address
6894 was still set to the sender address. This is wrong, because when routing to
6895 send a bounce to the sender, it would be empty. Therefore, I have changed
6896 it so that, while verifying a sender address, $sender_address is set to <>.
6897 (There is no change to what happens when verifying a recipient address.)
6899 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6900 target A or AAAA records (if not already returned) without resetting the
6901 qualify_single or search_parents options of the DNS resolver. These are
6902 inappropriate in this case because the targets of MX and SRV records must
6903 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6904 target that, when qualified, matched something in the local domain. These
6905 two options are now turned off when doing these lookups.
6907 57. It seems that at least some releases of Reiserfs (which does not have the
6908 concept of a fixed number of inodes) returns zero and not -1 for the
6909 number of available inodes. This interacted badly with check_spool_inodes,
6910 which assumed that -1 was the "no such thing" setting. What I have done is
6911 to check that the total number of inodes is greater than zero before doing
6912 the test of how many are available.
6914 58. When a "warn" ACL statement has a log_message modifier, the message is
6915 remembered, and not repeated. This is to avoid a lot of repetition when a
6916 message has many recipients that cause the same warning to be written.
6917 However, Exim was preserving the list of already written lines for an
6918 entire SMTP session, which doesn't seem right. The memory is now reset if a
6919 new message is started.
6921 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6922 debugging output unless log_rewrite was also set.
6924 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6925 of (int)(handle) when we know that handle contains (void *)(-1).
6927 61. The Exim daemon panic-logs an error return when it closes the incoming
6928 connection. However "connection reset by peer" seems to be common, and
6929 isn't really an error worthy of noting specially, so that particular error
6932 62. When Exim is trying to find all the local interfaces, it used to panic and
6933 die if the ioctl to get the interface flags failed. However, it seems that
6934 on at least one OS (Solaris 9) it is possible to have an interface that is
6935 included in the list of interfaces, but for which you get a failure error
6936 for this call. This happens when the interface is not "plumbed" into a
6937 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6938 failure of the "get flags" call assumes that the interface is down.
6940 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6941 makes life easier for people who are doing arithmetic on fields extracted
6942 from dates, where you often get leading zeros that should not be
6943 interpreted as octal.
6945 64. Added qualify_domain to the redirect router, to override the global
6948 65. If a pathologically long header line contained very many addresses (the
6949 report of this problem mentioned 10 000) and each of them was rewritten,
6950 Exim could use up a very large amount of memory. (It kept on making new
6951 copies of the header line as it rewrote, and never released the old ones.)
6952 At the expense of a bit more processing, the header rewriting function has
6953 been changed so that it no longer eats memory in this way.
6955 66. The generation of the Received: header has been moved from the time that a
6956 message starts to be received, to the time that it finishes. The timestamp
6957 in the Received: header should now be very close to that of the <= log
6958 line. There are two side-effects of this change:
6960 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6961 logged header lines no longer include the local Received: line, because
6962 it has not yet been created. The same applies to a copy of the message
6963 that is returned to a non-SMTP sender when a message is rejected.
6965 (b) When a filter file is tested using -bf, no additional Received: header
6966 is added to the test message. After some thought, I decided that this
6969 This change does not affect the value of $received_for. It is still set
6970 after address rewriting, but before local_scan() is called.
6972 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6974 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6975 gave an unhelpful panic error message, and a defer error. I have managed to
6976 change this behaviour so that it now rejects any supplied certificate,
6977 which seems right, as the list of acceptable certificates is empty.
6979 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6980 gave an unhelpful defer error. I have not managed to make this reject any
6981 supplied certificates, but the error message it gives is "no certificate
6982 supplied", which is not helpful.
6984 70. exigrep's output now also includes lines that are not associated with any
6985 message, but which match the given pattern. Implemented by a patch from
6986 Martin Sluka, which also tidied up the Perl a bit.
6988 71. Recipient callout verification, like sender verification, was using <> in
6989 the MAIL FROM command. This isn't really the right thing, since the actual
6990 sender may affect whether the remote host accepts the recipient or not. I
6991 have changed it to use the actual sender in the callout; this means that
6992 the cache record is now keyed on a recipient/sender pair, not just the
6993 recipient address. There doesn't seem to be a real danger of callout loops,
6994 since a callout by the remote host to check the sender would use <>.
6995 [SEE ABOVE: changed after hitting problems.]
6997 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6998 temporary errors. However, in the case of such a code being given after
6999 the end of a data transmission (i.e. after ".") Exim was failing to write
7000 a retry record for the message. (Yes, there was some broken host that was
7001 actually sending 8xx at this point.)
7003 73. An unknown lookup type in a host list could cause Exim to panic-die when
7004 the list was checked. (An example that provoked this was putting <; in the
7005 middle of a list instead of at the start.) If this happened during a DATA
7006 ACL check, a -D file could be left lying around. This kind of configuration
7007 error no longer causes Exim to die; instead it causes a defer error. The
7008 incident is still logged to the main and panic logs.
7010 74. Buglet left over from Exim 3 conversion. The message "too many messages
7011 in one connection" was written to the rejectlog but not the mainlog, except
7012 when address rewriting (yes!) was being logged.
7014 75. Added write_rejectlog option.
7016 76. When a system filter was run not as root (that is, when system_filter_user
7017 was set), the values of the $n variables were not being returned to the
7018 main process; thus, they were not subsequently available in the $sn
7021 77. Added +return_path_on_delivery log selector.
7023 78. A connection timeout was being treated differently from recipients deferred
7024 when testing hosts_max_try with a message that was older than the host's
7025 retry timeout. (The host should not be counted, thus allowing all hosts to
7026 be tried at least once before bouncing.) This may have been the cause of an
7027 occasionally reported bug whereby a message would remain on the queue
7028 longer than the retry timeout, but would be bounced if a delivery was
7029 forced. I say "may" because I never totally pinned down the problem;
7030 setting up timeout/retry tests is difficult. See also the next item.
7032 79. The ultimate address timeout was not being applied to errors that involved
7033 a combination of host plus message (for example, a timeout on a MAIL
7034 command). When an address resolved to a number of possible hosts, and they
7035 were not all tried for each delivery (e.g. because of hosts_max_try), a
7036 message could remain on the queue longer than the retry timeout.
7038 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
7041 81. Fixed an obscure SMTP outgoing bug which required at least the following
7042 conditions: (a) there was another message waiting for the same server;
7043 (b) the server returned 5xx to all RCPT commands in the first message so
7044 that the message was not completed; (c) the server dropped the connection
7045 or gave a negative response to the RSET that Exim sends to abort the
7046 transaction. The observed case was a dropped connection after DATA that had
7047 been sent in pipelining mode. That is, the server had advertised PIPELINING
7048 but was not implementing it correctly. The effect of the bug was incorrect
7049 behaviour, such as trying another host, and this could lead to a crash.
7055 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
7056 and daemon.c were passed as pointers to ints; they should have been
7057 pointers to socklen_t variables (which are typically unsigned ints).
7059 2. Some signed/unsigned type warnings in the os.c file for Linux have been
7062 3. Fixed a really odd bug that affected only the testing scheme; patching a
7063 certain fixed string in the binary changed the value of another string that
7064 happened to be identical to the end of the original first string.
7066 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
7067 name", it returns that address as the IP address. On some operating
7068 systems (e.g. Solaris), it also passes back the IP address string as the
7069 "host name". However, on others (e.g. Linux), it passes back an empty
7070 string. Exim wasn't checking for this, and was changing the host name to an
7071 empty string, assuming it had been canonicalized.
7073 5. Although rare, it is permitted to have more than one PTR record for a given
7074 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
7075 all the names associated with an address, because they do in Solaris.
7076 However, it seems that they do not in Linux for data that comes from the
7077 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
7078 I found this out when I moved to a new Linux workstation and tried to run
7079 the Exim test suite.
7081 To get round this problem I have changed the code so that it now does its
7082 own call to the DNS to look up PTR records when searching for a host name.
7083 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
7084 addresses that are only in /etc/hosts are still found.
7086 This behaviour is, however, controlled by an option called host_lookup_
7087 order, which defaults to "bydns:byaddr". If people want to use the other
7088 order, or indeed, just use one or the other means of lookup, they can
7089 specify it in this variable.
7091 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
7092 some operating systems, this comes back from gethostbyaddr() as an empty
7093 string, and this is what Exim used to test for. However, it seems that in
7094 other systems, "." is yielded. Exim now tests for this case too.
7096 7. The values of check_spool_space and check_log_space are now held internally
7097 as a number of kilobytes instead of an absolute number of bytes. If a
7098 numbers is specified without 'K' or 'M', it is rounded up to the nearest
7099 kilobyte. This means that much larger values can be stored.
7101 8. Exim monitor: an attempt to get the action menu when not actually pointing
7102 at a message produces an empty menu entitled "No message selected". This
7103 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
7104 no entries in it ("Shell widget menu has zero width and/or height"). So I
7105 have added a single, blank menu entry in this case.
7107 9. Added ${quote_local_part.
7109 10. MIME decoding is now applied to the contents of Subject: header lines when
7112 11. Now that a reference to $sender_host_address automatically causes a reverse
7113 lookup to occur if necessary (4.13/18), there is no need to arrange for a
7114 host lookup before query-style lookups in lists that might use this
7115 variable. This has therefore been abolished, and the "net-" prefix is no
7116 longer necessary for query-style lookups.
7118 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
7119 have been a typo for LFLAGS, so it has been changed.
7121 13. The install script calls Exim with "-C /dev/null" in order to find the
7122 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
7123 to be output. However, since Exim outputs its version number before the
7124 error, it didn't break the script. It just looked ugly. I fixed this by
7125 always allowing "-C /dev/null" if the caller is root.
7127 14. Ignore overlarge ACL variable number when reading spool file - insurance
7128 against a later release with more variables having written the file.
7130 15. The standard form for an IPv6 address literal was being rejected by EHLO.
7131 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
7132 this, as well as the form without the "IPv6" on the front.
7134 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
7135 OS/Makefile-Darwin file.
7137 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
7138 only to LDAP libraries that do not have LDAP_OPT_DEREF.
7140 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
7141 variable. However, some OS use ints for inodes. I've added cast to long int
7142 to get rid of the compiler warning.
7144 19. I had forgotten to lock out "/../" in configuration file names when
7145 ALT_CONFIG_PREFIX was set.
7147 20. Routers used for verification do not need to specify transports. However,
7148 if such a router generated a host list, and callout was configured, Exim
7149 crashed, because it could not find a port number from the (non-existent)
7150 transport. It now assumes port 25 in this circumstance.
7152 21. Added the -t option to exigrep.
7154 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
7155 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
7156 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
7157 them is set, LOOKUP_LSEARCH is forced.
7159 23. "exim -bV" now outputs a list of lookups that are included in the binary.
7161 24. Added sender and host information to the "rejected by local_scan()" log
7162 line; previously there was no indication of these.
7164 25. Added .include_if_exists.
7166 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
7167 Linux. It turns out that not all file systems support this. Apparently some
7168 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
7169 do it.) To cope with this, the error EINVAL, which means that sync-ing is
7170 not supported on the file descriptor, is now ignored when Exim is trying to
7171 sync a directory. This applies only to Linux.
7173 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
7175 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
7176 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
7179 29. Added support for maildirsize files from supplied patch (modified a bit).
7181 30. The use of :fail: followed by an empty string could lead Exim to respond to
7182 sender verification failures with (e.g.):
7184 550 Verification failed for <xxx>
7185 550 Sender verify failed
7187 where the first response line was missing the '-' that indicates it is not
7188 the final line of the response.
7190 31. The loop for finding the name of the user that called Exim had a hardwired
7191 limit of 10; it now uses the value of finduser_retries, which is used for
7192 all other user lookups.
7194 32. Added $received_count variable, available in data and not_smtp ACLs, and at
7197 33. Exim was neglecting to zero errno before one call of strtol() when
7198 expanding a string and expecting an integer value. On some systems this
7199 resulted in spurious "integer overflow" errors. Also, it was casting the
7200 result into an int without checking.
7202 34. Testing for a connection timeout using "timeout_connect" in the retry rules
7203 did not work. The code looks as if it has *never* worked, though it appears
7204 to have been documented since at least release 1.62. I have made it work.
7206 35. The "timeout_DNS" error in retry rules, also documented since at least
7207 1.62, also never worked. As it isn't clear exactly what this means, and
7208 clearly it isn't a major issue, I have abolished the feature by treating it
7209 as "timeout", and writing a warning to the main and panic logs.
7211 36. The display of retry rules for -brt wasn't always showing the error code
7214 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
7215 timeout_connect_A, timeout_connect_MX.
7217 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
7218 to the empty sender.
7220 39. The daemon was not analysing the content of -oX till after it had closed
7221 stderr and disconnected from the controlling terminal. This meant that any
7222 syntax errors were only noted on the panic log, and the return code from
7223 the command was 0. By re-arranging the code a little, I've made the
7224 decoding happen first, so such errors now appear on stderr, and the return
7225 code is 1. However, the actual setting up of the sockets still happens in
7226 the disconnected process, so errors there are still only recorded on the
7229 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
7230 connections (as happens on some IP stacks) was logged at start up time as
7231 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
7232 it from "IPv6 and IPv4", which means that two separate sockets are being
7235 41. The debug output for gethostbyname2() or getipnodebyname() failures now
7236 says whether AF_INET or AF_INET6 was passed as an argument.
7238 42. Exiwhat output was messed up when time zones were included in log
7241 43. Exiwhat now gives more information about the daemon's listening ports,
7242 and whether -tls-on-connect was used.
7244 44. The "port" option of the smtp transport is now expanded.
7246 45. A "message" modifier in a "warn" statement in a non-message ACL was being
7247 silently ignored. Now an error message is written to the main and panic
7250 46. There's a new ACL modifier called "logwrite" which writes to a log file
7251 as soon as it is encountered.
7253 47. Added $local_user_uid and $local_user_gid at routing time.
7255 48. Exim crashed when trying to verify a sender address that was being
7258 49. Exim was recognizing only a space character after ".include". It now also
7259 recognizes a tab character.
7261 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
7262 extracting the relevant information from the specification. The man page no
7263 longer contains scrambled data for the -d option, and I've added a section
7264 at the front about calling Exim under different names.
7266 51. Added "extra_headers" argument to the "mail" command in filter files.
7268 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
7271 53. Installed eximstats 1.29.
7273 54. Added transport_filter_timeout as a generic transport option.
7275 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
7276 Cc: header lines. This was required by RFC 822, but it not required by RFC
7279 56. Exim used to add From:, Date:, and Message-Id: header lines to any
7280 incoming messages that did not have them. Now it does so only if the
7281 message originates locally, that is, if there is no associated remote host
7282 address. When Resent- header lines are present, this applies to the Resent-
7283 lines rather than the non-Resent- lines.
7285 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
7286 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
7288 58. Messages for configuration errors now include the name of the main
7289 configuration file - useful now that there may be more than one file in a
7290 list (.included file names were always shown).
7292 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
7293 for those rare installations that do not start the daemon as root or run it
7294 setuid root. I've cut out the call to initgroups() if the daemon is not
7297 60. The Exim user and group can now be bound into the binary as text strings
7298 that are looked up at the start of Exim's processing.
7300 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
7302 62. Added $mailstore_basename variable.
7304 63. Installed patch to sieve.c from Michael Haardt.
7306 64. When Exim failed to open the panic log after failing to open the main log,
7307 the original message it was trying to log was written to stderr and debug
7308 output, but if they were not available (the usual case in production), it
7309 was lost. Now it is written to syslog before the two lines that record the
7310 failures to open the logs.
7312 65. Users' Exim filters run in subprocesses under the user's uid. It is
7313 possible for a "deliver" command or an alias in a "personal" command to
7314 provoke an address rewrite. If logging of address rewriting is configured,
7315 this fails because the process is not running as root or exim. There may be
7316 a better way of dealing with this, but for the moment (because 4.30 needs
7317 to be released), I have disabled address rewrite logging when running a
7318 filter in a non-root, non-exim process.
7324 1. The buildconfig auxiliary program wasn't quoting the value set for
7325 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
7326 not defined. This bug was masked in 4.22 by the effect that was fixed in
7329 2. Some messages that were rejected after a message id was allocated were
7330 shown as "incomplete" by exigrep. It no longer does this for messages that
7331 are rejected by local_scan() or the DATA or non-SMTP ACLs.
7333 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
7334 have allow_domain_literals set, the ID did not get logged in the <= line.
7335 Domain literals are now always recognized in Message-ID: header lines.
7337 4. The first argument for a ${extract expansion item is the key name or field
7338 number. Leading and trailing spaces in this item were not being ignored,
7339 causing some misleading effects.
7341 5. When deliver_drop_privilege was set, single queue runner processes started
7342 manually (i.e. by the command "exim -q") or by the daemon (which uses the
7343 same command in the process it spins off) were not dropping privilege.
7345 6. When the daemon running as "exim" started a queue runner, it always
7346 re-executed Exim in the spun-off process. This is a waste of effort when
7347 deliver_drop_privilege is set. The new process now just calls the
7348 queue-runner function directly.
7354 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
7357 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
7358 ignored. Though the use of -oP was forcing the writing of a pid file, it
7359 was always written to the default place.
7361 3. If the message "no IP address found for host xxxx" is generated during
7362 incoming verification, it is now followed by identification of the incoming
7363 connection (so you can more easily find what provoked it).
7365 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
7367 5. Added some features to "harden" Exim a bit more against certain attacks:
7369 (a) There is now a build-time option called FIXED_NEVER_USERS that can
7370 be put in Local/Makefile. This is like the never_users runtime option,
7371 but it cannot be overridden. The default setting is "root".
7373 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
7374 prefix string with which any file named in a -C command line option
7377 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
7378 is retained for -C and -D only if the caller of Exim is root. Without
7379 it, the exim user may also use -C and -D and retain privilege.
7381 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
7382 command line option is disabled.
7384 6. Macro names set by the -D option must start with an upper case letter, just
7385 like macro names defined in the configuration file.
7387 7. Added "dereference=" facility to LDAP.
7389 8. Two instances of the typo "uknown" in the source files are fixed.
7391 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
7392 the Configure-Makefile script screwed up while processing it.
7394 10. Incorporated PCRE 4.4.
7396 11. The SMTP synchronization check was not operating right at the start of an
7397 SMTP session. For example, it could not catch a HELO sent before the client
7398 waited for the greeting. There is now a check for outstanding input at the
7399 point when the greeting is written. Because of the duplex, asynchronous
7400 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
7401 way, but not yet received, when the check is performed.
7403 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
7404 on TCP/IP sockets, because this apparently causes some broken clients to
7407 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
7408 unchanged) from the Cygwin maintainer.
7410 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
7411 was supported instead of testing for mbx. Effectively a typo.
7413 15. The spa authenticator server code was not checking that the input it
7414 received was valid base64.
7416 16. The debug output line for the "set" modifier in ACLs was not showing the
7417 name of the variable that was being set.
7419 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
7421 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
7422 lookup on its own. Something else was needed to trigger the lookup. For
7423 example, a match in host_lookup or the need for a host name in a host list.
7424 Now, if $sender_host_name is referenced and the host name has not yet been
7425 looked up, a lookup is performed. If the lookup fails, the variable remains
7426 empty, and $host_lookup_failed is set to "1".
7428 19. Added "eqi" as a case-independent comparison operator.
7430 20. The saslauthd authentication condition could segfault if neither service
7431 nor realm was specified.
7433 21. If an overflowing value such as "2048M" was set for message_size_limit, the
7434 error message that was logged was misleading, and incoming SMTP
7435 connections were dropped. The message is now more accurate, and temporary
7436 errors are given to SMTP connections.
7438 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
7439 (except RSET) with a 421 error, until QUIT is received. However, it was
7440 failing to send a response to QUIT.
7442 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
7443 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
7444 ACL is now run after the helo_try_verify_hosts code.
7446 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
7447 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
7448 case-independent, so other case variants are also recognized). Apparently
7449 some systems use these upper case variants.
7451 25. If more than two messages were waiting for the same host, and a transport
7452 filter was specified for the transport, Exim sent two messages over the
7453 same TCP/IP connection, and then failed with "socket operation on non-
7454 socket" when it tried to send the third.
7456 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
7458 27. The extern definition of crypt16() in expand.c was not being excluded when
7459 the OS had its own crypt16() function.
7461 28. Added bounce_return_body as a new option, and bounce_return_size_limit
7462 as a preferred synonym for return_size_limit, both as an option and as an
7465 29. Added LIBS=-liconv to OS/Makefile-OSF1.
7467 30. Changed the default configuration ACL to relax the local part checking rule
7468 for addresses that are not in any local domains. For these addresses,
7469 slashes and pipe symbols are allowed within local parts, but the sequence
7470 /../ is explicitly forbidden.
7472 31. SPA server authentication was not clearing the challenge buffer before
7475 32. log_message in a "warn" ACL statement was writing to the reject log as
7476 well as to the main log, which contradicts the documentation and doesn't
7477 seem right (because no rejection is happening). So I have stopped it.
7479 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
7480 However, I am unable to do any testing of this.
7482 34. Fixed an infelicity in the appendfile transport. When checking directories
7483 for a mailbox, to see if any needed to be created, it was accidentally
7484 using path names with one or more superfluous leading slashes; tracing
7485 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
7487 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
7488 contents are added to the log line that is written for every discarded
7489 recipient. (Previously a log_message setting was ignored.)
7491 36. The ${quote: operator now quotes the string if it is empty.
7493 37. The install script runs exim in order to find its version number. If for
7494 some reason other than non-existence or emptiness, which it checks, it
7495 could not run './exim', it was installing it with an empty version number,
7496 i.e. as "exim-". This error state is now caught, and the installation is
7499 38. An argument was missing from the function that creates an error message
7500 when Exim fails to connect to the socket for saslauthd authentication.
7501 This could cause Exim to crash, or give a corrupted message.
7503 39. Added isip, isip4, and isip6 to ${if conditions.
7505 40. The ACL variables $acl_xx are now saved with the message, and can be
7506 accessed later in routers, transports, and filters.
7508 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
7509 strings in the file are not string-expanded.
7511 42. If a MAIL command specified a SIZE value that was too large to fit into an
7512 int variable, the check against message_size_limit failed. Such values are
7513 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
7514 day this will have to be increased, but I don't think I want to be around
7515 when emails are that large.
7522 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
7523 iconv() is not standard in FreeBSD.
7525 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
7526 IPv6 enabled. The observed symptom was a segmentation fault on return from
7527 the function os_common_find_running_interfaces() in src/os.c.
7529 3. In the check_special_case() function in daemon.c I had used "errno" as an
7530 argument name, which causes warnings on some systems. This was basically a
7531 typo, since it was named "eno" in the comments!
7533 4. The code that waits for the clock to tick (at a resolution of some fraction
7534 of a second) so as to ensure message-id uniqueness was always waiting for
7535 at least one whole tick, when it could have waited for less. [This is
7536 almost certainly not relevant at current processor speeds, where it is
7537 unlikely to ever wait at all. But we try to future-proof.]
7539 5. The function that sleeps for a time interval that includes fractions of a
7540 second contained a race. It did not block SIGALRM between setting the
7541 timer, and suspending (a couple of lines later). If the interval was short
7542 and the sigsuspend() was delayed until after it had expired, the suspension
7543 never ended. On busy systems this could lead to processes getting stuck for
7546 6. Some uncommon configurations may cause a lookup to happen in a queue runner
7547 process, before it forks any delivery processes. The open lookup caching
7548 mechanism meant that the open file or database connection was passed into
7549 the delivery process. The problem was that delivery processes always tidy
7550 up cached lookup data. This could cause a problem for the next delivery
7551 process started by the queue runner, because the external queue runner
7552 process does not know about the closure. So the next delivery process
7553 still has data in the lookup cache. In the case of a file lookup, there was
7554 no problem because closing a file descriptor in a subprocess doesn't affect
7555 the parent. However, if the lookup was caching a connection to a database,
7556 the connection was closed, and the second delivery process was likely to
7557 see errors such as "PGSQL: query failed: server closed the connection
7558 unexpectedly". The problem has been fixed by closing all cached lookups
7559 in a queue runner before running a delivery process.
7561 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
7562 seem to have the "const" qualifier which it has on other OS. I've
7565 8. Change 4.21/2 was too strict. It is only if there are two authenticators
7566 *of the same type* (client or server) with the same public name that an
7567 error should be diagnosed.
7569 9. When Exim looked up a host name for an IP address, but failed to find the
7570 original IP address when looking up the host name (a safety check), it
7571 output the message "<ip address> does not match any IP for NULL", which was
7572 confusing, to say the least. The bug was that the host name should have
7573 appeared instead of "NULL".
7575 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
7576 user that is built into the binary, and the -C or -D options is used, root
7577 privilege is dropped before the configuration file is read. In addition,
7578 logging is switched to stderr instead of the normal log files. If the
7579 configuration then re-defines the Exim user, the unprivileged environment
7580 is probably not what is expected, so Exim logs a panic warning message (but
7583 However, if deliver_drop_privilege is set, the unprivileged state may well
7584 be exactly what is intended, so the warning has been cut out in that case,
7585 and Exim is allowed to try to write to its normal log files.
7591 1. smtp_return_error_details was not giving details for temporary sender
7592 or receiver verification errors.
7594 2. Diagnose a configuration error if two authenticators have the same public
7597 3. Exim used not to create the message log file for a message until the first
7598 delivery attempt. This could be confusing when incoming messages were held
7599 for policy or load reasons. The message log file is now created at the time
7600 the message is received, and an initial "Received" line is written to it.
7602 4. The automatically generated man page for command line options had a minor
7603 bug that caused no ill effects; however, a more serious problem was that
7604 the procedure for building the man page automatically didn't always
7605 operate. Consequently, release 4.20 contains an out-of-date version. This
7606 shouldn't happen again.
7608 5. When building Exim with embedded Perl support, the script that builds the
7609 Makefile was calling 'perl' to find its compile-time parameters, ignoring
7610 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
7612 6. The freeze_tell option was not being used for messages that were frozen on
7613 arrival, either by an ACL or by local_scan().
7615 7. Added the smtp_incomplete_transaction log selector.
7617 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
7618 was accepting AUTH without a new EHLO.
7620 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
7621 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
7624 10. Logging of TCP/IP connections (when configured) now happens in the main
7625 daemon process instead of the child process, so that the TCP/IP connection
7626 count is more accurate (but it can never be perfect).
7628 11. The use of "drop" in a nested ACL was not being handled correctly in the
7629 outer ACL. Now, if condition failure induced by the nested "drop" causes
7630 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
7631 or "require"), the connection is dropped.
7633 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
7634 that yield "discard" can now be used with an "accept" or a "discard" verb,
7635 but an error is generated for any others (because I can't see a useful way
7636 to define what should happen).
7638 13. When an ACL is read dynamically from a file (or anywhere else), the lines
7639 are now processed in the same way as lines in the Exim configuration file.
7640 In particular, continuation lines are supported.
7642 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
7644 15. Added -ti meaning -t -i.
7646 16. Check for letters, digits, hyphens, and dots in the names of dnslist
7647 domains, and warn by logging if others are found.
7649 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
7650 returned from GIFCONF when Exim is trying to find the list of interfaces on
7651 a host. The code in os.c has been modified to copy each ifreq to an aligned
7652 structure in all cases.
7654 Also, in some cases, the returned ifreq's were being copied to a 'struct
7655 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
7656 means the last couple of bytes of an IPv6 address could be chopped if the
7657 ifreq contained only a normal sockaddr (14 bytes storage).
7659 18. Named domain lists were not supported in the hosts_treat_as_local option.
7660 An entry such as +xxxx was not recognized, and was treated as a literal
7663 19. Ensure that header lines added by a DATA ACL are included in the reject log
7664 if the ACL subsequently rejects the message.
7666 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
7667 MD5 (which is deprecated).
7669 21. When testing a filter file using -bf, Exim was writing a message when it
7670 took the sender from a "From " line in the message, but it was not doing so
7671 when it took $return_path from a Return-Path: header line. It now does.
7673 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
7674 with a valid header line field name (a series of printing characters
7675 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
7677 23. Changed "disc" in the source to "disk" to conform to the documentation and
7678 the book and for uniformity.
7680 24. Ignore Sendmail's -Ooption=value command line item.
7682 25. When execve() failed while trying to run a command in a pipe transport,
7683 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
7684 could be confused with a return value of 69 from the command itself. This
7685 has been changed to 127, the value the shell returns if it is asked to run
7686 a non-existent command. The wording for the related log line suggests a
7687 non-existent command as the problem.
7689 26. If received_header_text expands to an empty string, do not add a Received:
7690 header line to the message. (Well, it adds a token one on the spool, but
7691 marks it "old" so that it doesn't get used or transmitted.)
7693 27. Installed eximstats 1.28 (addition of -nt option).
7695 28. There was no check for failure on the call to getsockname() in the daemon
7696 code. This can fail if there is a shortage of resources on the system, with
7697 ENOMEM, for example. A temporary error is now given on failure.
7699 29. Contrary to the C standard, it seems that in some environments, the
7700 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
7701 program. Exim now does this explicitly; it affects the formatting of
7702 timestamps using strftime().
7704 30. If exiqsumm was given junk data, it threw up some uninitialized variable
7705 complaints. I've now initialized all the variables, to avoid this.
7707 32. Header lines added by a system filter were not being "seen" during
7708 transport-time rewrites.
7710 33. The info_callback() function passed to OpenSSL is set up with type void
7711 (*)(SSL *, int, int), as described somewhere. However, when calling the
7712 function (actually a macro) that sets it up, the type void(*)() is
7713 expected. I've put in a cast to prevent warnings from picky compilers.
7715 34. If a DNS black list lookup found a CNAME record, but there were no A
7716 records associated with the domain it pointed at, Exim crashed.
7718 35. If a DNS black list lookup returned more than one A record, Exim ignored
7719 all but the first. It now scans all returned addresses if a particular IP
7720 value is being sought. In this situation, the contents of the
7721 $dnslist_value variable are a list of all the addresses, separated by a
7724 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
7725 to accept a host name and all its aliases if the forward lookup for any of
7726 them yielded the IP address of the incoming connection. Now it accepts only
7727 those names whose forward lookup yields the correct IP address. Any other
7728 names are discarded. This closes a loophole whereby a rogue DNS
7729 administrator could create reverse DNS records to break through a
7730 wildcarded host restriction in an ACL.
7732 37. If a user filter or a system filter that ran in a subprocess used any of
7733 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
7734 the wrong values were passed to the pipe command ($thisaddress had the
7735 value of $0, $0 had the value of $1, etc). This bug was introduced by
7736 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
7738 38. Improved the line breaking for long SMTP error messages from ACLs.
7739 Previously, if there was no break point between 40 and 75 characters, Exim
7740 left the rest of the message alone. Two changes have been made: (a) I've
7741 reduced the minimum length to 35 characters; (b) if it can't find a break
7742 point between 35 and 75 characters, it looks ahead and uses the first one
7743 that it finds. This may give the occasional overlong line, but at least the
7744 remaining text gets split now.
7746 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
7747 file descriptors might be low, and that setting 1000 would always raise it.
7748 It turns out that in some environments, the limit is already over 1000 and
7749 that lowering it causes trouble. So now Exim takes care not to decrease it.
7751 40. When delivering a message, the value of $return_path is set to $sender_
7752 address at the start of routing (routers may change the value). By an
7753 oversight, this default was not being set up when an address was tested by
7754 -bt or -bv, which affected the outcome if any router or filter referred to
7757 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
7758 log only when "message" or "log_message" are set. However, if one of the
7759 conditions was an address verification, or a call to a nested ACL, the
7760 messages generated by the underlying test were being passed through. This
7761 no longer happens. The underlying message is available in $acl_verify_
7762 message for both "message" and "log_message" expansions, so it can be
7763 passed through if needed.
7765 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
7766 new expansion $bh_ to give the encoded byte string without charset
7767 translation. Translation happens only if iconv() is available; HAVE_ICONV
7768 indicates this at build time. HEADERS_CHARSET gives the charset to
7769 translate to; headers_charset can change it in the configuration, and
7770 "headers charset" can change it in an individual filter file.
7772 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
7773 that creates RFC 2047 encoded "words" labels them as that charset instead
7774 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
7775 expansion operator; (ii) when Exim creates a From: line for a local
7776 message; (iii) when a header line is rewritten to include a "phrase" part.
7778 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
7779 buggy, causing it to skip the first lines of messages whose message ID
7780 ended in 'D'. This would not have bitten before Exim release 4.14, because
7781 message IDs were unlikely to end in 'D' before then. The effect was to have
7782 incorrect size information for certain domains.
7784 45. #include "config.h" was missing at the start of the crypt16.c module. This
7785 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
7788 46. If there was a timeout during a "random" callout check, Exim treated it as
7789 a failure of the random address, and carried on sending RSET and the real
7790 address. If the delay was just some slowness somewhere, the response to the
7791 original RCPT would be taken as a response to RSET and so on, causing
7792 mayhem of various kinds.
7794 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
7795 when I implemented it. It didn't allow for the fact that some option values
7796 may legitimately be negative (e.g. size_addition), and it didn't even do
7797 the right test for positive values.
7799 48. Domain names in DNS records are case-independent. Exim always looks them up
7800 in lower case. Some resolvers return domain names in exactly the case they
7801 appear in the zone file, that is, they may contain uppercase letters. Not
7802 all resolvers do this - some return always lower case. Exim was treating a
7803 change of case by a resolver as a change of domain, similar to a widening
7804 of a domain abbreviation. This triggered its re-routing code and so it was
7805 trying to route what was effectively the same domain again. This normally
7806 caused routing to fail (because the router wouldn't handle the domain
7807 twice). Now Exim checks for this case specially, and just changes the
7808 casing of the domain that it ultimately uses when it transmits the message
7811 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
7814 50. If a filter generated a file delivery with a non-absolute name (possible if
7815 no home directory exists for the router), the forbid_file option was not
7818 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
7819 the existing equality matching.
7821 52. Exim was using ints instead of ino_t variables in some places where it was
7822 dealing with inode numbers.
7824 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
7825 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
7826 called TMPDIR, and if it finds it is different, it changes its value.
7828 54. The smtp_printf() function is now made available to local_scan() so
7829 additional output lines can be written before returning. There is also an
7830 smtp_fflush() function to enable the detection of a dropped connection.
7831 The variables smtp_input and smtp_batched_input are exported to
7834 55. Changed the default runtime configuration: the message "Unknown user"
7835 has been removed from the ACL, and instead placed on the localuser router,
7836 using the cannot_route_message feature. This means that any verification
7837 failures that generate their own messages won't get overridden. Similarly,
7838 the "Unrouteable address" message that was in the ACL for unverifiable
7839 relay addresses has also been removed.
7841 56. Added hosts_avoid_esmtp to the smtp transport.
7843 57. The exicyclog script was not checking for the esoteric option
7844 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7845 will work only if exicyclog is run under the appropriate euid.
7847 58. Following a discussion on the list, the rules by which Exim recognises line
7848 endings on incoming messages have been changed. The -dropcr and drop_cr
7849 options are now no-ops, retained only for backwards compatibility. The
7850 following line terminators are recognized: LF CRLF CR. However, special
7851 processing applies to CR:
7853 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7854 nor a local message in the state where . is a terminator.
7856 (ii) If a bare CR is encountered in a header line, an extra space is added
7857 after the line terminator so as not to end the header. The reasoning
7858 behind this is that bare CRs in header lines are most likely either
7859 to be mistakes, or people trying to play silly games.
7861 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7862 was being incorrectly given as 18 bytes larger than it should have been.
7863 This is a VOB (very old bug).
7865 60. This may never have affected anything current, but just in case it has:
7866 When the local host is found other than at the start of a list of hosts,
7867 the local host, those with the same MX, and any that follow, are discarded.
7868 When the list in question was part of a longer list of hosts, the following
7869 hosts (not currently being processed) were also being discarded. This no
7870 longer happens. I'm not sure if this situation could ever has previously
7873 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7876 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7877 header. This is something that is recommended in a new Internet Draft, and
7878 is something that is documented as being done by Sendmail. There are two
7879 possible values. For messages generated by the autoreply transport, Exim
7882 Auto-Submitted: auto-replied
7884 whereas for all other generated messages (e.g. bounces) it adds
7886 Auto-Submitted: auto-generated
7888 63. The "personal" condition in filters now includes a test for the
7889 Auto-Submitted: header. If it contains the string "auto-" the message it
7890 not considered personal.
7892 64. Added rcpt_include_affixes as a generic transport option.
7894 65. Added queue_only_override (default true).
7896 66. Added the syslog_duplication option.
7898 67. If what should have been the first header line of a message consisted of
7899 a space followed by a colon, Exim was mis-interpreting it as a header line.
7900 It isn't of course - it is syntactically invalid and should therefore be
7901 treated as the start of the message body. The misbehaviour could have
7902 caused a number of strange effects, including loss of data in subsequent
7903 header lines, and spool format errors.
7905 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7906 client host had authenticated. This control can now be exercised by an ACL
7907 for more flexibility.
7909 69. By default, callouts do not happen when testing with -bh. There is now a
7910 variant, -bhc, which does actually run the callout code, including
7911 consulting and updating the callout cache.
7913 70. Added support for saslauthd authentication, courtesy of Alexander
7916 71. If statvfs() failed on the spool or log directories while checking their
7917 size for availability, Exim confusingly gave the error "space shortage".
7918 Furthermore, in debugging mode it crashed with a floating point exception.
7919 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7920 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7921 serious problem, Exim now writes to the main and panic logs when this
7922 happens, with details of the failure. It then refuses to accept the
7923 incoming message, giving the message "spool directory problem" or "log
7924 directory problem" with a 421 code for SMTP messages.
7926 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7927 0, 1, and 2 exist, because some OS complain for execs without them (see
7928 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7929 descriptors. However, the code omitted to check that the open succeeded,
7930 causing mysterious errors if for some reason the permissions on /dev/null
7931 got screwed. Now Exim writes a message to the main and panic logs, and
7932 bombs out if it can't open /dev/null.
7934 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7935 interact so that it is all more flexible. It is supposed to remain
7936 backwards compatible. Also added extra_local_interfaces.
7938 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7939 to bomb out with an assertion failure - to the client this appears as a
7940 connection drop. This problem occurs in the part of the code that was taken
7941 from the Samba project. Fortunately, the assertion is in a very simple
7942 function, so I have fixed this by reproducing the function inline in the
7943 one place where it is called, and arranging for authentication to fail
7944 instead of killing the process with assert().
7946 75. The SPA client code was not working when the server requested OEM rather
7947 than Unicode encoding.
7949 76. Added code to make require_files with a specific uid setting more usable in
7950 the case where statting the file as root fails - usually a non-root-mounted
7951 NFS file system. When this happens and the failure is EACCES, Exim now
7952 forks a subprocess and does the per-uid checking as the relevant uid.
7954 77. Added process_log_path.
7956 78. If log_file_path was not explicitly set, a setting of check_log_space or
7957 check_log_inodes was ignored.
7959 79. If a space check for the spool or log partitions fails, the incident is now
7960 logged. Of course, in the latter case the data may get lost...
7962 80. Added the %p formatting code to string_format() so that it can be used to
7963 print addresses in debug_print(). Adjusted all the address printing in the
7964 debugging in store.c to use %p rather than %d.
7966 81. There was a concern that a line of code in smtp_in.c could overflow a
7967 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7968 initially expressed, the concern was not well-founded, because trailing
7969 spaces are removed early. However, if the trailing spaces were followed by
7970 a NULL, they did not get removed, so the overflow was possible. Two fixes
7973 (a) I re-wrote the offending code in a cleaner fashion.
7974 (b) If an incoming SMTP command contains a NULL character, it is rejected
7977 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7978 runs initgroups(), so that if the Exim user is in any additional groups,
7979 they will be used during message reception.
7985 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob