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.
9 JH/01 Bug 1329: Fix format of Maildir-format filenames to match other mail-
10 related applications. Previously an "H" was used where available info
11 says that "M" should be, so change to match.
13 JH/02 Bug 2587: Fix pam expansion condition. Tainted values are commonly used
14 as arguments, so an implementation trying to copy these into a local
15 buffer was taking a taint-enforcement trap. Fix by using dynamically
16 created buffers. Similar fix for radius expansion condition.
18 JH/03 Bug 2586: Fix listcount expansion operator. Using tainted arguments is
19 reasonable, eg. to count headers. Fix by using dynamically created
20 buffers rather than a local. Do similar fixes for ACL actions "dcc",
21 "log_reject_target", "malware" and "spam"; the arguments are expanded
22 so could be handling tainted values.
24 JH/04 Bug 2590: Fix -bi (newaliases). A previous code rearrangement had
25 broken the (no-op) support for this sendmail command. Restore it
26 to doing nothing, silently, and returning good status.
28 JH/05 Bug 2593: Fix "vacation" in Exim filter. Previously, when a "once"
29 record path was given (or the default used) without a leading directory
30 path, an error occurred on trying to open it. Use the transport's working
33 JH/06 Bug 2594: Change the name used for certificate name checks in the smtp
34 transport. Previously it was the name on the DNS A-record; use instead
35 the head of the CNAME chain leading there (if there is one). This seems
36 to align better with RFC 6125.
38 JH/07 Bug 2597: Fix a resource leak. Using a lookup in obtaining a value for
39 smtp_accept_max_per_host allocated resources which were not released
40 when the limit was exceeded. This eventually crashed the daemon. Fix
41 by adding a relase action in that path.
43 JH/08 Bug 2598: Fix verify ACL condition. The options for the condition are
44 expanded; previously using tainted values was rejected. Fix by using
45 dynamically-created buffers.
47 JH/09 Relax restrictions on ACL verify condition needing access to message
48 headers. Previously they were only permitted in data and non-smtp ACLs;
49 permit also mime, dkim, prdr quit and notquit. Applies to header-syntax,
50 not_blind, header_sender and header_names_ascii verification.
52 JH/10 Bug 2603: Fix coding of string copying to only evaluate arguments once.
53 Previously a macro used one argument twice; when called with the
54 argument as an expression having side-effects, incorrect operation
55 resulted. Use an inlineable function.
57 JH/11 Bug 2604: Fix request to cutthrough-deliver when a connection is already
58 held open for a verify callout. Previously this wan not accounted for
59 and a corrupt onward SMTP conversation resulted.
61 JH/12 Bug 2607: Fix the ${srs_encode } expansion to handle quoted local_parts.
62 Previously they were embedded naively in the constructed address; when
63 needed, strip the quoting and quote the entire local_part.
64 Also make the inbound_srs expansion condition handle quoting.
66 JH/13 Fix dsearch "subdir" filter to ignore ".". Previously only ".." was
67 excluded, not matching the documentation.
69 JH/14 Bug 2606: Fix a segfault in sqlite lookups. When no, or a bad, filename
70 was given for the sqlite_dbfile a trap resulted.
72 JH/15 Bug 2620: Fix "spam" ACL condition. Previously, tainted values for the
73 "name" argument resulted in a trap. There is no reason to disallow such;
74 this was a coding error.
76 JH/16 Bug 2615: Fix pause during message reception, on systems that have been
77 suspended/resumed. The Linux CLOCK_MONOTONIC does not account for time
78 spent suspended, ignoring the Posix definition. Previously we assumed
79 it did and a constant offset from real time could be used as a correction.
80 Change to using the same clock source for the start-of-message and the
81 post-message next-tick-wait. Also change to using CLOCK_BOOTTIME if it
82 exists, just to get a clock slightly more aligned to reality.
84 JH/17 Bug 2295: Fix DKIM signing to always semicolon-terminate. Although the
85 RFC says it is optional some validators care. The missing char was not
86 intended but triggered by a line-wrap alignement. Discovery and fix by
87 Guillaume Outters, hacked on by JH.
89 JH/18 Bug 2617: Fix a taint trap in parse_fix_phrase(). Previously when the
90 name being quoted was tainted a trap would be taken. Fix by using
91 dynamicaly created buffers. The routine could have been called by a
92 rewrite with the "h" flag, by using the "-F" command-line option, or
93 by using a "name=" option on a control=submission ACL modifier.
95 JH/19 SPF: change the Authentication-Results expansion component to give
96 smtp.helo when the sender domain is empty. Previously it gave
99 JH/20 Bug 2631: ACL dnslist conditions now ignore and log any lookups returns
100 not in 127.0.0.0/8 to help in spotting list domains taken over by a
101 domain-parking registrar.
103 JH/21 Bug 2630: Fix eol-replacement string for the ${readsocket } expansion.
104 Previously when a whitespace character was specified it was not inserted
105 after removing the newline.
107 JH/22 Bug 2265: Force SNI usage for smtp transport DANE'd connections, to be
108 the domain part of the recipient address. This overrides any tls_sni
109 option set, which was previously used.
111 JH/23 Logging: with the +tls_sni log_selector, do not wrap the received SNI
114 JH/24 Bug 2634: Fix a taint trap seen on NetBSD: the testing coded for
115 is_tainted() had an off-by-one error in the overenthusiastic direction.
116 Find and fix by Gavan. Although NetBSD is not a supported platform for
117 4.94 this bug could affect other platforms.
119 PP/01 Fix default prime selection to be consistent.
120 One path used ike23 still, instead of exim.dev.20160529.3; now both
121 execution flows will use the same DH primes (currently
122 exim.dev.20160529.3).
124 JH/25 OpenSSL: Fix back-compatibility behaviour surrounding tls_certificates
125 option in smtp transport, to match the documentation. Previously
126 verification was not being done in some cases where it should have been.
128 JH/26 Bug 2646: fix a memory usage issue in ldap lookups. Previously, when more
129 than one server was defined and depending on the platform memory layout
130 details, an internal consistency trap could be hit while walking the list
133 JH/27 Bug 2648: fix the passing of an authenticator public-name through spool
134 files. The value is used by the authresults expansion item. Previously
135 if this was used in a router or transport, a crash could result.
137 JH/28 Fix spurious logging of select error. Some platforms, notably FreeBSD,
138 have a sufficient incidence of EINTR returns from select that an
139 interaction with other operations done by the main daemon loop exposed
140 a bug in the error-handling. This was benign apart from the log
143 JH/29 Bug 2675: add outgoing-interface I= element to deferred "==" log lines,
144 for consistency with delivered "=>" and failed "**" lines. While we're
145 there, handle PRX and TFO.
147 JH/30 Bug 2677: fix matching of long addresses. Since 4.93 a limit of 256 was
148 applied. This resulted, if any header-line rewrite rules were configured,
149 in a panic-log trigerrable by sending a message with a long address in
150 a header. Fix by increaing the arbitrary limit to larger than a single
151 (dewrapped) 5322 header line maximum size.
153 JH/31 The ESMTP option name advertised for the SUPPORT_EARLY_PIPE build option
154 is changed from X_PIPE_CONNECT to PIPE_CONNECT. This is in line with
155 RFC 6648 which deprecates X- options in protocols as a general practice.
156 Changeover between the implementations is handled by the mechanisms
159 JH/32 Bug 2599: fix delay of delivery to a local address where there is also
160 a remote which uses callout/hold. Previously the local was queued.
162 JH/33 Fix a taint trap in the ${listextract } expansion when the source data
165 JH/34 Fix the placement of a multiple-message delivery marker in the delivery
166 log line. The asterisk is now consistently appended to the remote IP
167 (and port, if given), and will also be provided on defer and fail log
168 lines. Previously it could be placed on the local IP if that was being
169 logged, and was only provided on delivery lines.
171 JH/35 Bug 2343: Harden exim_tidydb against corrupt wait- files.
173 JH/36 Bug 2687: Fix interpretation of multiple ^ chars in a plaintext
174 authenticator client_send option. Previously the next char, after a pair
175 was collapsed, was taken verbatim (so ^^^foo became ^^foo; ^^^^foo became
176 ^^\x00foo). Fixed to get ^\x00foo and ^^foo respectively to match the
177 documentation. There is still no way to get a leading ^ immediately
178 after a NUL (ie. for the password of a PLAIN method authenticator.
180 JH/37 Enforce the expected size, for fixed-size records read from hints-DB
181 files. For bad sizes read, delete the record and whine to paniclog.
183 JH/38 When logging an AUTH failure, as server, do not include sensitive
184 information. Previously, the credentials would be included if given
185 as part of the AUTH command line and an ACL denied authentidcation.
187 JH/39 Bug 2691: fix $local_part_data. When the matching list element
188 referred to a file, bad data was returned. This likely also affected
191 JH/40 The gsasl authenticator now supports caching of the salted password
192 generated by the client-side implementation. This required the addition
193 of a new variable: $auth4.
195 JH/41 Fix daemon SIGHUP on FreeBSD. Previously, a named socket for IPC was
196 left undeleted; the attempt to re-create it then failed - resulting in
197 the usual "SIGHUP tp have daemon reload configuration" to not work.
198 This affected any platform not supporting "abstract" Unix-domain
199 sockets (i.e. not Linux).
201 JH/42 Bug 2693: Harden against a peer which reneges on a 452 "too many
202 recipients" response to RCPT in a later response, with a 250. The
203 previous coding assumed this would not happen, and under PIPELINING
204 would result in both lost and duplicate recipients for a message.
206 JH/43 Bug 2694: Fix weighted distribution of work to multiple spamd servers.
207 Previously the weighting was incorrectly applied. Found and fixed by
214 JH/01 Avoid costly startup code when not strictly needed. This reduces time
215 for some exim process initialisations. It does mean that the logging
216 of TLS configuration problems is only done for the daemon startup.
218 JH/02 Early-pipelining support code is now included unless disabled in Makefile.
220 JH/03 DKIM verification defaults no long accept sha1 hashes, to conform to
221 RFC 8301. They can still be enabled, using the dkim_verify_hashes main
224 JH/04 Support CHUNKING from an smtp transport using a transport_filter, when
225 DKIM signing is being done. Previously a transport_filter would always
226 disable CHUNKING, falling back to traditional DATA.
228 JH/05 Regard command-line recipients as tainted.
230 JH/06 Bug 340: Remove the daemon pid file on exit, when due to SIGTERM.
232 JH/07 Bug 2489: Fix crash in the "pam" expansion condition. It seems that the
233 PAM library frees one of the arguments given to it, despite the
234 documentation. Therefore a plain malloc must be used.
236 JH/08 Bug 2491: Use tainted buffers for the transport smtp context. Previously
237 on-stack buffers were used, resulting in a taint trap when DSN information
238 copied from a received message was written into the buffer.
240 JH/09 Bug 2493: Harden ARC verify against Outlook, whick has been seen to mix
241 the ordering of its ARC headers. This caused a crash.
243 JH/10 Bug 2492: Use tainted memory for retry record when needed. Previously when
244 a new record was being constructed with information from the peer, a trap
247 JH/11 Bug 2494: Unset the default for dmarc_tld_file. Previously a naiive
248 installation would get error messages from DMARC verify, when it hit the
249 nonexistent file indicated by the default. Distros wanting DMARC enabled
250 should both provide the file and set the option.
251 Also enforce no DMARC verification for command-line sourced messages.
253 JH/12 Fix an uninitialised flag in early-pipelining. Previously connections
254 could, depending on the platform, hang at the STARTTLS response.
256 JH/13 Bug 2498: Reset a counter used for ARC verify before handling another
257 message on a connection. Previously if one message had ARC headers and
258 the following one did not, a crash could result when adding an
259 Authentication-Results: header.
261 JH/14 Bug 2500: Rewind some of the common-coding in string handling between the
262 Exim main code and Exim-related utities. The introduction of taint
263 tracking also did many adjustments to string handling. Since then, eximon
264 frequently terminated with an assert failure.
266 JH/15 When PIPELINING, synch after every hundred or so RCPT commands sent and
267 check for 452 responses. This slightly helps the inefficieny of doing
268 a large alias-expansion into a recipient-limited target. The max_rcpt
269 transport option still applies (and at the current default, will override
270 the new feature). The check is done for either cause of synch, and forces
271 a fast-retry of all 452'd recipients using a new MAIL FROM on the same
272 connection. The new facility is not tunable at this time.
274 JH/16 Fix the variables set by the gsasl authenticator. Previously a pointer to
275 library live data was being used, so the results became garbage. Make
276 copies while it is still usable.
278 JH/17 Logging: when the deliver_time selector ise set, include the DT= field
279 on delivery deferred (==) and failed (**) lines (if a delivery was
280 attemtped). Previously it was only on completion (=>) lines.
282 JH/18 Authentication: the gsasl driver not provides the $authN variables in time
283 for the expansion of the server_scram_iter and server_scram_salt options.
285 WB/01 SPF: DNS lookups for the obsolete SPF RR type done by the libspf2 library
286 are now specifically given a NO_DATA response without hitting the system
287 resolver. The library goes on to do the now-standard TXT lookup.
288 Use of dnsdb lookups is not affected.
290 JH/19 Bug 2507: Modules: on handling a dynamic-module (lookups) open failure,
291 only retrieve the errormessage once. Previously two calls to dlerror()
292 were used, and the second one (for mainlog/paniclog) retrieved null
295 JH/20 Taint checking: disallow use of tainted data for
296 - the appendfile transport file and directory options
297 - the pipe transport command
298 - the autoreply transport file, log and once options
299 - file names used by the redirect router (including filter files)
301 - paths used by single-key lookups
302 Previously this was permitted.
304 JH/21 Bug 2501: Fix init call in the heimdal authenticator. Previously it
305 adjusted the size of a major service buffer; this failed because the
306 buffer was in use at the time. Change to a compile-time increase in the
307 buffer size, when this authenticator is compiled into exim.
309 JH/22 Taint-checking: move to safe-mode taint checking on all platforms. The
310 previous fast-mode was untenable in the face of glibs using mmap to
311 support larger malloc requests.
313 PP/01 Update the openssl_options possible values through OpenSSL 1.1.1c.
314 New values supported, if defined on system where compiled:
315 allow_no_dhe_kex, cryptopro_tlsext_bug, enable_middlebox_compat,
316 no_anti_replay, no_encrypt_then_mac, prioritize_chacha, tlsext_padding
318 JH/23 Performance improvement in the initial phase of a two-pass queue run. By
319 running a limited number of proceses in parallel, a benefit is gained. The
320 amount varies with the platform hardware and load. The use of the option
321 queue_run_in_order means we cannot do this, as ordering becomes
324 JH/24 Bug 2524: fix the cyrus_sasl auth driver gssapi usage. A previous fix
325 had introduced a string-copy (for ensuring NUL-termination) which was not
326 appropriate for that case, which can include embedded NUL bytes in the
327 block of data. Investigation showed the copy to actually be needless, the
328 data being length-specified.
330 JH/25 Fix use of concurrent TLS connections under GnuTLS. When a callout was
331 done during a receiving connection, and both used TLS, global info was
332 used rather than per-connection info for tracking the state of data
333 queued for transmission. This could result in a connection hang.
335 JH/26 Fix use of the SIZE parameter on MAIL commands, on continued connections.
336 Previously, when delivering serveral messages down a single connection
337 only the first would provide a SIZE. This was due to the size information
338 not being properly tracked.
340 JH/27 Bug 2530: When operating in a timezone with sub-minute offset, such as
341 TAI (at 37 seconds currently), pretend to be in UTC for time-related
342 expansion and logging. Previously, spurious values such as a future
343 minute could be seen.
345 JH/28 Bug 2533: Fix expansion of ${tr } item. When called in some situations
346 it could crash from a null-deref. This could also affect the
347 ${addresses: } operator and ${readsock } item.
349 JH/29 Bug 2537: Fix $mime_part_count. When a single connection had a non-mime
350 message following a mime one, the variable was not reset.
352 JH/30 When an pipelined-connect fails at the first response, assume incorrect
353 cached capability (perhaps the peer reneged?) and immediately retry in
356 JH/31 Fix spurious detection of timeout while writing to transport filter.
358 JH/32 Bug 2541: Fix segfault on bad cmdline -f (sender) argument. Previously
359 an attempt to copy the string was made before checking it.
361 JH/33 Fix the dsearch lookup to return an untainted result. Previously the
362 taint of the lookup key was maintained; we now regard the presence in the
363 filesystem as sufficient validation.
365 JH/34 Fix the readsocket expansion to not segfault when an empty "options"
366 argument is supplied.
368 JH/35 The dsearch lookup now requires that the directory is an absolute path.
369 Previously this was not checked, and nonempty relative paths made an
370 access under Exim's current working directory.
372 JH/36 Bug 2554: Fix msg:defer event for the hosts_max_try_hardlimit case.
373 Previously no event was raised.
375 JH/37 Bug 2552: Fix the check on spool space during reception to use the SIZE
376 parameter supplied by the sender MAIL FROM command. Previously it was
377 ignored, and only the check_spool_space option value for the required
380 JH/38 Fix $dkim_key_length. This should, after a DKIM verification, present
381 the size of the signing public-key. Previously it was instead giving
382 the size of the signature hash.
384 JH/39 DKIM verification: the RFC 8301 restriction on sizes of RSA keys is now
385 the default. See the (new) dkim_verify_min_keysizes option.
387 JH/40 Fix a memory-handling bug: when a connection carried multiple messages
388 and an ACL use a lookup for checking either the local_part or domain,
389 stale data could be accessed. Ensure that variable references are
390 dropped between messages.
392 JH/41 Bug 2571: Fix SPA authenticator. Running as a server, an offset supplied
393 by the client was not checked as pointing within response data before
394 being used. A malicious client could thus cause an out-of-bounds read and
395 possibly gain authentication. Fix by adding the check.
397 JH/42 Internationalisation: change the default for downconversion in the smtp
398 transport to be "if needed". Previously it was "as previously set" for
399 the message, which usually meant "if needed" for message-submission but
400 "no" for everything else. However, MTAs have been seen using SMTPUTF8
401 even when the envelope addresses did not need it, resulting in forwarding
402 failures to non-supporting MTAs. A downconvert in such cases will be
403 a no-op on the addresses, merely dropping the use of SMTPUTF8 by the
404 transport. The change does mean that addresses needing conversion will
405 be converted when previously a delivery failure would occur.
407 JH/43 Fix possible long line in DSN. Previously when a very long SMTP error
408 response was received it would be used unchecked in a fail-DSN, violating
409 standards on line-length limits. Truncate if needed.
411 HS/01 Remove parameters of the link to www.open-spf.org. The linked form
412 doesn't work. (Additionally add a new main config option to configure the
419 JH/01 OpenSSL: With debug enabled output keying information sufficient, server
420 side, to decode a TLS 1.3 packet capture.
422 JH/02 OpenSSL: Suppress the sending of (stateful) TLS1.3 session tickets.
423 Previously the default library behaviour applied, sending two, each in
426 JH/03 Debug output for ACL now gives the config file name and line number for
429 JH/04 The default received_header_text now uses the RFC 8314 tls cipher clause.
431 JH/05 DKIM: ensure that dkim_domain elements are lowercased before use.
433 JH/06 Fix buggy handling of autoreply bounce_return_size_limit, and a possible
434 buffer overrun for (non-chunking) other transports.
436 JH/07 GnuTLS: Our use of late (post-handshake) certificate verification, under
437 TLS1.3, means that a server rejecting a client certificate is not visible
438 to the client until the first read of encrypted data (typically the
439 response to EHLO). Add detection for that case and treat it as a failed
440 TLS connection attempt, so that the normal retry-in-clear can work (if
441 suitably configured).
443 JB/01 Bug 2375: fix expansions of 822 addresses having comments in local-part
444 and/or domain. Found and fixed by Jason Betts.
446 JH/08 Add hardening against SRV & TLSA lookups the hit CNAMEs (a nonvalid
447 configuration). If a CNAME target was not a wellformed name pattern, a
450 JH/09 Logging: Fix initial listening-on line for multiple ports for an IP when
451 the OS reports them interleaved with other addresses.
453 JH/10 OpenSSL: Fix aggregation of messages. Previously, when PIPELINING was
454 used both for input and for a verify callout, both encrypted, SMTP
455 responses being sent by the server could be lost. This resulted in
456 dropped connections and sometimes bounces generated by a peer sending
459 JH/11 Harden plaintext authenticator against a badly misconfigured client-send
460 string. Previously it was possible to cause undefined behaviour in a
461 library routine (usually a crash). Found by "zerons".
463 JH/12 Bug 2384: fix "-bP smtp_receive_timeout". Previously it returned no
466 JH/13 Bug 2386: Fix builds with Dane under LibreSSL 2.9.0 onward. Some old
467 API was removed, so update to use the newer ones.
469 JH/14 Bug 1891: Close the log file if receiving a non-smtp message, without
470 any timeout set, is taking a long time. Previously we would hang on to a
471 rotated logfile "forever" if the input was arriving with long gaps
472 (a previous attempt to fix addressed lack, for a long time, of initial
475 HS/01 Bug 2390: Use message_id for tempfile creation to avoid races in a
476 shared (NFS) environment. The length of the tempfile name is now
477 4 + 16 ("hdr.$message_exim_id") which might break on file
478 systems which restrict the file name length to lower values.
481 HS/02 Bug 2390: Use message_id for tempfile creation to avoid races in a
482 shared (NFS) environment.
484 HS/03 Bug 2392: exigrep does case sensitive *option* processing (as it
485 did for all versions <4.90). Notably -M, -m, --invert, -I may be
488 JH/15 Use unsigned when creating bitmasks in macros, to avoid build errors
489 on some platforms for bit 31.
491 JH/16 GnuTLS: rework ciphersuite strings under recent library versions. Thanks
492 to changes apparently associated with TLS1.3 handling some of the APIs
493 previously used were either nonfunctional or inappropriate. Strings
494 like TLS1.3:ECDHE_SECP256R1__RSA_PSS_RSAE_SHA256__AES_256_GCM__AEAD:256
495 and TLS1.2:ECDHE_SECP256R1__RSA_SHA256__AES_128_CBC__SHA256:128 replace
496 the previous TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256 .
497 This affects log line X= elements, the $tls_{in,out}_cipher variables,
498 and the use of specific cipher names in the encrypted= ACL condition.
500 JH/17 OpenSSL: the default openssl_options now disables ssl_v3.
502 JH/18 GnuTLS: fix $tls_out_ocsp under hosts_request_ocsp. Previously the
503 verification result was not updated unless hosts_require_ocsp applied.
505 JH/19 Bug 2398: fix listing of a named-queue. Previously, even with the option
506 queue_list_requires_admin set to false, non-admin users were denied the
509 JH/20 Bug 2389: fix server advertising of usable certificates, under GnuTLS in
510 directory-of-certs mode. Previously they were advertised despite the
513 JH/21 The smtp transport option "hosts_noproxy_tls" is now unset by default.
514 A single TCP connection by a client will now hold a TLS connection open
515 for multiple message deliveries, by default. Previously the default was to
518 JH/22 The smtp transport option "hosts_try_dane" now enables all hosts by
519 default. If built with the facility, DANE will be used. The facility
520 SUPPORT_DANE is now enabled in the prototype build Makefile "EDITME".
522 JH/23 The build default is now for TLS to be included; the SUPPORT_TLS define
523 is replaced with DISABLE_TLS. Either USE_GNUTLS or (the new) USE_OPENSSL
524 must be defined and you must still, unless you define DISABLE_TLS, manage
525 the the include-dir and library-file requirements that go with that
526 choice. Non-TLS builds are still supported.
528 JH/24 Fix duplicated logging of peer name/address, on a transport connection-
531 JH/25 The smtp transport option "hosts_try_fastopen" now enables all hosts by
532 default. If the platform supports and has the facility enabled, it will
533 be requested on all coneections.
535 JH/26 The PIPE_CONNECT facility is promoted from experimental status and is now
536 controlled by the build-time option SUPPORT_PIPE_CONNECT.
538 PP/01 Unbreak heimdal_gssapi, broken in 4.92.
540 JH/27 Bug 2404: Use the main-section configuration option "dsn_from" for
541 success-DSN messages. Previously the From: header was always the default
542 one for these; the option was ignored.
544 JH/28 Fix the timeout on smtp response to apply to the whole response.
545 Previously it was reset for every read, so a teergrubing peer sending
546 single bytes within the time limit could extend the connection for a
547 long time. Credit to Qualsys Security Advisory Team for the discovery.
549 JH/29 Fix DSN Final-Recipient: field. Previously it was the post-routing
550 delivery address, which leaked information of the results of local
551 forwarding. Change to the original envelope recipient address, per
554 JH/30 Bug 2411: Fix DSN generation when RFC 3461 failure notification is
555 requested. Previously not bounce was generated and a log entry of
556 error ignored was made.
558 JH/31 Avoid re-expansion in ${sort } expansion. (CVE-2019-13917)
560 JH/32 Introduce a general tainting mechanism for values read from the input
561 channel, and values derived from them. Refuse to expand any tainted
562 values, to catch one form of exploit.
564 JH/33 Bug 2413: Fix dkim_strict option. Previously the expansion result
565 was unused and the unexpanded text used for the test. Found and
566 fixed by Ruben Jenster.
568 JH/34 Fix crash after TLS shutdown. When the TCP/SMTP channel was left open,
569 an attempt to use a TLS library read routine dereffed a nul pointer,
572 JH/35 Bug 2409: filter out-of-spec chars from callout response before using
573 them in our smtp response.
575 JH/36 Have the general router option retry_use_local_part default to true when
576 any of the restrictive preconditions are set (to anything). Previously it
577 was only for check_local user. The change removes one item of manual
578 configuration which is required for proper retries when a remote router
579 handles a subset of addresses for a domain.
581 JH/37 Appendfile: when evaluating quota use (non-quota_size_regex) take the file
582 link count into consideration.
584 HS/04 Fix handling of very log lines in -H files. If a -<key> <value> line
585 caused the extension of big_buffer, the following lines were ignored.
587 JH/38 Bug 1395: Teach the DNS negative-cache about TTL value from the SOA in
588 accordance with RFC 2308. Previously there was no expiry, so a longlived
589 receive process (eg. due to ACL delays) versus a short SOA value could
592 HS/05 Handle trailing backslash gracefully. (CVE-2019-15846)
594 JH/39 Promote DMARC support to mainline.
596 JH/40 Bug 2452: Add a References: header to DSNs.
598 JH/41 With GnuTLS 3.6.0 (and later) do not attempt to manage Diffie-Hellman
599 parameters. The relevant library call is documented as "Deprecated: This
600 function is unnecessary and discouraged on GnuTLS 3.6.0 or later. Since
601 3.6.0, DH parameters are negotiated following RFC7919."
603 HS/06 Change the default of dnssec_request_domains to "*"
605 JH/42 Bug 2545: Fix CHUNKING for all RCPT commands rejected. Previously we
606 carried on and emitted a BDAT command, even when PIPELINING was not
609 JH/43 Bug 2465: Fix taint-handling in dsearch lookup. Previously a nontainted
610 buffer was used for the filename, resulting in a trap when tainted
611 arguments (eg. $domain) were used.
613 JH/44 With OpenSSL 1.1.1 (onwards) disable renegotiation for TLS1.2 and below;
614 recommended to avoid a possible server-load attack. The feature can be
615 re-enabled via the openssl_options main cofiguration option.
617 JH/45 local_scan API: documented the current smtp_printf() call. This changed
618 for version 4.90 - adding a "more data" boolean to the arguments.
619 Bumped the ABI version number also, this having been missed previously;
620 release versions 4.90 to 4.92.3 inclusive were effectively broken in
621 respect of usage of smtp_printf() by either local_scan code or libraries
622 accessed via the ${dlfunc } expansion item. Both will need coding
623 adjustment for any calls to smtp_printf() to match the new function
624 signature; a FALSE value for the new argument is always safe.
626 JH/46 FreeBSD: fix use of the sendfile() syscall. The shim was not updating
627 the file-offset (which the Linux syscall does, and exim expects); this
628 resulted in an indefinite loop.
630 JH/47 ARC: fix crash in signing, triggered when a configuration error failed
631 to do ARC verification. The Authentication-Results: header line added
632 by the configuration then had no ARC item.
638 JH/01 Remove code calling the customisable local_scan function, unless a new
639 definition "HAVE_LOCAL_SCAN=yes" is present in the Local/Makefile.
641 JH/02 Bug 1007: Avoid doing logging from signal-handlers, as that can result in
642 non-signal-safe functions being used.
644 JH/03 Bug 2269: When presented with a received message having a stupidly large
645 number of DKIM-Signature headers, disable DKIM verification to avoid
646 a resource-consumption attack. The limit is set at twenty.
648 JH/04 Add variables $arc_domains, $arc_oldest_pass for ARC verify. Fix the
649 report of oldest_pass in ${authres } in consequence, and separate out
650 some descriptions of reasons for verification fail.
652 JH/05 Bug 2273: Cutthrough delivery left a window where the received messsage
653 files in the spool were present and unlocked. A queue-runner could spot
654 them, resulting in a duplicate delivery. Fix that by doing the unlock
655 after the unlink. Investigation by Tim Stewart. Take the opportunity to
656 add more error-checking on spoolfile handling while that code is being
659 PP/01 Refuse to open a spool data file (*-D) if it's a symlink.
660 No known attacks, no CVE, this is defensive hardening.
662 JH/06 Bug 2275: The MIME ACL unlocked the received message files early, and
663 a queue-runner could start a delivery while other operations were ongoing.
664 Cutthrough delivery was a common victim, resulting in duplicate delivery.
665 Found and investigated by Tim Stewart. Fix by using the open message data
666 file handle rather than opening another, and not locally closing it (which
667 releases a lock) for that case, while creating the temporary .eml format
668 file for the MIME ACL. Also applies to "regex" and "spam" ACL conditions.
670 JH/07 Bug 177: Make a random-recipient callout success visible in ACL, by setting
671 $sender_verify_failure/$recipient_verify_failure to "random".
673 JH/08 When generating a selfsigned cert, use serial number 1 since zero is not
676 JH/09 Bug 2274: Fix logging of cmdline args when starting in an unlinked cwd.
677 Previously this would segfault.
679 JH/10 Fix ARC signing for case when DKIM signing failed. Previously this would
682 JH/11 Bug 2264: Exim now only follows CNAME chains one step by default. We'd
683 like zero, since the resolver should be doing this for us, But we need one
684 as a CNAME but no MX presence gets the CNAME returned; we need to check
685 that doesn't point to an MX to declare it "no MX returned" rather than
686 "error, loop". A new main option is added so the older capability of
687 following some limited number of chain links is maintained.
689 JH/12 Add client-ip info to non-pass iprev ${authres } lines.
691 JH/13 For receent Openssl versions (1.1 onward) use modern generic protocol
692 methods. These should support TLS 1.3; they arrived with TLS 1.3 and the
693 now-deprecated earlier definitions used only specified the range up to TLS
694 1.2 (in the older-version library docs).
696 JH/14 Bug 2284: Fix DKIM signing for body lines starting with a pair of dots.
698 JH/15 Rework TLS client-side context management. Stop using a global, and
699 explicitly pass a context around. This enables future use of TLS for
700 connections to service-daemons (eg. malware scanning) while a client smtp
701 connection is using TLS; with cutthrough connections this is quite likely.
703 JH/16 Fix ARC verification to do AS checks in reverse order.
705 JH/17 Support a "tls" option on the ${readsocket } expansion item.
707 JH/18 Bug 2287: Fix the protocol name (eg utf8esmtp) for multiple messages
708 using the SMTPUTF8 option on their MAIL FROM commands, in one connection.
709 Previously the "utf8" would be re-prepended for every additional message.
711 JH/19 Reject MAIL FROM commands with SMTPUTF8 when the facility was not advertised.
712 Previously thery were accepted, resulting in issues when attempting to
713 forward messages to a non-supporting MTA.
715 PP/02 Let -n work with printing macros too, not just options.
717 JH/20 Bug 2296: Fix cutthrough for >1 address redirection. Previously only
718 one parent address was copied, and bogus data was used at delivery-logging
719 time. Either a crash (after delivery) or bogus log data could result.
720 Discovery and analysis by Tim Stewart.
722 PP/03 Make ${utf8clean:} expansion operator detect incomplete final character.
723 Previously if the string ended mid-character, we did not insert the
724 promised '?' replacement.
726 PP/04 Documentation: current string operators work on bytes, not codepoints.
728 JH/21 Change as many as possible of the global flags into one-bit bitfields; these
729 should pack well giving a smaller memory footprint so better caching and
730 therefore performance. Group the declarations where this can't be done so
731 that the byte-sized flag variables are not interspersed among pointer
732 variables, giving a better chance of good packing by the compiler.
734 JH/22 Bug 1896: Fix the envelope from for DMARC forensic reports to be possibly
735 non-null, to avoid issues with sites running BATV. Previously reports were
736 sent with an empty envelope sender so looked like bounces.
738 JH/23 Bug 2318: Fix the noerror command within filters. It wasn't working.
739 The ignore_error flag wasn't being returned from the filter subprocess so
740 was not set for later routers. Investigation and fix by Matthias Kurz.
742 JH/24 Bug 2310: Raise a msg:fail:internal event for each undelivered recipient,
743 and a msg:complete for the whole, when a message is manually removed using
744 -Mrm. Developement by Matthias Kurz, hacked on by JH.
746 JH/25 Avoid fixed-size buffers for pathnames in DB access. This required using
747 a "Gnu special" function, asprintf() in the DB utility binary builds; I
748 hope that is portable enough.
750 JH/26 Bug 2311: Fix DANE-TA verification under GnuTLS. Previously it was also
751 requiring a known-CA anchor certificate; make it now rely entirely on the
752 TLSA as an anchor. Checking the name on the leaf cert against the name
753 on the A-record for the host is still done for TA (but not for EE mode).
755 JH/27 Fix logging of proxy address. Previously, a pointless "PRX=[]:0" would be
756 included in delivery lines for non-proxied connections, when compiled with
757 SUPPORT_SOCKS and running with proxy logging enabled.
759 JH/28 Bug 2314: Fire msg:fail:delivery event even when error is being ignored.
760 Developement by Matthias Kurz, tweaked by JH. While in that bit of code,
761 move the existing event to fire before the normal logging of message
762 failure so that custom logging is bracketed by normal logging.
764 JH/29 Bug 2322: A "fail" command in a non-system filter (file) now fires the
765 msg:fail:internal event. Developement by Matthias Kurz.
767 JH/30 Bug 2329: Increase buffer size used for dns lookup from 2k, which was
768 far too small for todays use of crypto signatures stored there. Go all
769 the way to the max DNS message size of 64kB, even though this might be
770 overmuch for IOT constrained device use.
772 JH/31 Fix a bad use of a copy function, which could be used to pointlessly
773 copy a string over itself. The library routine is documented as not
774 supporting overlapping copies, and on MacOS it actually raised a SIGABRT.
776 JH/32 For main options check_spool_space and check_inode_space, where the
777 platform supports 64b integers, support more than the previous 2^31 kB
778 (i.e. more than 2 TB). Accept E, P and T multipliers in addition to
779 the previous G, M, k.
781 JH/33 Bug 2338: Fix the cyrus-sasl authenticator to fill in the
782 $authenticated_fail_id variable on authentication failure. Previously
785 JH/34 Increase RSA keysize of autogen selfsign cert from 1024 to 2048. RHEL 8.0
786 OpenSSL didn't want to use such a weak key. Do for GnuTLS also, and for
787 more-modern GnuTLS move from GNUTLS_SEC_PARAM_LOW to
788 GNUTLS_SEC_PARAM_MEDIUM.
790 JH/35 OpenSSL: fail the handshake when SNI processing hits a problem, server
791 side. Previously we would continue as if no SNI had been received.
793 JH/36 Harden the handling of string-lists. When a list consisted of a sole
794 "<" character, which should be a list-separator specification, we walked
795 off past the nul-terimation.
797 JH/37 Bug 2341: Send "message delayed" warning MDNs (restricted to external
798 causes) even when the retry time is not yet met. Previously they were
799 not, meaning that when (say) an account was over-quota and temp-rejecting,
800 and multiple senders' messages were queued, only one sender would get
801 notified on each configured delay_warning cycle.
803 JH/38 Bug 2351: Log failures to extract envelope addresses from message headers.
805 JH/39 OpenSSL: clear the error stack after an SSL_accept(). With anon-auth
806 cipher-suites, an error can be left on the stack even for a succeeding
807 accept; this results in impossible error messages when a later operation
810 AM/01 Bug 2359: GnuTLS: repeat lowlevel read and write operations while they
811 return error codes indicating retry. Under TLS1.3 this becomes required.
813 JH/40 Fix the feature-cache refresh for EXPERIMENTAL_PIPE_CONNECT. Previously
814 it only wrote the new authenticators, resulting in a lack of tracking of
815 peer changes of ESMTP extensions until the next cache flush.
817 JH/41 Fix the loop reading a message header line to check for integer overflow,
818 and more-often against header_maxsize. Previously a crafted message could
819 induce a crash of the recive process; now the message is cleanly rejected.
821 JH/42 Bug 2366: Fix the behaviour of the dkim_verify_signers option. It had
822 been totally disabled for all of 4.91. Discovery and fix by "Mad Alex".
828 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
829 When redis_servers is set to a list of > 1 element, and the Redis servers
830 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
831 case of MOVED into a DEFER case instead, thus moving the query onto the
832 next server in the list. For a cluster of N elements, all N servers must
833 be defined in redis_servers.
835 GF/02 Catch and remove uninitialized value warning in exiqsumm
836 Check for existence of @ARGV before looking at $ARGV[0]
838 JH/01 Replace the store_release() internal interface with store_newblock(),
839 which internalises the check required to safely use the old one, plus
840 the allocate and data copy operations duplicated in both (!) of the
841 extant use locations.
843 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
844 modifier. This matches the restriction on the commandline.
846 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
847 Previously only the last row was returned.
849 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
850 we assumed that tags in the header were well-formed, and parsed the
851 element content after inspecting only the first char of the tag.
852 Assumptions at that stage could crash the receive process on malformed
855 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
856 While running the DKIM ACL we operate on the Permanent memory pool so that
857 variables created with "set" persist to the DATA ACL. Also (at any time)
858 DNS lookups that fail create cache records using the Permanent pool. But
859 expansions release any allocations made on the current pool - so a dnsdb
860 lookup expansion done in the DKIM ACL releases the memory used for the
861 DNS negative-cache, and bad things result. Solution is to switch to the
862 Main pool for expansions.
863 While we're in that code, add checks on the DNS cache during store_reset,
864 active in the testsuite.
865 Problem spotted, and debugging aided, by Wolfgang Breyha.
867 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
868 When none of the hosts presented to a transport match an already-open
869 connection, close it and proceed with the list. Previously we would
870 queue the message. Spotted by Lena with Yahoo, probably involving
873 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
874 Previously a spurious "250 OK id=" response was appended to the proper
877 JH/08 The "support for" informational output now, which built with Content
878 Scanning support, has a line for the malware scanner interfaces compiled
879 in. Interface can be individually included or not at build time.
881 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
882 by the template makefile "src/EDITME". The "STREAM" support for an older
883 ClamAV interface method is removed.
885 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
886 rows affected is given instead).
888 JH/11 The runtime Berkeley DB library version is now additionally output by
889 "exim -d -bV". Previously only the compile-time version was shown.
891 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
892 SMTP connection. Previously, when one had more recipients than the
893 first, an abortive onward connection was made. Move to full support for
894 multiple onward connections in sequence, handling cutthrough connection
895 for all multi-message initiating connections.
897 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
898 routers. Previously, a multi-recipient message would fail to match the
899 onward-connection opened for the first recipient, and cause its closure.
901 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
902 a timeout on read on a GnuTLS initiating connection, resulting in the
903 initiating connection being dropped. This mattered most when the callout
904 was marked defer_ok. Fix to keep the two timeout-detection methods
907 JH/15 Relax results from ACL control request to enable cutthrough, in
908 unsupported situations, from error to silently (except under debug)
909 ignoring. This covers use with PRDR, frozen messages, queue-only and
912 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
914 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
915 metadata, resulting in a crash in free().
917 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
918 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
919 Broken also in d185889f4, with init system revamp.
921 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
922 Previously we abruptly closed the connection after reading a malware-
923 found indication; now we go on to read the "scan ok" response line,
926 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
927 ACL. Previously, a crash would result.
929 JH/19 Speed up macro lookups during configuration file read, by skipping non-
930 macro text after a replacement (previously it was only once per line) and
931 by skipping builtin macros when searching for an uppercase lead character.
933 JH/20 DANE support moved from Experimental to mainline. The Makefile control
934 for the build is renamed.
936 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
937 was allocated for every new TLS startup, meaning one per message. Fix
938 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
940 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
941 reported the original. Fix to report (as far as possible) the ACL
942 result replacing the original.
944 JH/23 Fix memory leak during multi-message connections using STARTTLS under
945 OpenSSL. Certificate information is loaded for every new TLS startup,
946 and the resources needed to be freed.
948 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
950 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
953 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
954 DATA response info to the (existing) per-recipient response info for
955 the "C=" log element. It can have useful tracking info from the
956 destination system. Patch from Simon Arlott.
958 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
959 length value. Previously this would segfault.
961 HS/02 Support Avast multiline protoocol, this allows passing flags to
962 newer versions of the scanner.
964 JH/28 Ensure that variables possibly set during message acceptance are marked
965 dead before release of memory in the daemon loop. This stops complaints
966 about them when the debug_store option is enabled. Discovered specifically
967 for sender_rate_period, but applies to a whole set of variables.
968 Do the same for the queue-runner and queue-list loops, for variables set
969 from spool message files. Do the same for the SMTP per-message loop, for
970 certain variables indirectly set in ACL operations.
972 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
973 as a multi-recipient message from a mailinglist manager). The coding had
974 an arbitrary cutoff number of characters while checking for more input;
975 enforced by writing a NUL into the buffer. This corrupted long / fast
976 input. The problem was exposed more widely when more pipelineing of SMTP
977 responses was introduced, and one Exim system was feeding another.
978 The symptom is log complaints of SMTP syntax error (NUL chars) on the
979 receiving system, and refused recipients seen by the sending system
980 (propating to people being dropped from mailing lists).
981 Discovered and pinpointed by David Carter.
983 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
984 replaced by the ${authresults } expansion.
986 JH/31 Bug 2257: Fix pipe transport to not use a socket-only syscall.
988 HS/03 Set a handler for SIGTERM and call exit(3) if running as PID 1. This
989 allows proper process termination in container environments.
991 JH/32 Bug 2258: Fix spool_wireformat in combination with LMTP transport.
992 Previously the "final dot" had a newline after it; ensure it is CR,LF.
994 JH/33 SPF: remove support for the "spf" ACL condition outcome values "err_temp"
995 and "err_perm", deprecated since 4.83 when the RFC-defined words
996 "temperror" and "permerror" were introduced.
998 JH/34 Re-introduce enforcement of no cutthrough delivery on transports having
999 transport-filters or DKIM-signing. The restriction was lost in the
1000 consolidation of verify-callout and delivery SMTP handling.
1001 Extend the restriction to also cover ARC-signing.
1003 JH/35 Cutthrough: for a final-dot response timeout (and nonunderstood responses)
1004 in defer=pass mode supply a 450 to the initiator. Previously the message
1007 PP/02 DANE: add dane_require_tls_ciphers SMTP Transport option; if unset,
1008 tls_require_ciphers is used as before.
1010 HS/03 Malware Avast: Better match the Avast multiline protocol. Add
1011 "pass_unscanned". Only tmpfails from the scanner are written to
1012 the paniclog, as they may require admin intervention (permission
1013 denied, license issues). Other scanner errors (like decompression
1014 bombs) do not cause a paniclog entry.
1016 JH/36 Fix reinitialisation of DKIM logging variable between messages.
1017 Previously it was possible to log spurious information in receive log
1020 JH/37 Bug 2255: Revert the disable of the OpenSSL session caching. This
1021 triggered odd behaviour from Outlook Express clients.
1023 PP/03 Add util/renew-opendmarc-tlds.sh script for safe renewal of public
1026 JH/38 DKIM: accept Ed25519 pubkeys in SubjectPublicKeyInfo-wrapped form,
1027 since the IETF WG has not yet settled on that versus the original
1028 "bare" representation.
1030 JH/39 Fix syslog logging for syslog_timestamp=no and log_selector +millisec.
1031 Previously the millisecond value corrupted the output.
1032 Fix also for syslog_pid=no and log_selector +pid, for which the pid
1033 corrupted the output.
1039 JH/01 Rework error string handling in TLS interface so that the caller in
1040 more cases is responsible for logging. This permits library-sourced
1041 string to be attached to addresses during delivery, and collapses
1042 pairs of long lines into single ones.
1044 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
1045 during configuration. Wildcards are allowed and expanded.
1047 JH/02 Rework error string handling in DKIM to pass more info back to callers.
1048 This permits better logging.
1050 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
1051 do not close it down and have the child transport start it up again on
1052 the passed-on TCP connection. Instead, proxy the child (and any
1053 subsequent ones) for TLS via a unix-domain socket channel. Logging is
1054 affected: the continued delivery log lines do not have any DNSSEC, TLS
1055 Certificate or OCSP information. TLS cipher information is still logged.
1057 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
1058 identical IP addresses on different listening ports. Will also affect
1061 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
1062 add noisy ifdef guards to special-case this sillyness.
1063 Patch from Bernd Kuhls.
1065 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
1066 than 255 are no longer allowed.
1068 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
1069 Disable the session-cache too, which might reduce our load. Since we
1070 currrectly use a new context for every connection, both as server and
1071 client, there is no benefit for these.
1072 GnuTLS appears to not support tickets server-side by default (we don't
1073 call gnutls_session_ticket_enable_server()) but client side is enabled
1074 by default on recent versions (3.1.3 +) unless the PFS priority string
1077 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
1078 <https://reproducible-builds.org/specs/source-date-epoch/>.
1080 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
1081 the check for any unsuccessful recipients did not notice the limit, and
1082 erroneously found still-pending ones.
1084 JH/08 Pipeline CHUNKING command and data together, on kernels that support
1085 MSG_MORE. Only in-clear (not on TLS connections).
1087 JH/09 Avoid using a temporary file during transport using dkim. Unless a
1088 transport-filter is involved we can buffer the headers in memory for
1089 creating the signature, and read the spool data file once for the
1090 signature and again for transmission.
1092 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
1093 4.77 as the kernel support then wasn't solid, having issues in 64bit
1094 mode. Now, it's been long enough. Add support for FreeBSD also.
1096 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
1097 case where the routing stage had gathered several addresses to send to
1098 a host before calling the transport for the first, we previously failed
1099 to close down TLS in the old transport process before passing the TCP
1100 connection to the new process. The new one sent a STARTTLS command
1101 which naturally failed, giving a failed delivery and bloating the retry
1102 database. Investigation and fix prototype from Wolfgang Breyha.
1104 JH/12 Fix check on SMTP command input synchronisation. Previously there were
1105 false-negatives in the check that the sender had not preempted a response
1106 or prompt from Exim (running as a server), due to that code's lack of
1107 awareness of the SMTP input buffering.
1109 PP/04 Add commandline_checks_require_admin option.
1110 Exim drops privileges sanely, various checks such as -be aren't a
1111 security problem, as long as you trust local users with access to their
1112 own account. When invoked by services which pass untrusted data to
1113 Exim, this might be an issue. Set this option in main configuration
1114 AND make fixes to the calling application, such as using `--` to stop
1117 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
1118 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
1119 responses to those, into a single TLS record each way (this usually means
1120 a single packet). As a side issue, smtp_enforce_sync now works on TLS
1123 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
1124 affects you only if you're dancing at the edge of the param size limits.
1125 If you are, and this message makes sense to you, then: raise the
1126 configured limit or use OpenSSL 1.1. Nothing we can do for older
1129 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
1130 cmdline element to get the documented default one. Previously it was
1133 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
1134 are unsuable for non-message-specific reasons - eg. connection timeout,
1135 banner-time rejection.
1137 JH/16 Fix logging of delivery remote port, when specified by router, under
1140 PP/06 Repair manualroute's ability to take options in any order, even if one
1141 is the name of a transport.
1144 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
1146 JH/17 Change the list-building routines interface to use the expanding-string
1147 triplet model, for better allocation and copying behaviour.
1149 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
1150 Previously it was constructed the first time a possibly-matching string
1151 was met in the configuration file input during startup; now it is done
1154 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
1155 compatible one, to avoid the (poorly documented) possibility of a config
1156 file in the working directory redirecting the DB files, possibly correpting
1157 some existing file. CVE-2017-10140 assigned for BDB.
1159 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
1160 cache-hot. Previously, although the result was properly cached, the
1161 initial verify call returned a defer.
1163 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
1164 the main verify for receipient in uncached-mode.
1166 JH/22 Retire historical build files to an "unsupported" subdir. These are
1167 defined as "ones for which we have no current evidence of testing".
1169 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
1170 if present. Previously it was ignored.
1172 JH/24 Start using specified-initialisers in C structure init coding. This is
1173 a C99 feature (it's 2017, so now considered safe).
1175 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
1176 if was a fixed-sized field and bitmask ops via macros; it is now more
1179 PP/07 GitHub PR 56: Apply MariaDB build fix.
1180 Patch provided by Jaroslav Škarvada.
1182 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
1183 during Coverity cleanups [4.87 JH/47]
1184 Diagnosis and fix provided by Michael Fischer v. Mollard.
1186 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
1187 the right size to place the terminating semicolon on its own folded
1188 line, the header hash was calculated to an incorrect value thanks to
1189 the (relaxed) space the fold became.
1191 HS/02 Fix Bug 2130: large writes from the transport subprocess were chunked
1192 and confused the parent.
1194 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
1195 which could crash as a result. This could lead to undeliverable messages.
1197 JH/28 Logging: "next input sent too soon" now shows where input was truncated
1200 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
1201 matters on fast-turnover and PID-randomising systems, which were getting
1202 out-of-order delivery.
1204 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
1205 a possibly-overlapping copy. The symptom was that "Remote host closed
1206 connection in response to HELO" was logged instead of the actual 4xx
1209 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
1210 Previously only that bufferd was discarded, resulting in SYMTP command
1213 JH/32 DKIM: when a message has multiple signatures matching an identity given
1214 in dkim_verify_signers, run the dkim acl once for each. Previously only
1215 one run was done. Bug 2189.
1217 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
1218 "panic the current process" to "deliberately defer". The panic log is
1219 still written with the problem list name; the mail and reject logs now
1220 get a temp-reject line for the message that was being handled, saying
1221 something like "domains check lookup or other defer". The SMTP 451
1222 message is still "Temporary local problem".
1224 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
1225 A crafted sequence of BDAT commands could result in in-use memory beeing
1226 freed. CVE-2017-16943.
1228 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
1229 from SMTP input. Previously it was always done; now only done for DATA
1230 and not BDAT commands. CVE-2017-16944.
1232 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
1233 to the message (such as an overlong header line). Previously this was
1234 not done and we did not exit BDAT mode. Followon from the previous item
1235 though a different problem.
1241 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
1242 than -2003 did; needs libidn2 in addition to libidn.
1244 JH/02 The path option on a pipe transport is now expanded before use.
1246 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
1247 Patch provided by "Björn", documentation fix added too.
1249 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
1250 missing a wire-to-host endian conversion.
1252 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
1253 close after a BDAT command line could be taken as a following command,
1254 giving a synch failure. Fix by only checking for synch immediately
1255 before acknowledging the chunk.
1257 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
1258 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
1259 macro. Patches provided by Josh Soref.
1261 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
1262 Previously we did not; the RFC seems ambiguous and VRFY is not listed
1263 by IANA as a service extension. However, John Klensin suggests that we
1266 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
1267 the dkim code may be unix-mode line endings rather than smtp wire-format
1268 CRLF, so prepend a CR to any bare LF.
1270 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
1271 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
1273 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
1274 the first were themselves being wrongly included in the feed into dkim
1275 processing; with most chunk sizes in use this resulted in an incorrect
1276 body hash calculated value.
1278 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
1279 DKIM signature block, for verification. Although advised against by
1280 standards it is specifically not ruled illegal.
1282 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
1284 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
1285 missing a body hash (the bh= tag).
1287 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
1288 It seems that HAProxy sends the Proxy Protocol information in clear and
1289 only then does a TLS startup, so do the same.
1291 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
1292 TCP connections (such as for Spamd) unless the daemon successfully set
1293 Fast Open mode on its listening sockets. This fixes breakage seen on
1294 too-old kernels or those not configured for Fast Open, at the cost of
1295 requiring both directions being enabled for TFO, and TFO never being used
1296 by non-daemon-related Exim processes.
1298 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
1299 endings, at least on the first header line. Try to canonify any that get
1300 past that check, despite the cost.
1302 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
1303 now limited to an arbitrary five deep, while parsing addresses with the
1304 strip_excess_angle_brackets option enabled.
1306 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
1307 instead leave the unprompted TLS handshake in socket buffer for the
1308 TLS library to consume.
1310 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
1312 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
1314 JH/16 Drop variables when they go out of scope. Memory management drops a whole
1315 region in one operation, for speed, and this leaves assigned pointers
1316 dangling. Add checks run only under the testsuite which checks all
1317 variables at a store-reset and panics on a dangling pointer; add code
1318 explicitly nulling out all the variables discovered. Fixes one known
1319 bug: a transport crash, where a dangling pointer for $sending_ip_address
1320 originally assigned in a verify callout, is re-used.
1322 PP/06 Drop '.' from @INC in various Perl scripts.
1324 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
1326 PP/08 Reduce a number of compilation warnings under clang; building with
1327 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
1328 should be warning-free.
1330 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
1332 HS/01 Fix portability problems introduced by PP/08 for platforms where
1333 realloc(NULL) is not equivalent to malloc() [SunOS et al].
1335 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
1336 chunk. This allows us to accept broken chunked messages. We need a more
1337 general solution here.
1339 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
1340 already-broken messages in the queue.
1342 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
1344 JH/19 Fix reference counting bug in routing-generated-address tracking.
1350 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
1351 supports it and a size is available (ie. the sending peer gave us one).
1353 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
1354 years of being deprecated). The replacements are the ACLs
1355 acl_smtp_mime and acl_not_smtp_mime.
1357 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
1358 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
1359 or even an in-clear connection were permitted. Now, if the host lookup
1360 was dnssec and dane was requested then the host is only used if the
1361 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
1362 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
1363 if one fails this test.
1364 This means that a poorly-configured remote DNS will make it incommunicado;
1365 but it protects against a DNS-interception attack on it.
1367 JH/04 Bug 1810: make continued-use of an open smtp transport connection
1368 non-noisy when a race steals the message being considered.
1370 JH/05 If main configuration option tls_certificate is unset, generate a
1371 self-signed certificate for inbound TLS connections.
1373 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
1374 in rewrites and routers.
1376 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
1377 and logged a warning sing 4.83; now they are a configuration file error.
1379 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
1380 (lacking @domain). Apply the same qualification processing as RCPT.
1382 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
1384 JH/10 Support ${sha256:} applied to a string (as well as the previous
1387 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
1388 a cutthrough deliver is pending, as we always want to make a connection.
1389 This also avoids re-routing the message when later placing the cutthrough
1390 connection after a verify cache hit.
1391 Do not update it with the verify result either.
1393 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
1394 when routing results in more than one destination address.
1396 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
1397 signing (which inhibits the cutthrough capability). Previously only
1398 the presence of an option was tested; now an expansion evaluating as
1399 empty is permissible (obviously it should depend only on data available
1400 when the cutthrough connection is made).
1402 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
1403 the relevant preceding SMTP command did not note the pipelining mode.
1405 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
1406 Previously they were not counted.
1408 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
1409 as one having no matching records. Previously we deferred the message
1410 that needed the lookup.
1412 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
1413 distinguished as "(=".
1415 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
1416 for missing MX records. Previously it only worked for missing A records.
1418 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
1420 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
1421 after the data-go-ahead and data-ack. Patch from Jason Betts.
1423 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
1424 even for a "none" policy. Patch from Tony Meyer.
1426 JH/22 Fix continued use of a connection for further deliveries. If a port was
1427 specified by a router, it must also match for the delivery to be
1430 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
1431 When one of the recipients of a message was unsuitable for the connection
1432 (has no matching addresses), we lost track of needing to mark it
1433 deferred. As a result mail would be lost.
1435 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
1437 JH/25 Decoding ACL controls is now done using a binary search; the source code
1438 takes up less space and should be simpler to maintain. Merge the ACL
1439 condition decode tables also, with similar effect.
1441 JH/26 Fix problem with one_time used on a redirect router which returned the
1442 parent address unchanged. A retry would see the parent address marked as
1443 delivered, so not attempt the (identical) child. As a result mail would
1446 JH/27 Fix a possible security hole, wherein a process operating with the Exim
1447 UID can gain a root shell. Credit to http://www.halfdog.net/ for
1448 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
1451 JH/28 Enable {spool,log} filesystem space and inode checks as default.
1452 Main config options check_{log,spool}_{inodes,space} are now
1453 100 inodes, 10MB unless set otherwise in the configuration.
1455 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
1456 Previously it only applied to the main-section connection policy
1459 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
1461 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
1462 by me. Added RFC7919 DH primes as an alternative.
1464 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
1465 are not in the system include path.
1467 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
1468 GnuTLS, when a session startup failed (eg because the client disconnected)
1469 Exim did stdio operations after fclose. This was exposed by a recent
1470 change which nulled out the file handle after the fclose.
1472 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
1473 signed directly by the cert-signing cert, rather than an intermediate
1474 OCSP-signing cert. This is the model used by LetsEncrypt.
1476 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
1478 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
1479 an incoming connection.
1481 HS/02 Bug 1802: Do not half-close the connection after sending a request
1484 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
1485 fallback to "prime256v1".
1487 JH/34 SECURITY: Use proper copy of DATA command in error message.
1488 Could leak key material. Remotely exploitable. CVE-2016-9963.
1494 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
1495 and 3.4.4 - once the server is enabled to respond to an OCSP request
1496 it does even when not requested, resulting in a stapling non-aware
1497 client dropping the TLS connection.
1499 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
1500 support variable-length bit vectors. No functional change.
1502 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
1503 The I= interface field on outgoing lines is now after the H= remote
1504 host field, same as incoming lines. There is a separate
1505 outgoing_interface log selector which allows you to disable the
1508 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
1509 If not running log_selector +smtp_connection the mainlog would be held
1510 open indefinitely after a "too many connections" event, including to a
1511 deleted file after a log rotate. Leave the per net connection logging
1512 leaving it open for efficiency as that will be quickly detected by the
1513 check on the next write.
1515 HS/01 Bug 1671: Fix post transport crash.
1516 Processing the wait-<transport> messages could crash the delivery
1517 process if the message IDs didn't exist for some reason. When
1518 using 'split_spool_directory=yes' the construction of the spool
1519 file name failed already, exposing the same netto behaviour.
1521 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
1522 mime_regex ACL conditions.
1524 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
1525 to DSN fail messages (bounces): remote IP, remote greeting, remote response
1526 to HELO, local diagnostic string.
1528 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
1529 log line to debug. Even when configured with a tls authenticator many
1530 client connections are expected to not authenticate in this way, so
1531 an authenticate fail is not an error.
1533 HS/02 Add the Exim version string to the process info. This way exiwhat
1534 gives some more detail about the running daemon.
1536 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
1537 matter for fast-change records such as DNSBLs.
1539 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
1540 retry record, even if constant. There may be multiple transports with
1541 different interface settings and the retry behaviour needs to be kept
1544 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
1546 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
1548 JH/10 Bug 840: fix log_defer_output option of pipe transport
1550 JH/11 Bug 830: use same host for all RCPTS of a message, even under
1551 hosts_randomize. This matters a lot when combined with mua_wrapper.
1553 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
1554 ${quote_pgsql:<string>} operator.
1556 JH/13 Bug 1708: avoid misaligned access in cached lookup.
1558 JH/14 Change header file name for freeradius-client. Relevant if compiling
1559 with Radius support; from the Gentoo tree and checked under Fedora.
1561 JH/15 Bug 1712: Introduce $prdr_requested flag variable
1563 JH/16 Bug 1714: Permit an empty string as expansion result for transport
1564 option transport_filter, meaning no filtering.
1566 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
1568 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
1569 defaults to "*" (all hosts). The variable is now available when not built
1570 with TLS, default unset, mainly to enable keeping the testsuite sane.
1571 If a server certificate is not supplied (via tls_certificate) an error is
1572 logged, and clients will find TLS connections fail on startup. Presumably
1573 they will retry in-clear.
1574 Packagers of Exim are strongly encouraged to create a server certificate
1575 at installation time.
1577 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
1578 with the $config_file variable.
1580 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
1581 in transport context, after the attempt, and per-recipient. The latter type
1582 is per host attempted. The event data is the error message, and the errno
1583 information encodes the lookup type (A vs. MX) used for the (first) host,
1584 and the trailing two digits of the smtp 4xx response.
1586 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
1587 to write to mainlog (or rejectlog, paniclog) in the window between file
1588 creation and permissions/ownership being changed. Particularly affects
1589 installations where exicyclog is run as root, rather than exim user;
1590 result is that the running daemon panics and dies.
1592 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
1594 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
1595 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
1596 "pri" and "weight". Note that the previous implicit priority given by the
1597 list order is no longer honoured.
1599 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
1600 for DKIM processing.
1602 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
1603 by defining SUPPORT_SOCKS.
1605 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
1606 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
1607 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
1608 variables are renamed to proxy_{local,external}_{address,port}.
1610 JH/27 Move Internationalisation support from Experimental to mainline, enabled
1611 for a build by defining SUPPORT_I18N
1613 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
1614 of the query string, and make ${quote_redis:} do that quoting.
1616 JH/29 Move Events support from Experimental to mainline, enabled by default
1617 and removable for a build by defining DISABLE_EVENT.
1619 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
1621 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
1622 cached by the daemon.
1624 JH/32 Move Redis support from Experimental to mainline, enabled for a build
1625 by defining LOOKUP_REDIS. The libhiredis library is required.
1627 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
1628 keys are given for lookup.
1630 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
1631 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
1632 only supported when built with TLS support. The PolarSSL SHA routines
1633 are still used when the TLS library is too old for convenient support.
1635 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
1636 openssl_options), for security. OpenSSL forces this from version 1.1.0
1637 server-side so match that on older versions.
1639 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
1640 allocation for $value could be released as the expansion processing
1641 concluded, but leaving the global pointer active for it.
1643 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
1644 and to use the domains and local_parts ACL conditions.
1646 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
1647 incorrectly not doubled on cutthrough transmission, hence seen as a
1648 body-termination at the receiving system - resulting in truncated mails.
1649 Commonly the sender saw a TCP-level error, and retransmitted the message
1650 via the normal store-and-forward channel. This could result in duplicates
1651 received - but deduplicating mailstores were liable to retain only the
1652 initial truncated version.
1654 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
1656 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
1658 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
1659 we're in there, support oversigning also; bug 1309.
1661 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
1663 HS/04 Add support for keep_environment and add_environment options.
1665 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
1666 either intentional arithmetic overflow during PRNG, or testing config-
1669 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
1670 delivery resulted in actual delivery. Cancel cutthrough before DATA
1673 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
1674 rejects a recipient: pass the reject to the originator.
1676 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
1677 Many were false-positives and ignorable, but it's worth fixing the
1680 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
1681 for the new environment-manipulation done at startup. Move the routines
1682 from being local to tls.c to being global via the os.c file.
1684 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
1685 an extract embedded as result-arg for a map, the first arg for extract
1686 is unavailable so we cannot tell if this is a numbered or keyed
1687 extraction. Accept either.
1693 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
1696 JH/02 The smtp transport option "multi_domain" is now expanded.
1698 JH/03 The smtp transport now requests PRDR by default, if the server offers
1701 JH/04 Certificate name checking on server certificates, when exim is a client,
1702 is now done by default. The transport option tls_verify_cert_hostnames
1703 can be used to disable this per-host. The build option
1704 EXPERIMENTAL_CERTNAMES is withdrawn.
1706 JH/05 The value of the tls_verify_certificates smtp transport and main options
1707 default to the word "system" to access the system default CA bundle.
1708 For GnuTLS, only version 3.0.20 or later.
1710 JH/06 Verification of the server certificate for a TLS connection is now tried
1711 (but not required) by default. The verification status is now logged by
1712 default, for both outbound TLS and client-certificate supplying inbound
1715 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
1718 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
1719 Status Notification (bounce) messages are now MIME format per RFC 3464.
1720 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
1721 under the control of the dsn_advertise_hosts option, and routers may
1722 have a dsn_lasthop option.
1724 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
1725 default, modifiable by a malware= option. The list separator for
1726 the options can now be changed in the usual way. Bug 68.
1728 JH/10 The smtp_receive_timeout main option is now expanded before use.
1730 JH/11 The incoming_interface log option now also enables logging of the
1731 local interface on delivery outgoing connections.
1733 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
1734 if the interface and destination host and port all match.
1736 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
1739 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
1740 Patch from Andrew Lewis.
1742 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
1743 now supports optional time-restrictions, weighting, and priority
1744 modifiers per server. Patch originally by <rommer@active.by>.
1746 JH/16 The spamd_address main option now supports a mixed list of local
1747 and remote servers. Remote servers can be IPv6 addresses, and
1748 specify a port-range.
1750 JH/17 Bug 68: The spamd_address main option now supports an optional
1751 timeout value per server.
1753 JH/18 Bug 1581: Router and transport options headers_add/remove can
1754 now have the list separator specified.
1756 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
1759 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
1762 JH/21 Support for the A6 type of dns record is withdrawn.
1764 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
1765 rather than the verbs used.
1767 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
1768 from 255 to 1024 chars.
1770 JH/24 Verification callouts now attempt to use TLS by default.
1772 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
1773 are generic router options now. The defaults didn't change.
1775 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
1776 Original patch from Alexander Shikoff, worked over by JH.
1778 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
1779 files if ZCAT_COMMAND is not executable.
1781 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
1783 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
1785 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
1786 Normally benign, it bites when the pair was led to by a CNAME;
1787 modern usage is to not canonicalize the domain to a CNAME target
1788 (and we were inconsistent anyway for A-only vs AAAA+A).
1790 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
1792 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
1793 when evaluating $sender_host_dnssec.
1795 JH/31 Check the HELO verification lookup for DNSSEC, adding new
1796 $sender_helo_dnssec variable.
1798 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
1800 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
1802 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
1804 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
1805 documented as working, but never had. Support all but $spam_report.
1807 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
1808 added for tls authenticator.
1810 HS/03 Add perl_taintmode main config option
1816 TL/01 When running the test suite, the README says that variables such as
1817 no_msglog_check are global and can be placed anywhere in a specific
1818 test's script, however it was observed that placement needed to be near
1819 the beginning for it to behave that way. Changed the runtest perl
1820 script to read through the entire script once to detect and set these
1821 variables, reset to the beginning of the script, and then run through
1822 the script parsing/test process like normal.
1824 TL/02 The BSD's have an arc4random API. One of the functions to induce
1825 adding randomness was arc4random_stir(), but it has been removed in
1826 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
1827 function when detected.
1829 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
1830 cause callback expansion.
1832 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
1833 syntax errors in an expansion can be treated as a string instead of
1834 logging or causing an error, due to the internal use of bool_lax
1835 instead of bool when processing it.
1837 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
1838 server certificates when making smtp deliveries.
1840 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
1842 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
1844 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
1846 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
1847 Merged patch from Sebastian Wiedenroth.
1849 JH/05 Fix results-pipe from transport process. Several recipients, combined
1850 with certificate use, exposed issues where response data items split
1851 over buffer boundaries were not parsed properly. This eventually
1852 resulted in duplicates being sent. This issue only became common enough
1853 to notice due to the introduction of connection certificate information,
1854 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1856 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1857 size buffer was used, resulting in syntax errors when an expansion
1860 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1861 version 3.3.6 or later.
1863 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1864 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1865 both become "event_action", the variables become $event_name, $event_data
1866 and $event_defer_errno. There is a new variable $verify_mode, usable in
1867 routers, transports and related events. The tls:cert event is now also
1868 raised for inbound connections, if the main configuration event_action
1871 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1872 early OCSP support, but no stapling (appears to be less than 1.0.0).
1874 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1875 server certificate names available under the smtp transport option
1876 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1879 JH/10 Time-related extraction expansions from certificates now use the main
1880 option "timezone" setting for output formatting, and are consistent
1881 between OpenSSL and GnuTLS compilations. Bug 1541.
1883 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1884 encoded parameter in the incoming message. Bug 1558.
1886 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1887 include certificate info, eximon was claiming there were spoolfile
1890 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1892 JH/14 Log delivery-related information more consistently, using the sequence
1893 "H=<name> [<ip>]" wherever possible.
1895 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1896 are problematic for Debian distribution, omit them from the release
1899 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1901 JH/16 Fix string representation of time values on 64bit time_t architectures.
1904 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1905 output list separator was used.
1910 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1911 checkers that were complaining about end of non-void function with no
1914 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1915 This was a regression introduced in 4.83 by another bugfix.
1917 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1919 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1920 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1926 TF/01 Correctly close the server side of TLS when forking for delivery.
1928 When a message was received over SMTP with TLS, Exim failed to clear up
1929 the incoming connection properly after forking off the child process to
1930 deliver the message. In some situations the subsequent outgoing
1931 delivery connection happened to have the same fd number as the incoming
1932 connection previously had. Exim would try to use TLS and fail, logging
1933 a "Bad file descriptor" error.
1935 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1936 utilities have not been installed.
1938 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1939 temporary space as the ACL may create new global variables.
1941 TL/01 LDAP support uses per connection or global context settings, depending
1942 upon the detected version of the libraries at build time.
1944 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1945 to extract and use the src ip:port in logging and expansions as if it
1946 were a direct connection from the outside internet. PPv2 support was
1947 updated based on HAProxy spec change in May 2014.
1949 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1951 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1952 Properly escape header and check for NULL return.
1954 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1957 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1959 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1960 characters in header names, implemented as a verify condition.
1961 Contributed by Michael Fischer v. Mollard.
1963 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1964 results permerror and temperror. Previous values are deprecated but
1965 still accepted. In a future release, err_perm and err_temp will be
1966 completely removed, which will be a backward incompatibility if the
1967 ACL tests for either of these two old results. Patch contributed by
1968 user bes-internal on the mailing list.
1970 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1972 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1973 selectors, in both main and reject logs.
1975 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1978 JH/07 Add malware type "sock" for talking to simple daemon.
1980 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1982 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1983 routers/transports under cutthrough routing.
1985 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1986 numbers. Touch up "bool" conditional to keep the same definition.
1988 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1990 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1992 JH/12 Expand items in router/transport headers_add or headers_remove lists
1993 individually rather than the list as a whole. Bug 1452.
1995 Required for reasonable handling of multiple headers_ options when
1996 they may be empty; requires that headers_remove items with embedded
1997 colons must have them doubled (or the list-separator changed).
1999 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
2000 view the policy declared in the DMARC record. Currently, $dmarc_status
2001 is a combined value of both the record presence and the result of the
2004 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
2006 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
2007 dnslookup router and the smtp transport (applying to the forward
2010 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
2011 of ldap servers used for a specific lookup. Patch provided by Heiko
2014 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
2015 New variable $lookup_dnssec_authenticated for observability.
2017 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
2018 Patch submitted by Lars Timman.
2020 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
2022 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
2023 Requires trusted mode and valid format message id, aborts otherwise.
2024 Patch contributed by Heiko Schlichting.
2026 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
2027 certextract with support for various fields. Bug 1358.
2029 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
2030 is requested by default, modifiable by smtp transport option
2033 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
2034 operate on certificate variables to give certificate fingerprints
2035 Also new ${sha256:cert_variable}.
2037 JH/23 The PRDR feature is moved from being Experimental into the mainline.
2039 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
2040 Christian Aistleitner.
2042 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
2044 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
2045 file. Patch from Wolfgang Breyha.
2047 JH/25 Expand the coverage of the delivery $host and $host_address to
2048 client authenticators run in verify callout. Bug 1476.
2050 JH/26 Port service names are now accepted for tls_on_connect_ports, to
2051 align with daemon_smtp_ports. Bug 72.
2053 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
2054 support and error reporting did not work properly.
2056 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
2057 and is readable. Patch from Andrew Colin Kissa.
2059 TL/14 Enhance documentation of ${run expansion and how it parses the
2060 commandline after expansion, particularly in the case when an
2061 unquoted variable expansion results in an empty value.
2063 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
2065 PP/02 Fix internal collision of T_APL on systems which support RFC3123
2066 by renaming away from it. Addresses GH issue 15, reported by
2069 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
2071 TL/15 SECURITY: prevent double expansion in math comparison functions
2072 (can expand unsanitized data). Not remotely exploitable.
2079 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
2081 PP/02 Make -n do something, by making it not do something.
2082 When combined with -bP, the name of an option is not output.
2084 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
2087 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
2088 $sender_host_name and config options to manage this, and basic check
2091 PP/05 DSCP support for outbound connections and control modifier for inbound.
2093 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
2094 (Only plugin which currently uses this is kerberos4, which nobody should
2095 be using, but we should make it available and other future plugins might
2096 conceivably use it, even though it would break NAT; stuff *should* be
2097 using channel bindings instead).
2099 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
2100 name; added for Sendmail compatibility; requires admin caller.
2101 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
2102 just ignore it); requires trusted caller.
2103 Also parse but ignore: -Ac -Am -X<logfile>
2106 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
2108 TL/02 Add +smtp_confirmation as a default logging option.
2110 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
2111 Patch by Magnus Holmgren from 2007-02-20.
2113 TL/04 Bugzilla 1281 - Spec typo.
2114 Bugzilla 1283 - Spec typo.
2115 Bugzilla 1290 - Spec grammar fixes.
2117 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
2119 TL/06 Add Experimental DMARC support using libopendmarc libraries.
2121 TL/07 Fix an out of order global option causing a segfault. Reported to dev
2122 mailing list by by Dmitry Isaikin.
2124 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
2126 JH/02 Support "G" suffix to numbers in ${if comparisons.
2128 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
2130 NM/01 Bugzilla 1197 - Spec typo
2131 Bugzilla 1196 - Spec examples corrections
2133 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
2135 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
2136 gnutls_enable_pkcs11, but renamed to more accurately indicate its
2139 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
2140 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
2142 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
2143 "acl {{name}{arg}...}", and optional args on acl condition
2146 JH/05 Permit multiple router/transport headers_add/remove lines.
2148 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
2150 JH/07 Avoid using a waiting database for a single-message-only transport.
2151 Performance patch from Paul Fisher. Bugzilla 1262.
2153 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
2156 JH/09 Add $headers_added variable, with content from use of ACL modifier
2157 add_header (but not yet added to the message). Bugzilla 199.
2159 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
2160 Pulled from Bugzilla 817 by Wolfgang Breyha.
2162 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
2164 (nb: this is the same fix as in Exim 4.80.1)
2166 JH/11 Add A= logging on delivery lines, and a client_set_id option on
2169 JH/12 Add optional authenticated_sender logging to A= and a log_selector
2172 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
2174 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
2175 advertise SMTP AUTH mechanism to us, instead of a generic
2176 protocol violation error. Also, make Exim more robust to bad
2177 data from the Dovecot auth socket.
2179 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
2181 When a queue runner is handling a message, Exim first routes the
2182 recipient addresses, during which it prunes them based on the retry
2183 hints database. After that it attempts to deliver the message to
2184 any remaining recipients. It then updates the hints database using
2187 So if a recipient address works intermittently, it can get repeatedly
2188 deferred at routing time. The retry hints record remains fresh so the
2189 address never reaches the final cutoff time.
2191 This is a fairly common occurrence when a user is bumping up against
2192 their storage quota. Exim had some logic in its local delivery code
2193 to deal with this. However it did not apply to per-recipient defers
2194 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
2196 This change adds a proper retry rule check during routing so that the
2197 final cutoff time is checked against the message's age. We only do
2198 this check if there is an address retry record and there is not a
2199 domain retry record; this implies that previous attempts to handle
2200 the address had the retry_use_local_parts option turned on. We use
2201 this as an approximation for the destination being like a local
2202 delivery, as in LMTP.
2204 I suspect this new check makes the old local delivery cutoff check
2205 redundant, but I have not verified this so I left the code in place.
2207 TF/02 Correct gecos expansion when From: is a prefix of the username.
2209 Test 0254 submits a message to Exim with the header
2213 When I ran the test suite under the user fanf2, Exim expanded
2214 the header to contain my full name, whereas it should have added
2215 a Resent-Sender: header. It erroneously treats any prefix of the
2216 username as equal to the username.
2218 This change corrects that bug.
2220 GF/01 DCC debug and logging tidyup
2221 Error conditions log to paniclog rather than rejectlog.
2222 Debug lines prefixed by "DCC: " to remove any ambiguity.
2224 TF/03 Avoid unnecessary rebuilds of lookup-related code.
2226 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
2227 Bug spotted by Jeremy Harris; was flawed since initial commit.
2228 Would have resulted in OCSP responses post-SNI triggering an Exim
2229 NULL dereference and crash.
2231 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
2233 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
2234 Bug detection, analysis and fix by Samuel Thibault.
2235 Bugzilla 1331, Debian bug #698092.
2237 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
2239 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
2240 Server implementation by Todd Lyons, client by JH.
2241 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
2242 config variable "prdr_enable" controls whether the server
2243 advertises the facility. If the client requests PRDR a new
2244 acl_data_smtp_prdr ACL is called once for each recipient, after
2245 the body content is received and before the acl_smtp_data ACL.
2246 The client is controlled by both of: a hosts_try_prdr option
2247 on the smtp transport, and the server advertisement.
2248 Default client logging of deliveries and rejections involving
2249 PRDR are flagged with the string "PRDR".
2251 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
2252 fclose(). Diagnosis by Todd Lyons.
2254 PP/17 Update configure.default to handle IPv6 localhost better.
2255 Patch by Alain Williams (plus minor tweaks).
2258 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
2259 This is now consistent with GnuTLS, and is now documented: the
2260 previous undocumented portable approach to treating the option as
2261 unset was to force an expansion failure. That still works, and
2262 an empty string is now equivalent.
2264 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
2265 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
2266 not performing validation itself.
2268 PP/20 Added force_command boolean option to pipe transport.
2269 Patch from Nick Koston, of cPanel Inc.
2271 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
2274 TF/04 Added udpsend ACL modifier and hexquote expansion operator
2276 PP/21 Fix eximon continuous updating with timestamped log-files.
2277 Broken in a format-string cleanup in 4.80, missed when I repaired the
2278 other false fix of the same issue.
2279 Report and fix from Heiko Schlichting.
2282 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
2283 Report from Prashanth Katuri.
2285 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
2286 It's SecureTransport, so affects any MacOS clients which use the
2287 system-integrated TLS libraries, including email clients.
2289 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
2290 using a MIME ACL for non-SMTP local injection.
2291 Report and assistance in diagnosis by Warren Baker.
2293 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
2295 JH/16 Fix comparisons for 64b. Bugzilla 1385.
2297 TL/09 Add expansion variable $authenticated_fail_id to keep track of
2298 last id that failed so it may be referenced in subsequent ACL's.
2300 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
2303 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
2304 ldap library initialization, allowing self-signed CA's to be
2305 used. Also properly sets require_cert option later in code by
2306 using NULL (global ldap config) instead of ldap handle (per
2307 session). Bug diagnosis and testing by alxgomz.
2309 TL/12 Enhanced documentation in the ratelimit.pl script provided in
2310 the src/util/ subdirectory.
2312 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
2313 renamed to Transport Post Delivery Action by Jeremy Harris, as
2316 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
2317 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
2318 redis_servers = needs to be configured which will be used by the redis
2319 lookup. Patch from Warren Baker, of The Packet Hub.
2321 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
2323 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
2324 hostname or reverse DNS when processing a host list. Used suggestions
2325 from multiple comments on this bug.
2327 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
2329 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
2330 Missed a few lines, added it to make the runtest require no keyboard
2333 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
2334 contains upper case chars. Make router use caseful_local_part.
2336 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
2337 support when GnuTLS has been built with p11-kit.
2343 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
2345 This, or similar/improved, will also be change PP/11 of 4.82.
2351 PP/01 Handle short writes when writing local log-files.
2352 In practice, only affects FreeBSD (8 onwards).
2353 Bugzilla 1053, with thanks to Dmitry Isaikin.
2355 NM/01 Bugzilla 949 - Documentation tweak
2357 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
2360 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
2362 PP/02 Implemented gsasl authenticator.
2364 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
2366 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
2367 `pkg-config foo` for cflags/libs.
2369 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
2370 with rest of GSASL and with heimdal_gssapi.
2372 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
2373 `pkg-config foo` for cflags/libs for the TLS implementation.
2375 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
2376 properties get this fed in as external SSF. A number of robustness
2377 and debugging improvements to the cyrus_sasl authenticator.
2379 PP/08 cyrus_sasl server now expands the server_realm option.
2381 PP/09 Bugzilla 1214 - Log authentication information in reject log.
2382 Patch by Jeremy Harris.
2384 PP/10 Added dbmjz lookup type.
2386 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
2388 PP/12 MAIL args handles TAB as well as SP, for better interop with
2389 non-compliant senders.
2390 Analysis and variant patch by Todd Lyons.
2392 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
2393 Bug report from Lars Müller <lars@samba.org> (via SUSE),
2394 Patch from Dirk Mueller <dmueller@suse.com>
2396 PP/13 tls_peerdn now print-escaped for spool files.
2397 Observed some $tls_peerdn in wild which contained \n, which resulted
2398 in spool file corruption.
2400 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
2401 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
2402 or write after TLS renegotiation, which otherwise led to messages
2405 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
2406 as a tracking header (ie: a signed header comes before the signature).
2407 Patch from Wolfgang Breyha.
2409 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
2410 comma-sep list; embedded commas doubled.
2412 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
2414 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
2416 Report and patch from Dmitry Banschikov.
2418 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
2419 Removed SSL_clear() after SSL_new() which led to protocol negotiation
2420 failures. We appear to now support TLS1.1+ with Exim.
2422 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
2423 lets Exim select keys and certificates based upon TLS SNI from client.
2424 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
2425 before an outbound SMTP session. New log_selector, +tls_sni.
2427 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
2428 NULL dereference. Report and patch from Alun Jones.
2430 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
2431 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
2432 Not seeing resolver debug output on NetBSD, but suspect this is a
2433 resolver implementation change.
2435 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
2436 Left warnings. Added "eximon gdb" invocation mode.
2438 PP/21 Defaulting "accept_8bitmime" to true, not false.
2440 PP/22 Added -bw for inetd wait mode support.
2442 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
2443 locate the relevant includes and libraries. Made this the default.
2445 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
2446 Bugzilla 1246, report and most of solution from Tomasz Kusy.
2448 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
2449 This may cause build issues on older platforms.
2451 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
2452 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
2453 gnutls_require_mac & gnutls_require_protocols (no longer supported).
2454 Added SNI support via GnuTLS too.
2455 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
2457 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
2459 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
2460 Applied second patch from Janne, implementing suggestion to default
2461 multiple-strings-in-record handling to match SPF spec.
2463 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
2465 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
2466 read-only, out of scope).
2467 Patch from Wolfgang Breyha, report from Stuart Northfield.
2469 PP/29 Fix three issues highlighted by clang analyser static analysis.
2470 Only crash-plausible issue would require the Cambridge-specific
2471 iplookup router and a misconfiguration.
2472 Report from Marcin Mirosław.
2474 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
2476 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
2477 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
2478 As part of this, removing so much warning spew let me fix some minor
2479 real issues in debug logging.
2481 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
2482 assignment on my part. Fixed.
2484 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
2485 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
2486 Janne Snabb (who went above and beyond: thank you).
2488 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
2489 string otherwise requires a connection and a bunch more work and it's
2490 relatively easy to get wrong. Should also expose TLS library linkage
2493 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
2494 64-bit ${eval} (JH/03).
2496 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
2497 GNU libc to support some of the 64-bit stuff, should not lead to
2498 conflicts. Defined before os.h is pulled in, so if a given platform
2499 needs to override this, it can.
2501 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
2502 protection layer was required, which is not implemented.
2503 Bugzilla 1254, patch from Wolfgang Breyha.
2505 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
2506 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
2507 tls_dhparam take prime identifiers. Also unbreak combination of
2508 OpenSSL+DH_params+TLSSNI.
2510 PP/39 Disable SSLv2 by default in OpenSSL support.
2516 PP/01 Solaris build fix for Oracle's LDAP libraries.
2517 Bugzilla 1109, patch from Stephen Usher.
2519 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
2521 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
2524 TF/02 Fix a couple more cases where we did not log the error message
2525 when unlink() failed. See also change 4.74-TF/03.
2527 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
2528 lock up or crash if it happened to be inside a call to libc when it
2529 got a SIGUSR1 from exiwhat.
2531 The SIGUSR1 handler appends the current process status to the process
2532 log which is later printed by exiwhat. It used to use the general
2533 purpose logging code to do this, but several functions it calls are
2534 not safe for signals.
2536 The new output code in the SIGUSR1 handler is specific to the process
2537 log, and simple enough that it's easy to inspect for signal safety.
2538 Removing some special cases also simplifies the general logging code.
2539 Removing the spurious timestamps from the process log simplifies
2542 TF/04 Improved ratelimit ACL condition.
2544 The /noupdate option has been deprecated in favour of /readonly which
2545 has clearer semantics. The /leaky, /strict, and /readonly update modes
2546 are mutually exclusive. The update mode is no longer included in the
2547 database key; it just determines when the database is updated. (This
2548 means that when you upgrade Exim will forget old rate measurements.)
2550 Exim now checks that the per_* options are used with an update mode that
2551 makes sense for the current ACL. For example, when Exim is processing a
2552 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
2553 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
2554 must specify per_mail/readonly. If you omit the update mode it defaults to
2555 /leaky where that makes sense (as before) or /readonly where required.
2557 The /noupdate option is now undocumented but still supported for
2558 backwards compatibility. It is equivalent to /readonly except that in
2559 ACLs where /readonly is required you may specify /leaky/noupdate or
2560 /strict/noupdate which are treated the same as /readonly.
2562 A useful new feature is the /count= option. This is a generalization
2563 of the per_byte option, so that you can measure the throughput of other
2564 aggregate values. For example, the per_byte option is now equivalent
2565 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
2567 The per_rcpt option has been generalized using the /count= mechanism
2568 (though it's more complicated than the per_byte equivalence). When it is
2569 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
2570 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
2571 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
2572 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
2573 Note that using per_rcpt with a non-readonly update mode in more than
2574 one ACL will cause the recipients to be double-counted. (The per_mail
2575 and per_byte options don't have this problem.)
2577 The handling of very low rates has changed slightly. If the computed rate
2578 is less than the event's count (usually one) then this event is the first
2579 after a long gap. In this case the rate is set to the same as this event's
2580 count, so that the first message of a spam run is counted properly.
2582 The major new feature is a mechanism for counting the rate of unique
2583 events. The new per_addr option counts the number of different
2584 recipients that someone has sent messages to in the last time period. It
2585 behaves like per_rcpt if all the recipient addresses are different, but
2586 duplicate recipient addresses do not increase the measured rate. Like
2587 the /count= option this is a general mechanism, so the per_addr option
2588 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
2589 example, measure the rate that a client uses different sender addresses
2590 with the options per_mail/unique=$sender_address. There are further
2591 details in the main documentation.
2593 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
2595 TF/06 Removed a few PCRE remnants.
2597 TF/07 Automatically extract Exim's version number from tags in the git
2598 repository when doing development or release builds.
2600 PP/02 Raise smtp_cmd_buffer_size to 16kB.
2601 Bugzilla 879. Patch from Paul Fisher.
2603 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
2604 Heavily based on revision 40f9a89a from Simon Arlott's tree.
2607 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
2609 PP/05 Variable $av_failed, true if the AV scanner deferred.
2610 Bugzilla 1078. Patch from John Horne.
2612 PP/06 Stop make process more reliably on build failure.
2613 Bugzilla 1087. Patch from Heiko Schlittermann.
2615 PP/07 Make maildir_use_size_file an _expandable_ boolean.
2616 Bugzilla 1089. Patch from Heiko Schlittermann.
2618 PP/08 Handle ${run} returning more data than OS pipe buffer size.
2619 Bugzilla 1131. Patch from Holger Weiß.
2621 PP/09 Handle IPv6 addresses with SPF.
2622 Bugzilla 860. Patch from Wolfgang Breyha.
2624 PP/10 GnuTLS: support TLS 1.2 & 1.1.
2626 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
2629 PP/11 match_* no longer expand right-hand-side by default.
2630 New compile-time build option, EXPAND_LISTMATCH_RHS.
2631 New expansion conditions, "inlist", "inlisti".
2633 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
2635 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
2637 PP/14 fix log_write() format string regression from TF/03.
2638 Bugzilla 1152. Patch from Dmitry Isaikin.
2644 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
2646 PP/02 Harmonised TLS library version reporting; only show if debugging.
2647 Layout now matches that introduced for other libraries in 4.74 PP/03.
2649 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
2651 PP/04 New "dns_use_edns0" global option.
2653 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
2656 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
2657 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
2659 TK/01 Updated PolarSSL code to 0.14.2.
2660 Bugzilla 1097. Patch from Andreas Metzler.
2662 PP/07 Catch divide-by-zero in ${eval:...}.
2663 Fixes bugzilla 1102.
2665 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
2668 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
2669 format-string attack -- SECURITY: remote arbitrary code execution.
2671 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
2672 time unintentionally subject to list matching rules, letting the header
2673 cause arbitrary Exim lookups (of items which can occur in lists, *not*
2674 arbitrary string expansion). This allowed for information disclosure.
2676 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
2677 INT_MIN/-1 -- value coerced to INT_MAX.
2683 NM/01 Workaround for PCRE version dependency in version reporting
2686 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
2687 This fixes portability to compilers other than gcc, notably
2688 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
2690 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
2691 makefiles for portability to HP-UX and POSIX correctness.
2693 PP/01 Permit LOOKUP_foo enabling on the make command-line.
2694 Also via indented variable definition in the Makefile.
2695 (Debugging by Oliver Heesakkers).
2697 PP/02 Restore caching of spamd results with expanded spamd_address.
2698 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
2700 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
2701 Improves build reliability. Fix from: Frank Elsner
2703 NM/02 Fix wide character breakage in the rfc2047 coding
2704 Fixes bug 1064. Patch from Andrey N. Oktyabrski
2706 NM/03 Allow underscore in dnslist lookups
2707 Fixes bug 1026. Patch from Graeme Fowler
2709 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
2710 Code patches from Adam Ciarcinski of NetBSD.
2712 NM/04 Fixed exiqgrep to cope with mailq missing size issue
2715 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
2716 is logged, to avoid truncation. Patch from John Horne.
2718 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
2719 Patch from Jakob Hirsch.
2721 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
2722 SQL string expansion failure details.
2723 Patch from Andrey Oktyabrski.
2725 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
2726 Patch from Simon Arlott.
2728 PP/09 New lookups functionality failed to compile on old gcc which rejects
2729 extern declarations in function scope.
2730 Patch from Oliver Fleischmann
2732 PP/10 Use sig_atomic_t for flags set from signal handlers.
2733 Check getgroups() return and improve debugging.
2734 Fixed developed for diagnosis in bug 927 (which turned out to be
2737 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
2738 Patch from Mark Zealey.
2740 PP/12 Bugzilla 1056: Improved spamd server selection.
2741 Patch from Mark Zealey.
2743 PP/13 Bugzilla 1086: Deal with maildir quota file races.
2744 Based on patch from Heiko Schlittermann.
2746 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
2747 Patch from Uwe Doering, sign-off by Michael Haardt.
2749 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
2750 variable declaration deep within a block. Bug and patch from
2753 PP/15 lookups-Makefile IRIX compatibility coercion.
2755 PP/16 Make DISABLE_DKIM build knob functional.
2757 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
2758 Patch by Simon Arlott
2760 TF/03 Fix valgrind.h portability to C89 compilers that do not support
2761 variable argument macros. Our copy now differs from upstream.
2767 TF/01 Failure to get a lock on a hints database can have serious
2768 consequences so log it to the panic log.
2770 TF/02 Log LMTP confirmation messages in the same way as SMTP,
2771 controlled using the smtp_confirmation log selector.
2773 TF/03 Include the error message when we fail to unlink a spool file.
2775 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
2776 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
2777 for maintaining out-of-tree patches for some time.
2779 PP/01 Bugzilla 139: Documentation and portability issues.
2780 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
2781 Handle per-OS dynamic-module compilation flags.
2783 PP/02 Let /dev/null have normal permissions.
2784 The 4.73 fixes were a little too stringent and complained about the
2785 permissions on /dev/null. Exempt it from some checks.
2786 Reported by Andreas M. Kirchwitz.
2788 PP/03 Report version information for many libraries, including
2789 Exim version information for dynamically loaded libraries. Created
2790 version.h, now support a version extension string for distributors
2791 who patch heavily. Dynamic module ABI change.
2793 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
2794 privilege escalation vulnerability whereby the Exim run-time user
2795 can cause root to append content of the attacker's choosing to
2798 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
2801 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
2802 If dropping privileges for untrusted macros, we disabled normal logging
2803 on the basis that it would fail; for the Exim run-time user, this is not
2804 the case, and it resulted in successful deliveries going unlogged.
2805 Fixed. Reported by Andreas Metzler.
2811 PP/01 Date: & Message-Id: revert to normally being appended to a message,
2812 only prepend for the Resent-* case. Fixes regression introduced in
2813 Exim 4.70 by NM/22 for Bugzilla 607.
2815 PP/02 Include check_rfc2047_length in configure.default because we're seeing
2816 increasing numbers of administrators be bitten by this.
2818 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
2820 PP/03 Bugzilla 994: added openssl_options main configuration option.
2822 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
2824 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
2826 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
2828 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
2829 without a peer certificate, leading to a segfault because of an
2830 assumption that peers always have certificates. Be a little more
2831 paranoid. Problem reported by Martin Tscholak.
2833 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
2834 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
2835 NB: ClamAV planning to remove STREAM in "middle of 2010".
2836 CL also introduces -bmalware, various -d+acl logging additions and
2837 more caution in buffer sizes.
2839 PP/09 Implemented reverse_ip expansion operator.
2841 PP/10 Bugzilla 937: provide a "debug" ACL control.
2843 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
2845 PP/12 Bugzilla 973: Implement --version.
2847 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2849 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2851 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2853 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2854 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2855 ignore trailing whitespace.
2857 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2859 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2862 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2863 Notification from Dr Andrew Aitchison.
2865 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2866 ExtendedDetectionInfo response format.
2867 Notification from John Horne.
2869 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2872 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2873 XSL and documented dependency on system catalogs, with examples of how
2876 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2879 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2880 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2881 configuration file which is writeable by the Exim user or group.
2883 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2884 of configuration files to cover files specified with the -C option if
2885 they are going to be used with root privileges, not just the default
2888 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2889 option (effectively making it always true).
2891 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2892 files to be used while preserving root privileges.
2894 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2895 that rogue child processes cannot use them.
2897 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2898 run-time user, instead of root.
2900 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2901 Exim run-time user without dropping privileges.
2903 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2904 result string, instead of calling string_vformat() twice with the same
2907 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2908 for other users. Others should always drop root privileges if they use
2909 -C on the command line, even for a whitelisted configure file.
2911 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2913 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2919 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2920 $data_path, and $header_path variables; fixed documentation bugs and
2923 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2924 exipick to access non-standard spools, including the "frozen" queue
2927 NM/01 Bugzilla 965: Support mysql stored procedures.
2928 Patch from Alain Williams
2930 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2932 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2933 Patch from Andreas Metzler
2935 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2936 Patch from Kirill Miazine
2938 NM/05 Bugzilla 671: Added umask to procmail example.
2940 JJ/03 installed exipick 20100323.0, fixing doc bug
2942 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2943 directory. Notification and patch from Dan Rosenberg.
2945 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2947 TK/02 Improve log output when DKIM signing operation fails.
2949 MH/01 Treat the transport option dkim_domain as a colon separated
2950 list, not as a single string, and sign the message with each element,
2951 omitting multiple occurences of the same signer.
2953 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2954 Bugzilla 985, 986. Patch by Simon Arlott
2956 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2957 Patch by Simon Arlott
2959 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2960 MBX locking. Notification from Dan Rosenberg.
2966 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2968 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2970 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2972 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2974 NM/04 Bugzilla 915: Flag broken perl installation during build.
2980 TK/01 Added patch by Johannes Berg that expands the main option
2981 "spamd_address" if it starts with a dollar sign.
2983 TK/02 Write list of recipients to X-Envelope-Sender header when building
2984 the mbox-format spool file for content scanning (suggested by Jakob
2987 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2988 (http://www.dcc-servers.net/) support via dccifd. Activated by
2989 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2991 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2992 by Mark Daniel Reidel <mr@df.eu>.
2994 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2995 When building exim an external PCRE library is now needed -
2996 PCRE is a system library on the majority of modern systems.
2997 See entry on PCRE_LIBS in EDITME file.
2999 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
3000 conversation. Added nologin parameter to request.
3001 Patch contributed by Kirill Miazine.
3003 TF/01 Do not log submission mode rewrites if they do not change the address.
3005 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
3007 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
3008 log files in place. Contributed by Roberto Lima.
3010 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
3012 TF/03 Bugzilla 615: When checking the local_parts router precondition
3013 after a local_part_suffix or local_part_prefix option, Exim now
3014 does not use the address's named list lookup cache, since this
3015 contains cached lookups for the whole local part.
3017 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
3018 Robert Millan. Documentation is in experimental-spec.txt.
3020 TF/04 Bugzilla 668: Fix parallel build (make -j).
3022 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
3024 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
3025 Patch provided by Jan Srzednicki.
3027 TF/05 Leading white space used to be stripped from $spam_report which
3028 wrecked the formatting. Now it is preserved.
3030 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
3031 that they are available at delivery time.
3033 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
3035 TF/08 TLS error reporting now respects the incoming_interface and
3036 incoming_port log selectors.
3038 TF/09 Produce a more useful error message if an SMTP transport's hosts
3039 setting expands to an empty string.
3041 NM/06 Bugzilla 744: EXPN did not work under TLS.
3042 Patch provided by Phil Pennock.
3044 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
3045 Patch provided by Richard Godbee.
3047 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
3048 acl_smtp_notquit, added index entry.
3050 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
3051 Patch provided by Eugene Bujak.
3053 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
3054 accept(). Patch provided by Maxim Dounin.
3056 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
3057 Patch provided by Phil Pennock.
3059 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
3061 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
3062 Patch provided by Brad "anomie" Jorsch.
3064 NM/14 Bugzilla 622: Added timeout setting to transport filter.
3065 Patch provided by Dean Brooks.
3067 TK/05 Add native DKIM support (does not depend on external libraries).
3069 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
3070 Patch provided by Graeme Fowler.
3072 NM/16 Bugzilla 851: Documentation example syntax fix.
3074 NM/17 Changed NOTICE file to remove references to embedded PCRE.
3076 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
3079 NM/19 Bugzilla 745: TLS version reporting.
3080 Patch provided by Phil Pennock.
3082 NM/20 Bugzilla 167: bool: condition support.
3083 Patch provided by Phil Pennock.
3085 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
3086 clients. Patch provided by Phil Pennock.
3088 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
3089 Patch provided by Brad "anomie" Jorsch.
3091 NM/23 Bugzilla 687: Fix misparses in eximstats.
3092 Patch provided by Heiko Schlittermann.
3094 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
3095 Patch provided by Heiko Schlittermann.
3097 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
3098 plus update to original patch.
3100 NM/26 Bugzilla 799: Documentation correction for ratelimit.
3102 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
3103 Patch provided by David Brownlee.
3105 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
3107 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
3109 NM/30 Bugzilla 888: TLS documentation bugfixes.
3111 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
3113 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
3114 Unlike the original bugzilla I have changed all shell scripts in src tree.
3116 NM/33 Bugzilla 898: Transport filter timeout fix.
3117 Patch by Todd Rinaldo.
3119 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
3120 Patch by Serge Demonchaux.
3122 NM/35 Bugzilla 39: Base64 decode bug fixes.
3123 Patch by Jakob Hirsch.
3125 NM/36 Bugzilla 909: Correct connect() call in dcc code.
3127 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
3129 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
3131 NM/39 Bugzilla 911: Fixed MakeLinks build script.
3137 TK/01 Add preliminary DKIM support. Currently requires a forked version of
3138 ALT-N's libdkim that I have put here:
3139 http://duncanthrax.net/exim-experimental/
3141 Note to Michael Haardt: I had to rename some vars in sieve.c. They
3142 were called 'true' and it seems that C99 defines that as a reserved
3143 keyword to be used with 'bool' variable types. That means you could
3144 not include C99-style headers which use bools without triggering
3145 build errors in sieve.c.
3147 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
3148 as mailq or other aliases. Changed the --help handling significantly
3149 to do whats expected. exim_usage() emits usage/help information.
3151 SC/01 Added the -bylocaldomain option to eximstats.
3153 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
3155 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
3157 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
3163 PH/01 Another patch from the Sieve maintainer.
3165 PH/02 When an IPv6 address is converted to a string for single-key lookup
3166 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
3167 dots are used instead of colons so that keys in lsearch files need not
3168 contain colons. This was done some time before quoting was made available
3169 in lsearch files. However, iplsearch files do require colons in IPv6 keys
3170 (notated using the quote facility) so as to distinguish them from IPv4
3171 keys. This meant that lookups for IP addresses in host lists did not work
3172 for iplsearch lookups.
3174 This has been fixed by arranging for IPv6 addresses to be expressed with
3175 colons if the lookup type is iplsearch. This is not incompatible, because
3176 previously such lookups could never work.
3178 The situation is now rather anomalous, since one *can* have colons in
3179 ordinary lsearch keys. However, making the change in all cases is
3180 incompatible and would probably break a number of configurations.
3182 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
3185 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
3186 conversion specification without a maximum field width, thereby enabling
3187 a rogue spamd server to cause a buffer overflow. While nobody in their
3188 right mind would setup Exim to query an untrusted spamd server, an
3189 attacker that gains access to a server running spamd could potentially
3190 exploit this vulnerability to run arbitrary code as the Exim user.
3192 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
3193 $primary_hostname instead of what libspf2 thinks the hosts name is.
3195 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
3196 a directory entry by the name of the lookup key. Previously, if a
3197 symlink pointed to a non-existing file or a file in a directory that
3198 Exim lacked permissions to read, a lookup for a key matching that
3199 symlink would fail. Now it is enough that a matching directory entry
3200 exists, symlink or not. (Bugzilla 503.)
3202 PH/03 The body_linecount and body_zerocount variables are now exported in the
3205 PH/04 Added the $dnslist_matched variable.
3207 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
3208 This means they are set thereafter only if the connection becomes
3211 PH/06 Added the client_condition to authenticators so that some can be skipped
3212 by clients under certain conditions.
3214 PH/07 The error message for a badly-placed control=no_multiline_responses left
3215 "_responses" off the end of the name.
3217 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
3219 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
3220 (without spaces) instead of just copying the configuration text.
3222 PH/10 Added the /noupdate option to the ratelimit ACL condition.
3224 PH/11 Added $max_received_linelength.
3226 PH/12 Added +ignore_defer and +include_defer to host lists.
3228 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
3229 way in which PCRE > 7.0 is built.
3231 PH/14 Implemented queue_only_load_latch.
3233 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
3234 MAIL command. The effect was to mangle the value on 64-bit systems.
3236 PH/16 Another patch from the Sieve maintainer.
3238 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
3240 PH/18 If a system quota error occurred while trying to create the file for
3241 a maildir delivery, the message "Mailbox is full" was not appended to the
3242 bounce if the delivery eventually timed out. Change 4.67/27 below applied
3243 only to a quota excession during the actual writing of the file.
3245 PH/19 It seems that peer DN values may contain newlines (and other non-printing
3246 characters?) which causes problems in log lines. The DN values are now
3247 passed through string_printing() before being added to log lines.
3249 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
3250 and InterBase are left for another time.)
3252 PH/21 Added message_body_newlines option.
3254 PH/22 Guard against possible overflow in moan_check_errorcopy().
3256 PH/23 POSIX allows open() to be a macro; guard against that.
3258 PH/24 If the recipient of an error message contained an @ in the local part
3259 (suitably quoted, of course), incorrect values were put in $domain and
3260 $local_part during the evaluation of errors_copy.
3266 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
3267 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
3270 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
3271 issue a MAIL command.
3273 PH/02 In an ACL statement such as
3275 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
3277 if a client was not listed at all, or was listed with a value other than
3278 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
3279 the condition was not true (as it should be), so access was not denied.
3280 The bug was that the ! inversion was incorrectly passed on to the second
3281 item. This has been fixed.
3283 PH/03 Added additional dnslists conditions == and =& which are different from
3284 = and & when the dns lookup returns more than one IP address.
3286 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
3287 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
3289 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
3290 FSYNC, which compiles an option called disable_fsync that allows for
3291 bypassing fsync(). The documentation is heavily laced with warnings.
3293 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
3295 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
3296 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
3297 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
3298 including adding "make clean"; (3) Added -fPIC when compiling the test
3299 dynamically loaded module, to get rid of a warning.
3301 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
3302 message fails, move_frozen_messages = true and ignore_bounce_errors_after
3303 = 0s. The bug is otherwise harmless.
3305 PH/07 There was a bug in the dovecot authenticator such that the value of
3306 $auth1 could be overwritten, and so not correctly preserved, after a
3307 successful authentication. This usually meant that the value preserved by
3308 the server_setid option was incorrect.
3310 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
3312 PH/09 Installed PCRE release 7.0.
3314 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
3315 run for batched SMTP input. It is now run at the start of every message
3316 in the batch. While fixing this I discovered that the process information
3317 (output by running exiwhat) was not always getting set for -bs and -bS
3318 input. This is fixed, and it now also says "batched" for BSMTP.
3320 PH/11 Added control=no_pipelining.
3322 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
3323 patch, slightly modified), and move the expansion of helo_data till after
3324 the connection is made in the smtp transport (so it can use these
3327 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
3329 PH/14 Added log_selector = +pid.
3331 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
3333 PH/16 Add ${if forany and ${if forall.
3335 PH/17 Added dsn_from option to vary the From: line in DSNs.
3337 PH/18 Flush SMTP output before performing a callout, unless control =
3338 no_callout_flush is set.
3340 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
3341 was true (the default) a successful delivery failed to delete the retry
3342 item, thus causing premature timeout of the address. The bug is now
3345 PH/20 Added hosts_avoid_pipelining to the smtp transport.
3347 PH/21 Long custom messages for fakedefer and fakereject are now split up
3348 into multiline responses in the same way that messages for "deny" and
3349 other ACL rejections are.
3351 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
3352 with slight modification.
3354 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
3355 draft, changing the syntax and factoring some duplicate code".
3357 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
3358 for deliveries of the second and subsequent messages over the same SMTP
3361 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
3362 ${reduce, with only minor "tidies".
3364 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
3366 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
3367 expansion side effects.
3369 PH/27 When a message times out after an over-quota error from an Exim-imposed
3370 quota, the bounce message says "mailbox is full". This message was not
3371 being given when it was a system quota that was exceeded. It now should
3374 MH/03 Made $recipients available in local_scan(). local_scan() already has
3375 better access to the recipient list through recipients_list[], but
3376 $recipients can be useful in postmaster-provided expansion strings.
3378 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
3379 in the case of a MAIL command with additional options following the
3380 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
3381 were accidentally chopped off.
3383 PH/29 SMTP synchronization checks are implemented when a command is read -
3384 there is a check that no more input is waiting when there shouldn't be
3385 any. However, for some commands, a delay in an ACL can mean that it is
3386 some time before the response is written. In this time, more input might
3387 arrive, invalidly. So now there are extra checks after an ACL has run for
3388 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
3389 pipelining has not been advertised.
3391 PH/30 MH's patch to allow iscntrl() characters to be list separators.
3393 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
3394 returned in the SMTP response when smtp_return_error_details was false.
3395 This has been fixed.
3397 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
3398 instead of the C I/O that was originally supplied, because problems were
3399 reported on Solaris.
3401 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
3402 Exim which did not show up earlier: it was assuming that a call to
3403 SSL_CTX_set_info_callback() might give an error value. In fact, there is
3404 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
3405 was a macro that became an assignment, so it seemed to work. This has
3406 changed to a proper function call with a void return, hence the compile
3407 error. Exim's code has been fixed.
3409 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
3412 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
3414 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
3416 PH/37 If a message is not accepted after it has had an id assigned (e.g.
3417 because it turns out to be too big or there is a timeout) there is no
3418 "Completed" line in the log. When some messages of this type were
3419 selected by exigrep, they were listed as "not completed". Others were
3420 picked up by some special patterns. I have improved the selection
3421 criteria to be more general.
3423 PH/38 The host_find_failed option in the manualroute router can now be set
3424 to "ignore", to completely ignore a host whose IP address cannot be
3425 found. If all hosts are ignored, the behaviour is controlled by the new
3426 host_all_ignored option.
3428 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
3429 homing or because of multiple MX records with /mx) generated more than
3430 one IP address, and the following item turned out to be the local host,
3431 all the secondary addresses of the first item were incorrectly removed
3432 from the list, along with the local host and any following hosts (which
3433 is what is supposed to happen).
3435 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
3436 whoever called Exim into the -H file. In the case of the daemon it was
3437 behaving confusingly. When first started, it used values for whoever
3438 started the daemon, but after a SIGHUP it used the Exim user (because it
3439 calls itself on a restart). I have changed the code so that it now always
3442 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
3443 message are rejected with the same error (e.g. no authentication or bad
3444 sender address), and a DATA command is nevertheless sent (as can happen
3445 with PIPELINING or a stupid MUA), the error message that was given to the
3446 RCPT commands is included in the rejection of the DATA command. This is
3447 intended to be helpful for MUAs that show only the final error to their
3450 PH/42 Another patch from the Sieve maintainer.
3452 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
3453 Eximstats - Fixed some broken HTML links and added missing column headers
3455 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
3456 columns for Rejects, Temp Rejects, Ham, and Spam rows.
3458 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
3460 PH/43 Yet another patch from the Sieve maintainer.
3462 PH/44 I found a way to check for a TCP/IP connection going away before sending
3463 the response to the final '.' that terminates a message, but only in the
3464 case where the client has not sent further data following the '.'
3465 (unfortunately, this is allowed). However, in many cases there won't be
3466 any further data because there won't be any more messages to send. A call
3467 to select() can be used: if it shows that the input is "ready", there is
3468 either input waiting, or the socket has been closed. An attempt to read
3469 the next input character can distinguish the two cases. Previously, Exim
3470 would have sent an OK response which the client would never have see.
3471 This could lead to message repetition. This fix should cure that, at
3472 least in a lot of common cases.
3474 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
3475 advertised in response to EHLO.
3481 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
3482 fixed by 4.65/MH/01 (is this a record?) are fixed:
3484 (i) An empty string was always treated as zero by the numeric comparison
3485 operators. This behaviour has been restored.
3487 (ii) It is documented that the numeric comparison operators always treat
3488 their arguments as decimal numbers. This was broken in that numbers
3489 starting with 0 were being interpreted as octal.
3491 While fixing these problems I realized that there was another issue that
3492 hadn't been noticed. Values of message_size_limit (both the global option
3493 and the transport option) were treated as octal if they started with 0.
3494 The documentation was vague. These values are now always treated as
3495 decimal, and I will make that clear in the documentation.
3501 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
3502 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
3505 MH/01 Don't check that the operands of numeric comparison operators are
3506 integers when their expansion is in "skipping" mode (fixes bug
3507 introduced by 4.64-PH/07).
3509 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
3510 child addresses, Exim now panics and dies. Previously, because the count
3511 is held in a short int, deliveries were likely to be lost. As such a
3512 large number of recipients for a single message is ridiculous
3513 (performance will be very, very poor), I have chosen to impose a limit
3514 rather than extend the field.
3520 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
3521 leftover -K file (the existence of which was triggered by #402).
3522 While we were at it, introduced process PID as part of the -K
3523 filename. This should rule out race conditions when creating
3526 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
3527 processing considerably. Previous code took too long for large mails,
3528 triggering a timeout which in turn triggers #401.
3530 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
3531 in the DK code in transports.c. sendfile() is not really portable,
3532 hence the _LINUX specificness.
3534 TF/01 In the add_headers option to the mail command in an Exim filter,
3535 there was a bug that Exim would claim a syntax error in any
3536 header after the first one which had an odd number of characters
3539 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
3540 callout verification, Exim cached a "reject" for the entire domain. This
3541 is correct for most verifications, but it is not correct for a recipient
3542 verification with use_sender or use_postmaster set, because in that case
3543 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
3544 case of MAIL FROM:<> rejection from other early rejections (e.g.
3545 rejection of HELO). When verifying a recipient using a non-null MAIL
3546 address, the cache is ignored if it shows MAIL FROM:<> rejection.
3547 Whatever the result of the callout, the value of the domain cache is
3548 left unchanged (for any other kind of callout, getting as far as trying
3549 RCPT means that the domain itself is ok).
3551 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
3554 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
3555 manifest itself as EPIPE rather than ECONNECT. When tidying away a
3556 session, the daemon ignores ECONNECT errors and logs others; it now
3557 ignores EPIPE as well.
3559 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
3560 (quoted-printable decoding).
3562 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
3563 later the small subsequent patch to fix an introduced bug.
3565 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
3567 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
3569 PH/08 An error is now given if message_size_limit is specified negative.
3571 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
3572 to be given (somewhat) arbitrary names.
3574 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
3577 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
3578 miscellaneous code fixes
3580 PH/10 Added the log_reject_target ACL modifier to specify where to log
3583 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
3584 hostname. This is wrong, because it relates to the incoming message (and
3585 probably the interface on which it is arriving) and not to the outgoing
3586 callout (which could be using a different interface). This has been
3587 changed to use the value of the helo_data option from the smtp transport
3588 instead - this is what is used when a message is actually being sent. If
3589 there is no remote transport (possible with a router that sets up host
3590 addresses), $smtp_active_hostname is used.
3592 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
3593 tweaks were necessary in order to get it to work (see also 21 below):
3594 (a) The code assumed that strncpy() returns a negative number on buffer
3595 overflow, which isn't the case. Replaced with Exim's string_format()
3597 (b) There were several signed/unsigned issues. I just did the minimum
3598 hacking in of casts. There is scope for a larger refactoring.
3599 (c) The code used strcasecmp() which is not a standard C function.
3600 Replaced with Exim's strcmpic() function.
3601 (d) The code set only $1; it now sets $auth1 as well.
3602 (e) A simple test gave the error "authentication client didn't specify
3603 service in request". It would seem that Dovecot has changed its
3604 interface. Fortunately there's a specification; I followed it and
3605 changed what the client sends and it appears to be working now.
3607 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
3610 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
3611 address A is aliased to B and C, where B exists and C does not. Without
3612 -v the output is "A verified" because verification stops after a
3613 successful redirection if more than one address is generated. However,
3614 with -v the child addresses are also verified. Exim was outputting "A
3615 failed to verify" and then showing the successful verification for C,
3616 with its parentage. It now outputs "B failed to verify", showing B's
3617 parentage before showing the successful verification of C.
3619 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
3620 look up a TXT record in a specific list after matching in a combined
3623 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
3624 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
3625 they consult the DNS. I had assumed they would set it the way they
3626 wanted; and indeed my experiments on Linux seem to show that in some
3627 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
3628 To be on the safe side, however, I have now made the interface to
3629 host_find_byname() similar to host_find_bydns(), with an argument
3630 containing the DNS resolver options. The host_find_byname() function now
3631 sets these options at its start, just as host_find_bydns() does. The smtp
3632 transport options dns_qualify_single and dns_search_parents are passed to
3633 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
3634 of host_find_byname() use the default settings of RES_DEFNAMES
3635 (qualify_single) but not RES_DNSRCH (search_parents).
3637 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
3638 spool_read_header() do less string testing, by means of a preliminary
3639 switch on the second character of optional "-foo" lines. (This is
3640 overdue, caused by the large number of possibilities that now exist.
3641 Originally there were few.) While I was there, I also converted the
3642 str(n)cmp tests so they don't re-test the leading "-" and the first
3643 character, in the hope this might squeeze out yet more improvement.
3645 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
3646 flag allowing group syntax was set by the header_syntax check but not
3647 turned off, possible causing trouble later; (2) The flag was not being
3648 set at all for the header_verify test, causing "group"-style headers to
3649 be rejected. I have now set it in this case, and also caused header_
3650 verify to ignore an empty address taken from a group. While doing this, I
3651 came across some other cases where the code for allowing group syntax
3652 while scanning a header line wasn't quite right (mostly, not resetting
3653 the flag correctly in the right place). These bugs could have caused
3654 trouble for malformed header lines. I hope it is now all correct.
3656 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
3657 with the "reply" argument non-NULL. The code, however (which originally
3658 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
3659 but it didn't always do it. This confused somebody who was copying the
3660 code for some other use. I have removed all the tests.
3662 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
3663 feature that was used to support insecure browsers during the U.S. crypto
3664 embargo. It requires special client support, and Exim is probably the
3665 only MTA that supported it -- and would never use it because real RSA is
3666 always available. This code has been removed, because it had the bad
3667 effect of slowing Exim down by computing (never used) parameters for the
3668 RSA_EXPORT functionality.
3670 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
3671 authenticator to fail if there's a tab character in the incoming data
3672 (there should never be unless someone is messing about, as it's supposed
3673 to be base64-encoded). Also added, on Timo's advice, the "secured" option
3674 if the connection is using TLS or if the remote IP is the same as the
3675 local IP, and the "valid-client-cert option" if a client certificate has
3678 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
3679 authenticators. This can be used for authorization after authentication
3680 succeeds. (In the case of plaintext, it servers for both authentication
3683 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
3684 if any retry times were supplied.
3686 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
3687 connection, where there is no client IP address to check. In this
3688 situation, the verify now always succeeds.
3690 PH/25 Applied John Jetmore's -Mset patch.
3692 PH/26 Added -bem to be like -Mset, but loading a message from a file.
3694 PH/27 In a string expansion for a processed (not raw) header when multiple
3695 headers of the same name were present, leading whitespace was being
3696 removed from all of them, but trailing whitespace was being removed only
3697 from the last one. Now trailing whitespace is removed from each header
3698 before concatenation. Completely empty headers in a concatenation (as
3699 before) are ignored.
3701 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
3702 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
3704 PH/29 [Removed. This was a change that I later backed out, and forgot to
3705 correct the ChangeLog entry (that I had efficiently created) before
3706 committing the later change.]
3708 PH/30 Exim was sometimes attempting to deliver messages that had suffered
3709 address errors (4xx response to RCPT) over the same connection as other
3710 messages routed to the same hosts. Such deliveries are always "forced",
3711 so retry times are not inspected. This resulted in far too many retries
3712 for the affected addresses. The effect occurred only when there were more
3713 hosts than the hosts_max_try setting in the smtp transport when it had
3714 the 4xx errors. Those hosts that it had tried were not added to the list
3715 of hosts for which the message was waiting, so if all were tried, there
3716 was no problem. Two fixes have been applied:
3718 (i) If there are any address or message errors in an SMTP delivery, none
3719 of the hosts (tried or untried) are now added to the list of hosts
3720 for which the message is waiting, so the message should not be a
3721 candidate for sending over the same connection that was used for a
3722 successful delivery of some other message. This seems entirely
3723 reasonable: after all the message is NOT "waiting for some host".
3724 This is so "obvious" that I'm not sure why it wasn't done
3725 previously. Hope I haven't missed anything, but it can't do any
3726 harm, as the worst effect is to miss an optimization.
3728 (ii) If, despite (i), such a delivery is accidentally attempted, the
3729 routing retry time is respected, so at least it doesn't keep
3730 hammering the server.
3732 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
3733 in ${readsocket because some servers need this prod.
3735 PH/32 Added some extra debug output when updating a wait-xxx database.
3737 PH/33 The hint "could be header name not terminated by colon", which has been
3738 given for certain expansion errors for a long time, was not being given
3739 for the ${if def:h_colon_omitted{... case.
3741 PH/34 The spec says: "With one important exception, whenever a domain list is
3742 being scanned, $domain contains the subject domain." There was at least
3743 one case where this was not true.
3745 PH/35 The error "getsockname() failed: connection reset by peer" was being
3746 written to the panic log as well as the main log, but it isn't really
3747 panic-worthy as it just means the connection died rather early on. I have
3748 removed the panic log writing for the ECONNRESET error when getsockname()
3751 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
3752 runs only) independently of the message's sender address. This meant
3753 that, if the 4xx error was in fact related to the sender, a different
3754 message to the same recipient with a different sender could confuse
3755 things. In particular, this can happen when sending to a greylisting
3756 server, but other circumstances could also provoke similar problems.
3757 I have changed the default so that the retry time for these errors is now
3758 based a combination of the sender and recipient addresses. This change
3759 can be overridden by setting address_retry_include_sender=false in the
3762 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
3763 remote server are returned as part of bounce messages. This was not
3764 happening for LMTP over a pipe (the lmtp transport), but now it is the
3765 same for both kinds of LMTP.
3767 PH/38 Despite being documented as not happening, Exim was rewriting addresses
3768 in header lines that were in fact CNAMEs. This is no longer the case.
3770 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
3771 and queue runs started by the daemon processed all messages. This has
3772 been fixed so that -R and -S can now usefully be given with -q<time>.
3774 PH/40 Import PCRE release 6.7 (fixes some bugs).
3776 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
3778 PH/42 Give an error if -q is specified more than once.
3780 PH/43 Renamed the variables $interface_address and $interface_port as
3781 $received_ip_address and $received_port, to make it clear that these
3782 values apply to message reception, and not to the outgoing interface when
3783 a message is delivered. (The old names remain recognized, of course.)
3785 PH/44 There was no timeout on the connect() call when using a Unix domain
3786 socket in the ${readsocket expansion. There now is.
3788 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
3789 be meaningful with "accept".
3791 SC/01 Eximstats V1.43
3792 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
3794 SC/02 Eximstats V1.44
3795 Use a glob alias rather than an array ref in the generated
3796 parser. This improves both readability and performance.
3798 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
3799 Collect SpamAssassin and rejection statistics.
3800 Don't display local sender or destination tables unless
3801 there is data to show.
3802 Added average volumes into the top table text output.
3804 SC/04 Eximstats V1.46
3805 Collect data on the number of addresses (recipients)
3806 as well as the number of messages.
3808 SC/05 Eximstats V1.47
3809 Added 'Message too big' to the list of mail rejection
3810 reasons (thanks to Marco Gaiarin).
3812 SC/06 Eximstats V1.48
3813 Mainlog lines which have GMT offsets and are too short to
3814 have a flag are now skipped.
3816 SC/07 Eximstats V1.49 (Alain Williams)
3817 Added the -emptyok flag.
3819 SC/08 Eximstats V1.50
3820 Fixes for obtaining the IP address from reject messages.
3822 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
3823 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
3824 whitespace changes from 4.64-PH/27
3826 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
3829 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
3830 are found, allow negative numbers in numeric criteria)
3832 JJ/06 exipick.20061117.2, added new $message_body_missing variable
3834 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
3835 to match changes made in 4.64-PH/43
3837 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
3839 PH/47 Put in an explicit test for a DNS lookup of an address record where the
3840 "domain" is actually an IP address, and force a failure. This locks out
3841 those revolvers/nameservers that support "A-for-A" lookups, in
3842 contravention of the specifications.
3844 PH/48 When a host name was looked up from an IP address, and the subsequent
3845 forward lookup of the name timed out, the host name was left in
3846 $sender_host_name, contrary to the specification.
3848 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3849 restricted to single-key lookups, Exim was not diagnosing an error if
3850 * or *@ was used with a query-style lookup.
3852 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3854 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3855 long ago, but noone interested enough thought of it. Let's just say that
3856 the "1.1" means that there are some new functions that weren't there at
3857 some point in the past.
3859 PH/51 Error processing for expansion failure of helo_data from an smtp
3860 transport during callout processing was broken.
3862 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3863 tested/used via the -bh/-bhc/-bs options.
3865 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3866 bug, fixed in subsequent PCRE releases).
3868 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3869 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3871 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3877 SC/01 Use a glob alias rather than an array ref in eximstats generated
3878 parser. This improves both readability and performance.
3880 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3881 Don't display local sender or destination tables in eximstats unless
3882 there is data to show.
3883 Added average volumes into the eximstats top table text output.
3885 SC/03 Collect data on the number of addresses (recipients) as well
3886 as the number of messages in eximstats.
3888 TF/01 Correct an error in the documentation for the redirect router. Exim
3889 does (usually) call initgroups() when daemonizing.
3891 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3892 with consistent privilege compared to when running as a daemon.
3894 TF/03 Note in the spec that $authenticated_id is not set for local
3895 submissions from trusted users.
3897 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3898 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3900 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3901 by adding some example configuration directives to the default
3902 configuration file. A little bit of work is required to uncomment the
3903 directives and define how usernames and passwords are checked, but
3904 there is now a framework to start from.
3906 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3907 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3908 without this. I don't know how relevant this is to other LDAP libraries.
3910 PH/02 Add the verb name to the "unknown ACL verb" error.
3912 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3914 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3916 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3917 directory not expanded when it should be if an expanded home directory
3918 was set for the address (which is overridden by the transport).
3920 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3923 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3924 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3925 because it is too late at that time, and has no effect.
3927 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3928 security issue with \' (bugzilla #107). I could not use the
3929 PQescapeStringConn() function, because it needs a PGconn value as one of
3932 PH/08 When testing addresses using -bt, indicate those final addresses that
3933 are duplicates that would not cause an additional delivery. At least one
3934 person was confused, thinking that -bt output corresponded to deliveries.
3935 (Suppressing duplicates isn't a good idea as you lose the information
3936 about possibly different redirections that led to the duplicates.)
3938 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3939 systems where poll() doesn't work, in particular OS X.
3941 PH/10 Added more information to debugging output for retry time not reached.
3943 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3944 operations in malware.c.
3946 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3949 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3950 syslog_duplication set false, log lines that would normally be written
3951 both the the main log and to the reject log were not written to syslog at
3954 PH/14 In the default configuration, change the use of "message" in ACL warn
3955 statements to "add_header".
3957 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3958 not followed by a command (e.g. "seen endif").
3960 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3961 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3964 PH/17 Added extra conditions to the default value of delay_warning_condition
3968 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3969 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3970 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3973 The Auto-Submitted: and various List- headers are standardised, whereas I
3974 don't think Precedence: ever was.
3976 PH/18 Refactored debugging code in route_finduser() to show more information,
3977 in particular, the error code if getpwnam() issues one.
3979 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3980 This is apparently needed in addition to the PH/07 change above to avoid
3981 any possible encoding problems.
3983 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3984 but not after initializing Perl.
3986 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3987 output them only if debugging. By default they are written stderr,
3988 apparently, which is not desirable.
3990 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3993 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3996 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3998 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3999 authenticated or an ident call has been made. Suppress the default
4000 values for $authenticated_id and $authenticated_sender (but permit -oMai
4001 and -oMas) when testing with -bh.
4003 PH/24 Re-jigged the order of the tests in the default configuration so that the
4004 tests for valid domains and recipients precede the DNS black list and CSA
4005 tests, on the grounds that those ones are more expensive.
4007 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
4008 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
4009 This bug exists in every version of Exim that I still have, right back to
4012 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
4013 However, an attempt to turn on case-sensitivity in a regex key by
4014 including (?-i) didn't work because the subject string was already
4015 lowercased, and the effects were non-intuitive. It turns out that a
4016 one-line patch can be used to allow (?-i) to work as expected.
4022 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
4023 other effects) broke the use of negated acl sub-conditions.
4025 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
4028 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
4029 "Deny" causes Exim to reject the incoming connection with a 554 error.
4030 Unfortunately, if there is a major crisis, such as a disk failure,
4031 tcp-wrappers gives "deny", whereas what one would like would be some
4032 kind of temporary error. A kludge has been added to help with this.
4033 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
4034 554 error is used if errno is still zero or contains ENOENT (which occurs
4035 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
4038 PH/03 Add -lutil to the default FreeBSD LIBS setting.
4040 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
4041 errors. Otherwise a message that provokes a temporary error (when other
4042 messages do not) can cause a whole host to time out.
4044 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
4045 addresses were routed directly to files or pipes from a redirect router.
4046 File deliveries just didn't batch; pipe deliveries might have suffered
4049 PH/06 A failure to get a lock for a hints database would erroneously always say
4050 "Failed to get write lock", even when it was really a read lock.
4052 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
4053 of 0600. This has been changed to use the value of the lockfile_mode
4054 option (which defaults to 0600).
4056 PH/08 Applied small patch from the Sieve maintainer.
4058 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
4059 folder from quota calculations, a direct delivery into this folder messed
4060 up the contents of the maildirsize file. This was because the regex was
4061 used only to exclude .Trash (or whatever) when the size of the mailbox
4062 was calculated. There was no check that a delivery was happening into an
4063 excluded directory. This bug has been fixed by ignoring all quota
4064 processing for deliveries into excluded directories.
4066 PH/10 Added the maildirfolder_create_regex option to appendfile.
4072 PH/01 The code for finding all the local interface addresses on a FreeBSD
4073 system running IPv6 was broken. This may well have applied to all BSD
4074 systems, as well as to others that have similar system calls. The broken
4075 code found IPv4 interfaces correctly, but gave incorrect values for the
4076 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
4077 that it would not match correctly against @[] and not recognize the IPv6
4080 PH/02 The ipliteral router was not recognizing addresses of the form user@
4081 [ipv6:....] because it didn't know about the "ipv6:" prefix.
4083 PH/03 Added disable_ipv6.
4085 PH/04 Changed $reply_address to use the raw form of the headers instead of the
4086 decoded form, because it is most often used to construct To: headers
4087 lines in autoreplies, and the decoded form may well be syntactically
4088 invalid. However, $reply_address has leading white space removed, and all
4089 newlines turned into spaces so that the autoreply transport does not
4092 PH/05 If group was specified without a user on a router, and no group or user
4093 was specified on a transport, the group from the router was ignored.
4095 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
4096 for visible compile-time settings that can be used to change these
4097 numbers, for those that want even more. Backwards compatibility with old
4098 spool files has been maintained. However, going back to a previous Exim
4099 release will lost any variables that are in spool files.
4101 PH/07 Two small changes when running in the test harness: increase delay when
4102 passing a TCP/IP connection to a new process, in case the original
4103 process has to generate a bounce, and remove special handling of
4104 127.0.0.2 (sic), which is no longer necessary.
4106 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
4107 be the same on different OS.
4109 PH/09 Moved a debug statement in filter processing to avoid a race problem when
4112 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
4113 whether --show-vars was specified or not
4115 JJ/02 exipick: Added support for new ACL variable spool format introduced
4118 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
4119 syntactically invalid From: or Reply-to: line, and a filter used this to
4120 generate an autoreply, and therefore failed to obtain an address for the
4121 autoreply, Exim could try to deliver to a non-existent relative file
4122 name, causing unrelated and misleading errors. What now happens is that
4123 it logs this as a hard delivery error, but does not attempt to create a
4126 PH/11 The exinext utility has a -C option for testing purposes, but although
4127 the given file was scanned by exinext itself; it wasn't being passed on
4128 when Exim was called.
4130 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
4131 an end-of-file indication when reading a command response.
4133 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
4134 compiled. In many other places in Exim, IPv6 addresses are always
4135 recognized, so I have changed this. It also means that IPv4 domain
4136 literals of the form [IPV4:n.n.n.n] are now always recognized.
4138 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
4139 used if the router is not running as root, for example, when verifying at
4140 ACL time, or when using -bh. The debugging output from this situation was
4141 non-existent - all you got was a failure to exec. I have made two
4144 (a) Failures to set uid/gid, the current directory, or a process leader
4145 in a subprocess such as that created by queryprogram now generate
4146 suitable debugging output when -d is set.
4148 (b) The queryprogram router detects when it is not running as root,
4149 outputs suitable debugging information if -d is set, and then runs
4150 the subprocess without attempting to change uid/gid.
4152 PH/15 Minor change to Makefile for building test_host (undocumented testing
4155 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
4156 additional section of a DNS packet that returns MX or SRV records.
4157 Instead, it always explicitly searches for A/AAAA records. This avoids
4158 major problems that occur when a DNS server includes only records of one
4159 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
4160 fixed another bug: if SRV records were looked up and the corresponding
4161 address records were *not* found in the additional section, the port
4162 values from the SRV records were lost.
4164 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
4165 using the correct key (the original address) when searching the retry
4166 rules in order to find which one to use for generating the retry hint.
4168 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
4169 adding the default one. Similarly, if it contains a Reply-To: header, the
4170 errors_reply_to option, if set, is not used.
4172 PH/19 When calculating a retry time, Exim used to measure the "time since
4173 failure" by looking at the "first failed" field in the retry record. Now
4174 it does not use this if it is later than than the arrival time of the
4175 message. Instead it uses the arrival time. This makes for better
4176 behaviour in cases where some deliveries succeed, thus re-setting the
4177 "first failed" field. An example is a quota failure for a huge message
4178 when small messages continue to be delivered. Without this change, the
4179 "time since failure" will always be short, possible causing more frequent
4180 delivery attempts for the huge message than are intended.
4181 [Note: This change was subsequently modified - see PH/04 for 4.62.]
4183 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
4184 $1, $2, $3) because the numerical variables can be reset during some
4185 expansion items (e.g. "match"), thereby losing the authentication data.
4187 PH/21 Make -bV show the size of off_t variables so that the test suite can
4188 decide whether to run tests for quotas > 2G.
4190 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
4191 mailbox_size, and mailbox_filecount in the appendfile transport. If a
4192 filecount value is greater than 2G or if a quota value is greater than 2G
4193 on a system where the size of off_t is not greater than 4, a panic error
4196 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
4197 never match. The debug and -bh output now contains an explicit error
4198 message indicating a malformed IPv4 address or mask.
4200 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
4201 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
4202 PH/23 above applies.
4204 PH/25 Do not write to syslog when running in the test harness. The only
4205 occasion when this arises is a failure to open the main or panic logs
4206 (for which there is an explicit test).
4208 PH/26 Added the /no_tell option to "control=freeze".
4210 PH/27 If a host name lookup failed very early in a connection, for example, if
4211 the IP address matched host_lookup and the reverse lookup yielded a name
4212 that did not have a forward lookup, an error message of the form "no IP
4213 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
4214 could be logged. Now it outputs the IP address instead of "NULL".
4216 PH/28 An enabling patch from MH: add new function child_open_exim2() which
4217 allows the sender and the authenticated sender to be set when
4218 submitting a message from within Exim. Since child_open_exim() is
4219 documented for local_scan(), the new function should be too.
4221 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
4222 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
4223 results in an empty string is now treated as unset.
4225 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
4227 PH/31 Added sender_verify_fail logging option.
4229 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
4230 needed by RFC 822 but not by RFC 2822 was commented out. I have now
4231 tidied the source and removed it altogether.
4233 PH/33 When a queue run was abandoned because the load average was too high, a
4234 log line was always written; now it is written only if the queue_run log
4235 selector is set. In addition, the log line for abandonment now contains
4236 information about the queue run such as the pid. This is always present
4237 in "start" and "stop" lines but was omitted from the "abandon" line.
4239 PH/34 Omit spaces between a header name and the colon in the error message that
4240 is given when verify = headers_syntax fails (if there are lots of them,
4241 the message gets confusing).
4243 PH/35 Change the default for dns_check_names_pattern to allow slashes within
4244 names, as there are now some PTR records that contain slashes. This check
4245 is only to protect against broken name servers that fall over on strange
4246 characters, so the fact that it applies to all lookups doesn't matter.
4248 PH/36 Now that the new test suite is complete, we can remove some of the
4249 special code in Exim that was needed for the old test suite. For example,
4250 sorting DNS records because real resolvers return them in an arbitrary
4251 order. The new test suite's fake resolver always returns records in the
4254 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
4255 bounces) except when queue_only is set, to avoid logging races between
4256 the different processes.
4258 PH/38 Panic-die if .include specifies a non-absolute path.
4260 PH/39 A tweak to the "H" retry rule from its user.
4262 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
4263 a label. They prevented compilation on older perls.
4265 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
4266 a warning to be raised on newish perls.
4268 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
4269 on queue. Changes to match documented behaviour of showing count of
4270 messages matching specified criteria.
4272 PH/40 Changed the default ident timeout from 30s to 5s.
4274 PH/41 Added support for the use of login_cap features, on those BSD systems
4275 that have them, for controlling the resources used by pipe deliveries.
4277 PH/42 The content-scanning code uses fopen() to create files in which to put
4278 message data. Previously it was not paying any attention to the mode of
4279 the files. Exim runs with umask(0) because the rest of the code creates
4280 files with open(), and sets the required mode explicitly. Thus, these
4281 files were ending up world-writeable. This was not a big issue, because,
4282 being within the spool directory, they were not world-accessible. I have
4283 created a function called modefopen, which takes an additional mode
4284 argument. It sets umask(777), creates the file, chmods it to the required
4285 mode, then resets the umask. All the relevant calls to fopen() in the
4286 content scanning code have been changed to use this function.
4288 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
4289 to 24 hours. This avoids potential overflow problems when processing G
4290 and H retry rules. I suspect nobody ever tinkers with this value.
4292 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
4294 PH/45 When the plaintext authenticator is running as a client, the server's
4295 challenges are checked to ensure they are valid base64 strings. By
4296 default, the authentication attempt is cancelled if an invalid string is
4297 received. Setting client_ignore_invalid_base64 true ignores these errors.
4298 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
4299 they are received. Thus, the responses can be made to depend on the
4300 challenges. If an invalid string is ignored, an empty string is placed in
4303 PH/46 Messages that are created by the autoreply transport now contains a
4304 References: header, in accordance with RFCs 2822 and 3834.
4306 PH/47 Added authenticated_sender_force to the smtp transport.
4308 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
4310 PH/49 Installed latest patch from the Sieve maintainer.
4312 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
4313 was also set, the appendfile transport was unnecessarily scanning a
4314 directory of message files (e.g. for maildir delivery) to find the count
4315 of files (along with the size), even though it did not need this
4316 information. It now does the scan only if it needs to find either the
4317 size of the count of files.
4319 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
4321 PH/52 Two bugs concerned with error handling when the smtp transport is
4324 (i) Exim was not creating retry information for temporary errors given
4325 for individual recipients after the DATA command when the smtp transport
4326 was used in LMTP mode. This meant that they could be retried too
4327 frequently, and not timed out correctly.
4329 (ii) Exim was setting the flag that allows error details to be returned
4330 for LMTP errors on RCPT commands, but not for LMTP errors for individual
4331 recipients that were returned after the DATA command.
4333 PH/53 This is related to PH/52, but is more general: for any failing address,
4334 when detailed error information was permitted to be returned to the
4335 sender, but the error was temporary, then after the final timeout, only
4336 "retry timeout exceeded" was returned. Now it returns the full error as
4337 well as "retry timeout exceeded".
4339 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
4340 do this, and (what is worse) MTAs that accept it.
4342 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
4343 will now be deprecated.
4345 PH/56 New os.c-cygwin from the Cygwin maintainer.
4347 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
4348 formats (previously only available in exim formats via -bpr, -bpru,
4349 and -bpra. Now also available in native and exiqgrep formats)
4351 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
4352 with very large, slow to parse queues
4354 JJ/08 exipick: added ! as generic prefix to negate any criteria format
4356 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
4358 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
4359 responses to authentication challenges, though it was showing the
4360 challenges; (ii) I've removed the CR characters from the debug output for
4363 PH/58 Allow for the insertion of a newline as well as a space when a string
4364 is turned into more than one encoded-word during RFC 2047 encoding. The
4365 Sieve code now uses this.
4367 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
4368 data_4xx, lost_connection, tls_required.
4370 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
4371 message was being sent as an SMTP response.
4373 PH/61 Add -l and -k options to exicyclog.
4375 PH/62 When verifying, if an address was redirected to one new address, so that
4376 verification continued, and the new address failed or deferred after
4377 having set something in $address_data, the value of $address_data was not
4378 passed back to the ACL. This was different to the case when no
4379 redirection occurred. The value is now passed back in both cases.
4381 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
4382 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
4383 use login_cap.h, so on its own it isn't the distinguishing feature. The
4384 new name refers directly to the setclassresources() function.
4386 PH/65 Added configuration files for NetBSD3.
4388 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
4390 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
4391 is preferred over IPv4.
4393 PH/68 The bounce_return_message and bounce_return_body options were not being
4394 honoured for bounces generated during the reception of non-SMTP messages.
4395 In particular, this applied to messages rejected by the ACL. This bug has
4396 been fixed. However, if bounce_return_message is true and bounce_return_
4397 body is false, the headers that are returned for a non-SMTP message
4398 include only those that have been read before the error was detected.
4399 (In the case of an ACL rejection, they have all been read.)
4401 PH/69 The HTML version of the specification is now built in a directory called
4402 spec_html instead of spec.html, because the latter looks like a path with
4403 a MIME-type, and this confuses some software.
4405 PH/70 Catch two compiler warnings in sieve.c.
4407 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
4408 function verify_get_ident() calls ip_connect() to connect a socket, but
4409 if the "connect()" function timed out, ip_connect() used to close the
4410 socket. However, verify_get_ident() also closes the socket later, and in
4411 between Exim writes to the log, which may get opened at this point. When
4412 the socket was closed in ip_connect(), the log could get the same file
4413 descriptor number as the socket. This naturally causes chaos. The fix is
4414 not to close the socket in ip_connect(); the socket should be closed by
4415 the function that creates it. There was only one place in the code where
4416 this was missing, in the iplookup router, which I don't think anybody now
4417 uses, but I've fixed it anyway.
4419 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
4420 well as to direct DNS lookups. Otherwise the handling of names in host
4421 lists is inconsistent and therefore confusing.
4427 PH/01 Two changes to the default runtime configuration:
4429 (1) Move the checks for relay_from_hosts and authenticated clients from
4430 after to before the (commented out) DNS black list checks.
4432 (2) Add control=submission to the relay_from_hosts and authenticated
4433 clients checks, on the grounds that messages accepted by these
4434 statements are most likely to be submissions.
4436 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
4438 (1) Generate an error if the third argument for the ${prvs expansion is
4441 (2) Treat a missing third argument of ${prvscheck as if it were an empty
4444 (3) Reset the variables that are obtained from the first argument of
4445 ${prvscheck and used in the second argument before leaving the code,
4446 because their memory is reclaimed, so using them afterwards may do
4449 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
4450 one (it's much easier than Tom thought :-).
4452 (5) Because of (4), we can now allow for the use of $prvscheck_result
4453 inside the third argument.
4455 PH/03 For some reason, the default setting of PATH when running a command from
4456 a pipe transport was just "/usr/bin". I have changed it to
4459 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
4460 anything to be listed in the output from -bV.
4462 PH/05 When a filter generated an autoreply, the entire To: header line was
4463 quoted in the delivery log line, like this:
4465 => >A.N.Other <ano@some.domain> <original@ddress> ...
4467 This has been changed so that it extracts the operative address. There
4468 may be more than one such address. If so, they are comma-separated, like
4471 => >ano@some.domain,ona@other.domain <original@ddress> ...
4473 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
4474 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
4475 not being looked up in the rDNS to get a host name, Exim was showing the
4476 IP address twice in Received: lines, even though the IP addresses were
4477 identical. For example:
4479 Received: from [1.2.3.4] (helo=[1.2.3.4])
4481 However, if the real host name was known, it was omitting the HELO data
4482 if it matched the actual IP address. This has been tidied up so that it
4483 doesn't show the same IP address twice.
4485 PH/07 When both +timestamp and +memory debugging was on, the value given by
4486 $tod_xxx expansions could be wrong, because the tod_stamp() function was
4487 called by the debug printing, thereby overwriting the timestamp buffer.
4488 Debugging no longer uses the tod_stamp() function when +timestamp is set.
4490 PH/08 When the original message was included in an autoreply transport, it
4491 always said "this is a copy of the message, including all the headers",
4492 even if body_only or headers_only was set. It now gives an appropriate
4495 PH/09 Applied a patch from the Sieve maintainer which:
4497 o fixes some comments
4498 o adds the (disabled) notify extension core
4499 o adds some debug output for the result of if/elsif tests
4500 o points to the current vacation draft in the documentation
4501 and documents the missing references header update
4505 o fixes a bug in processing the envelope test (when testing
4506 multiple envelope elements, the last element determined the
4509 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
4510 Electronic Mail") by including:
4512 Auto-submitted: auto-generated
4514 in the messages that it generates (bounce messages and others, such as
4515 warnings). In the case of bounce messages for non-SMTP messages, there was
4516 also a typo: it was using "Auto_submitted" (underscore instead of
4517 hyphen). Since every message generated by Exim is necessarily in response
4518 to another message, thes have all been changed to:
4520 Auto-Submitted: auto-replied
4522 in accordance with these statements in the RFC:
4524 The auto-replied keyword:
4526 - SHOULD be used on messages sent in direct response to another
4527 message by an automatic process,
4529 - MUST NOT be used on manually-generated messages,
4531 - MAY be used on Delivery Status Notifications (DSNs) and Message
4532 Disposition Notifications (MDNs),
4534 - MUST NOT be used on messages generated by automatic or periodic
4535 processes, except for messages which are automatic responses to
4538 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
4539 to the default Received: header definition.
4541 PH/12 Added log selector acl_warn_skipped (default on).
4543 PH/13 After a successful wildlsearch lookup, discard the values of numeric
4544 variables because (a) they are in the wrong storage pool and (b) even if
4545 they were copied, it wouldn't work properly because of the caching.
4547 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
4548 checking when decoding. Apparently there are clients that generate
4549 overlong encoded strings. Why am I not surprised?
4551 PH/15 If the first argument of "${if match_address" was not empty, but did not
4552 contain an "@" character, Exim crashed. Now it writes a panic log message
4553 and treats the condition as false.
4555 PH/16 In autoreply, treat an empty string for "once" the same as unset.
4557 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
4558 extension "envelope-auth". The code is finished and in agreement with
4559 other implementations, but there is no documentation so far and in fact,
4560 nobody wrote the draft yet. This extension is currently #undef'ed, thus
4561 not changing the active code.
4563 Print executed "if" and "elsif" statements when debugging is used. This
4564 helps a great deal to understand what a filter does.
4566 Document more things not specified clearly in RFC3028. I had all this
4567 sorted out, when out of a sudden new issues came to my mind. Oops."
4569 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
4572 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
4573 canonical form (as documented). However, after a host name lookup from
4574 the IP address, check_host() was doing a simple string comparison with
4575 addresses acquired from the DNS when checking that the found name did
4576 have the original IP as one of its addresses. Since any found IPv6
4577 addresses are likely to be in abbreviated form, the comparison could
4578 fail. Luckily, there already exists a function for doing the comparison
4579 by converting both addresses to binary, so now that is used instead of
4580 the text comparison.
4582 PH/20 There was another similar case to PH/19, when a complete host name was
4583 given in a host list; looking up its IP address could give an abbreviated
4584 form, whereas the current host's name might or might not be abbreviated.
4585 The same fix has been applied.
4591 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
4592 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
4595 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
4596 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
4598 PH/03 Typo: missing ".o" in src/pcre/Makefile.
4600 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
4601 header line, restrict the check to what is listed in RFCs 2369 and 2929.
4602 Also, for "Auto-Submitted", treat anything other than "no" as
4603 non-personal, in accordance with RFC 3834. (Previously it treated
4604 anything starting "auto-" as non-personal.)
4606 TF/01 The control=submission/name=... option had a problem with syntax
4607 errors if the name included a slash character. The /name= option
4608 now slurps the rest of the string, so it can include any characters
4609 but it must come last in the list of options (after /sender_retain
4612 PH/05 Some modifications to the interface to the fake nameserver for the new
4620 TK/01 Added the "success_on_redirect" address verification option. See
4621 NewStuff for rationale and an example.
4623 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
4625 PH/02 Patch to exigrep to allow it to work on syslog lines.
4627 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
4628 fread() to skip over the body file's header line, because in Cygwin the
4629 header line is locked and is inaccessible.
4631 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
4632 co-exist for some time) to make it clear that it is the Exim ID that is
4633 referenced, not the Message-ID: header line.
4635 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
4636 string_format() function, because snprintf() does not exist on all
4639 PH/06 The use of forbid_filter_existstest now also locks out the use of the
4640 ${stat: expansion item.
4642 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
4643 protocol synchronization error", to keep the pedants happy.
4645 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
4646 well as for IRIX systems, when gcc is being used. See the host.c source
4649 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
4651 PH/10 Named domain lists were not working if used in a queue_smtp_domains
4654 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
4655 transport and to the smtp transport in LMTP mode.
4657 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
4659 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
4660 run a filter in a subprocess. This could lead to confusion in subsequent
4661 lookups in the parent process. There should also be a search_tidyup() at
4662 the end of the subprocess.
4664 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
4665 only if the host matched helo_try_verify_hosts, which caused the
4666 verification to occur when the EHLO/HELO command was issued. The ACL just
4667 tested the remembered result. Now, if a previous verification attempt has
4668 not happened, "verify = helo" does it there and then.
4670 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
4672 TK/03 Fix log output including CR from clamd.
4674 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
4675 exist provoked a memory error which could cause a segfault.
4677 PH/15 Installed PCRE 6.2
4679 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
4681 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
4682 of the problem. Specifically, suggested +O2 rather than +O1 for the
4685 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
4687 PH/20 If a delivery was routed to a non-standard port by means of an SRV
4688 record, the port was not correctly logged when the outgoing_port log
4689 selector was set (it logged the transort's default port).
4691 PH/21 Added support for host-specific ports to manualroute, queryprogram,
4692 fallback_hosts, and "hosts" in the smtp transport.
4694 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
4695 host errors such as "Connection refused".
4697 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
4698 authentication with radiusclient 0.4.9:
4700 - Error returned from rc_read_config was caught wrongly
4701 - Username/password not passed on to radius server due to wrong length.
4703 The presumption is that some radiusclient API changes for 4.51/PH/17
4704 were not taken care of correctly. The code is still untested by me (my
4705 Linux distribution still has 0.3.2 of radiusclient), but it was
4706 contributed by a Radius user.
4708 PH/24 When doing a callout, the value of $domain wasn't set correctly when
4709 expanding the "port" option of the smtp transport.
4711 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
4712 while reading a MIME header. Thanks to Tom Hughes for a patch.
4714 PH/24 Include config.h inside local_scan.h so that configuration settings are
4717 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
4718 that in an ACL for RCPT (for example), you can examine exactly what was
4721 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
4722 commands, but it was not correctly comparing the address with the actual
4723 client host address. Thus, it would show the EHLO address in Received:
4724 header lines when this was not necessary.
4726 PH/27 Added the % operator to ${eval:}.
4728 PH/28 Exim tries to create and chdir to its spool directory when it starts;
4729 it should be ignoring failures (because with -C, for example, it has lost
4730 privilege). It wasn't ignoring creation failures other than "already
4733 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
4736 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
4737 because an input line was too long, either on its own, or by virtue of
4738 too many continuations, the temporary file was not being removed, and the
4739 return code was incorrect.
4741 PH/31 Missing "BOOL" in function definition in filtertest.c.
4743 PH/32 Applied Sieve patches from the maintainer.
4745 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
4747 PH/33 Added "verify = not_blind".
4749 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
4750 Local/Makefile (with some defaults set). These are used in built scripts
4751 such as exicyclog, but they have never been used in the exim_install
4752 script (though there are many overriding facilities there). I have
4753 arranged that the exim_install script now takes note of these two
4756 PH/35 Installed configuration files for Dragonfly.
4758 PH/36 When a locally submitted message by a trusted user did not contain a
4759 From: header, and the sender address was obtained from -f or from an SMTP
4760 MAIL command, and the trusted user did not use -F to supply a sender
4761 name, $originator_name was incorrectly used when constructing a From:
4762 header. Furthermore, $originator_name was used for submission mode
4763 messages from external hosts without From: headers in a similar way,
4764 which is clearly wrong.
4766 PH/37 Added control=suppress_local_fixups.
4768 PH/38 When log_selector = +received_sender was set, and the addition of the
4769 sender made the log line's construction buffer exactly full, or one byte
4770 less than full, an overflow happened when the terminating "\n" was
4773 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
4774 when the result of a list match is failure because a DNS lookup failed.
4776 PH/40 RM_COMMAND is now used in the building process.
4778 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
4779 the "build-* directories that it finds.
4781 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
4782 address in a domain literal was a prefix of an interface address.
4784 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
4785 when verifying a sender address, unless rewrite_headers is false.
4787 PH/44 Wrote a long comment about why errors_to addresses are verified as
4788 recipients, not senders.
4790 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
4791 the ratelimit ACL was added.
4793 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
4795 PH/46 Added extra information about PostgreSQL errors to the error string.
4797 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
4798 avoiding the need to install special zones in a real server. This is
4799 backwards compatible; if it can't find the fake resolver, it drops back.
4800 Thus, both old and new test suites can be run.
4802 TF/02 Added util/ratelimit.pl
4804 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
4805 clock is set back in time.
4807 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
4808 Candler <B.Candler@pobox.com>.
4810 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
4811 OK for addresses that are the result of redirections.
4813 PH/48 A number of further additions for the benefit of the new test suite,
4814 including a fake gethostbyname() that interfaces to the fake DNS resolver
4817 TF/06 The fix for widen_domains has also been applied to qualify_single and
4818 search_parents which are the other dnslookup options that can cause
4821 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
4824 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
4826 TF/07 Exim produced the error message "an SRV record indicated no SMTP
4827 service" if it encountered an MX record with an empty target hostname.
4828 The message is now "an MX or SRV record indicated no SMTP service".
4830 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
4831 if the DNS of the sending site is misconfigured. This is quite a
4832 common situation. This change restores the behaviour of treating a
4833 helo verification defer as a failure.
4835 PH/51 If self=fail was set on a router, the bounce message did not include the
4836 actual error message.
4842 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
4844 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
4845 command itself ended in error, the underlying message about the transport
4846 filter timeout was being overwritten with the pipe command error. Now the
4847 underlying error message should be appended to the second error message.
4849 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4851 PH/02 Reduce the amount of output that "make" produces by default. Full output
4852 can still be requested.
4854 PH/03 The warning log line about a condition test deferring for a "warn" verb
4855 was being output only once per connection, rather than after each
4856 occurrence (because it was using the same function as for successful
4857 "warn" verbs). This seems wrong, so I have changed it.
4859 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4860 it should not have, which might have caused a crash in the right
4861 circumstances, but probably never did.
4863 PH/04 Installed a modified version of Tony Finch's patch to make submission
4864 mode fix the return path as well as the Sender: header line, and to
4865 add a /name= option so that you can make the user's friendly name appear
4868 TF/03 Added the control = fakedefer ACL modifier.
4870 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4871 Mark Lowes for thorough testing.
4873 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4875 TK/03 Merged latest SRS patch from Miles Wilton.
4877 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4878 with the definition in sysexits.h (which is #included earlier).
4879 Fortunately, Exim does not actually use EX_OK. The code used to try to
4880 preserve the sysexits.h value, by assuming that macro definitions were
4881 scanned for macro replacements. I have been disabused of this notion,
4882 so now the code just undefines EX_OK before #including unistd.h.
4884 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4885 in the smtp transport. When a block could not be written in a single
4886 write() function, the timeout was being re-applied to each part-write.
4887 This seems wrong - if the receiver was accepting one byte at a time it
4888 would take for ever. The timeout is now adjusted when this happens. It
4889 doesn't have to be particularly precise.
4891 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4892 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4894 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4895 without a domain if the check to <postmaster@domain> fails.
4897 SC/01 Eximstats: added -xls and the ability to specify output files
4898 (patch written by Frank Heydlauf).
4900 SC/02 Eximstats: use FileHandles for outputting results.
4902 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4904 SC/04 Eximstats: fixed display of large numbers with -nvr option
4906 SC/05 Eximstats: fixed merging of reports with empty tables.
4908 SC/06 Eximstats: added the -include_original_destination flag
4910 SC/07 Eximstats: removed tabs and trailing whitespace.
4912 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4914 TK/06 MBOX spool code: Add real "From " MBOX separator line
4915 so the .eml file is really in mbox format (even though
4916 most programs do not really care). Patch from Alex Miller.
4918 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4919 The latter is generated from $received_to and is only set if the
4920 message has one envelope recipient. SA can use these headers,
4921 obviously out-of-the-box. Patch from Alex Miller.
4923 PH/08 The ${def test on a variable was returning false if the variable's
4924 value was "0", contrary to what the specification has always said!
4925 The result should be true unless the variable is empty.
4927 PH/09 The syntax error of a character other than { following "${if
4928 def:variable_name" (after optional whitespace) was not being diagnosed.
4929 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4930 accidental colon was present, for example, could give incorrect results.
4932 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4933 result is used (not including appendfile, where other changes are about
4936 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4937 This involved changing a lot of size variables from int to off_t. It
4938 should work with maildirs and everything.
4940 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4941 spamd dying while we are connected to it.
4943 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4946 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4947 function for BDB 4.3.
4949 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4951 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4952 the sources for PCRE have been re-arranged and more files are now
4955 PH/15 The code I had for printing potentially long long variables in PH/11
4956 above was not the best (it lost precision). The length of off_t variables
4957 is now inspected at build time, and an appropriate printing format (%ld
4958 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4959 to be "long long int" or "long int". This is needed for the internal
4960 formatting function string_vformat().
4962 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4963 the configuration file to be ":syslog", then the script "guesses" where
4964 the logs files are, rather than using the compiled in default. In our
4965 case the guess is not the same as the compiled default, so the script
4966 suddenly stopped working when I started to use syslog. The patch checks
4967 to see if log_file_path is "". If so, it attempts to read it from exim
4968 with no configuration file to get the compiled in version, before it
4969 falls back to the previous guessing code."
4971 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4972 implementing BATV in an Exim configuration. See NewStuff for the gory
4975 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4976 Makefile that are specific to HP-UX.
4978 PH/18 If the "use_postmaster" option was set for a recipient callout together
4979 with the "random" option, the postmaster address was used as the MAIL
4980 FROM address for the random test, but not for the subsequent recipient
4981 test. It is now used for both.
4983 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4984 patch removes a few documentation additions to RFC 3028, because the
4985 latest draft now contains them. It adds the new en;ascii-case comparator
4986 and a new error check for 8bit text in MIME parts. Comparator and
4987 require names are now matched exactly. I enabled the subaddress
4988 extension, but it is not well tested yet (read: it works for me)."
4990 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4991 rework some of the code of TK/09 above to avoid the hardwired use of
4992 "%lld" and "long long". Replaced the call to snprintf() with a call to
4995 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4996 records point to non-existent hosts", "retry timeout exceeded", and
4997 "retry time not reached for any host after a long failure period".
4999 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
5000 experimental DomainKeys support:
5002 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
5003 (2) On an error such as an illegally used "control", the wrong name for
5004 the control was given.
5006 These problems did NOT occur unless DomainKeys support was compiled.
5008 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
5010 PH/24 Added ${if match_ip condition.
5012 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
5013 databases so that it will be absolutely obvious if a crash occurs in the
5014 DB library. This is a regular occurrence (often caused by mis-matched
5017 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
5018 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
5019 picked up on a user's system that detects such things. There doesn't seem
5020 to be a gcc warning option for this - only an attribute that has to be
5021 put on the function's prototype. It seems that in Fedora Core 4 they have
5022 set this on a number of new functions. No doubt there will be more in due
5025 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
5026 specify a transport. However, if an address that was verified by such a
5027 router was the subject of a callout, Exim crashed because it tried to
5028 read the rcpt_include_affixes from the non-existent transport. Now it
5029 just assumes that the setting of that option is false. This bug was
5030 introduced by 4.51/PH/31.
5032 PH/28 Changed -d+all to exclude +memory, because that information is very
5033 rarely of interest, but it makes the output a lot bigger. People tend to
5034 do -d+all out of habit.
5036 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
5037 code in os-type was giving problems when libc.so lives in lib64, like on
5040 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
5041 aren't the modern standard, and it seems that some systems' include files
5042 don't always have them. Exim was already checking for some of the newer
5043 ones like T_AAAA, and defining it itself. I've added checks for all the
5044 record types that Exim uses.
5046 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
5047 not automatically generating a new one, as it is supposed to. This
5048 prevented TLS from working. If the file did exist, but contained invalid
5049 data, a new version was generated, as expected. It was only the case of a
5050 non-existent file that was broken.
5052 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
5053 with a change in libdomainkeys > 0.64.
5055 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
5056 from DNS. If the selector record carries the flag, it now has
5057 precedence over the domain-wide flag.
5059 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
5061 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
5062 the use of a port name that isn't defined in /etc/services) occurred, the
5063 message was deferred as in a normal delivery, and thus remained on the
5064 spool, instead of being failed because of the mua_wrapper setting. This
5065 is now fixed, and I tidied up some of the mua_wrapper messages at the
5068 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
5069 the messages in a hash of arrays rather than using individual hashes.
5070 This is a bit cleaner and results in dramatic memory savings, albeit
5071 at a slight CPU cost.
5073 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
5074 as requested by Marc Sherman.
5076 SC/10 Eximstats: added histograms for user specified patterns as requested
5079 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
5081 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
5082 fopen() in the content-scanning modules that did not already have it.
5088 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
5089 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
5091 TK/02 Fix ACL "control" statement not being available in MIME ACL.
5093 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
5095 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
5096 to test Sieve filters that use "vacation".
5098 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
5099 that changes the way the GnuTLS parameters are stored in the cache file.
5100 The new format can be generated externally. For backward compatibility,
5101 if the data in the cache doesn't make sense, Exim assumes it has read an
5102 old-format file, and it generates new data and writes a new file. This
5103 means that you can't go back to an older release without removing the
5106 PH/03 A redirect router that has both "unseen" and "one_time" set does not
5107 work if there are any delivery delays because "one_time" forces the
5108 parent to be marked "delivered", so its unseen clone is never tried
5109 again. For this reason, Exim now forbids the simultaneous setting of
5112 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
5113 redirected to themselves ("homonym" addresses). Read the long ChangeLog
5114 entry if you want to know the details. The fix, however, neglected to
5115 consider the case when local delivery batching is involved. The test for
5116 "previously delivered" was not happening when checking to see if an
5117 address could be batched with a previous (undelivered) one; under
5118 certain circumstances this could lead to multiple deliveries to the same
5121 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
5122 in its include files, and this causes problems building Exim.
5124 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
5125 header_syntax) but Exim was just ignoring anything given after a slash.
5126 In particular, this caused confusion with an attempt to use "verify =
5127 reverse_host_lookup/defer_ok". An error is now given when options are
5128 supplied for verify items that do not have them. (Maybe reverse_host_
5129 lookup should have a defer_ok option, but that's a different point.)
5131 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
5132 defined by RFC 821) to 2048, because there were problems with some AUTH
5133 commands, and RFC 1869 says the size should be increased for extended
5134 SMTP commands that take arguments.
5136 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
5139 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
5140 "unknown" error; now it says that the functionality isn't in the binary.
5142 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
5143 an address' error message when a string expansion fails (syntax or
5144 whatever). Otherwise the password may appear in the log. Following change
5145 PH/42 below, there is no longer a chance of it appearing in a bounce
5148 PH/11 Installed exipick version 20050225.0 from John Jetmore.
5150 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
5151 of its addresses was ever tried. (Bugzilla bug #2.)
5153 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
5154 the result incorrectly in the debug output. (It correctly added a newline
5155 to what was transported.)
5157 TF/01 Added $received_time.
5159 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
5160 commented out examples of how to interface to a virus scanner and to
5161 SpamAssassin. Also added commented examples of av_scanner and
5162 spamd_address settings.
5164 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
5165 and controls are allowed in which ACLs. There were a couple of minor
5166 errors. Some of the entries in the conditions table (which is a table of
5167 where they are NOT allowed) were getting very unwieldy; rewrote them as a
5168 negation of where the condition IS allowed.
5170 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
5172 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
5173 header file does not have a version number, so I've had to invent a new
5174 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
5175 API. The code is untested by me (my Linux distribution still has 0.3.2 of
5176 radiusclient), but it was contributed by a Radius user.
5178 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
5179 files or directories, for OpenSSL.
5181 PH/19 When an Exim process that is running as root has to create an Exim log
5182 file, it does so in a subprocess that runs as exim:exim so as to get the
5183 ownership right at creation (otherwise, other Exim processes might see
5184 the file with the wrong ownership). There was no test for failure of this
5185 fork() call, which would lead to the process getting stuck as it waited
5186 for a non-existent subprocess. Forks do occasionally fail when resources
5187 run out. I reviewed all the other calls to fork(); they all seem to check
5190 PH/20 When checking for unexpected SMTP input at connect time (before writing
5191 the banner), Exim was not dealing correctly with a non-positive return
5192 from the read() function. If the client had disconnected by this time,
5193 the result was a log entry for a synchronization error with an empty
5194 string after "input=" when read() returned zero. If read() returned -1
5195 (an event I could not check), uninitialized data bytes were printed.
5196 There were reports of junk text (parts of files, etc) appearing after
5199 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
5201 PH/22 Added support for macro redefinition, and (re)definition in between
5202 driver and ACL definitions.
5204 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
5205 forgetting to use the resulting value; it was using the unexpanded value.
5207 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
5208 hadn't been configured. The fix is from Juergen Kreileder, who
5209 understands it better than I do:
5211 "Here's what I see happening with three configured cyrus_sasl
5212 authenticators configured (plain, login, cram-md5):
5214 On startup auth_cyrus_sasl_init() gets called for each of these.
5215 This means three calls to sasl_listmech() without a specified mech_list.
5216 => SASL tests which mechs of all available mechs actually work
5217 => three warnings about OTP not working
5218 => the returned list contains: plain, login, cram-md5, digest-md5, ...
5220 With the patch, sasl_listmech() also gets called three times. But now
5221 SASL's mech_list option is set to the server_mech specified in the the
5222 authenticator. Or in other words, the answer from sasl_listmech()
5223 gets limited to just the mech you're testing for (which is different
5225 => the return list contains just 'plain' or 'login', 'cram-md5' or
5226 nothing depending on the value of ob->server_mech.
5228 I've just tested the patch: Authentication still works fine,
5229 unavailable mechs specified in the exim configuration are still
5230 caught, and the auth.log warnings about OTP are gone."
5232 PH/25 When debugging is enabled, the contents of the command line are added
5233 to the debugging output, even when log_selector=+arguments is not
5236 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
5237 answer is "GNU", and only if the return is "GNU/something" is the answer
5240 PH/27 $acl_verify_message is now set immediately after the failure of a
5241 verification in an ACL, and so is available in subsequent modifiers. In
5242 particular, the message can be preserved by coding like this:
5244 warn !verify = sender
5245 set acl_m0 = $acl_verify_message
5247 Previously, $acl_verify_message was set only while expanding "message"
5248 and "log_message" when a very denied access.
5250 PH/28 Modified OS/os.c-Linux with
5252 -#ifndef OS_LOAD_AVERAGE
5253 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
5255 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
5256 nomenclature these days.)
5258 PH/29 Installed patch from the Sieve maintainer that adds the options
5259 sieve_useraddress and sieve_subaddress to the redirect router.
5261 PH/30 In these circumstances:
5262 . Two addresses routed to the same list of hosts;
5263 . First host does not offer TLS;
5264 . First host accepts first address;
5265 . First host gives temporary error to second address;
5266 . Second host offers TLS and a TLS session is established;
5267 . Second host accepts second address.
5268 Exim incorrectly logged both deliveries with the TLS parameters (cipher
5269 and peerdn, if requested) that were in fact used only for the second
5272 PH/31 When doing a callout as part of verifying an address, Exim was not paying
5273 attention to any local part prefix or suffix that was matched by the
5274 router that accepted the address. It now behaves in the same way as it
5275 does for delivery: the affixes are removed from the local part unless
5276 rcpt_include_affixes is set on the transport.
5278 PH/32 Add the sender address, as F=<...>, to the log line when logging a
5279 timeout during the DATA phase of an incoming message.
5281 PH/33 Sieve envelope tests were broken for match types other than :is. I have
5282 applied a patch sanctioned by the Sieve maintainer.
5284 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
5285 the uid or gid is negative. A case of a negative gid caused this to be
5286 noticed. The fix allows for either to be negative.
5288 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
5289 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
5292 PH/36 The change PH/12 above was broken. Fixed it.
5294 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
5295 the grounds that routing the same address twice would always produce the
5296 same answer. This might have been true once, but it is certainly no
5297 longer true now. Routing a child address may depend on the previous
5298 routing that produced that child. Some complicated redirection strategies
5299 went wrong when messages had multiple recipients, and made Exim's
5300 behaviour dependent on the order in which the addresses were given.
5302 I have moved the duplicate checking until after the routing is complete.
5303 Exim scans the addresses that are assigned to local and remote
5304 transports, and removes any duplicates. This means that more work will be
5305 done, as duplicates will always all be routed, but duplicates are
5306 presumably rare, so I don't expect this is of any significance.
5308 For deliveries to pipes, files, and autoreplies, the duplicate checking
5309 still happens during the routing process, since they are not going to be
5312 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
5313 It corrects a timeout issue with spamd. This is Ian's comment: "The
5314 background is that sometimes spamd either never reads data from a
5315 connection it has accepted, or it never writes response data. The exiscan
5316 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
5317 blindly assumes that writes won't block so it may never time out."
5319 PH/39 Allow G after quota size as well as K and M.
5321 PH/40 The value set for $authenticated_id in an authenticator may not contain
5322 binary zeroes or newlines because the value is written to log lines and
5323 to spool files. There was no check on this. Now the value is run through
5324 the string_printing() function so that such characters are converted to
5325 printable escape sequences.
5327 PH/41 $message_linecount is a new variable that contains the total number of
5328 lines in the message. Compare $body_linecount, which is the count for the
5331 PH/42 Exim no longer gives details of delivery errors for specific addresses in
5332 bounce and delay warning messages, except in certain special cases, which
5335 (a) An SMTP error message from a remote host;
5336 (b) A message specified in a :fail: redirection;
5337 (c) A message specified in a "fail" command in a system filter;
5338 (d) A message specified in a FAIL return from the queryprogram router;
5339 (e) A message specified by the cannot_route_message router option.
5341 In these cases only, Exim does include the error details in bounce and
5342 warning messages. There are also a few cases where bland messages such
5343 as "unrouteable address" or "local delivery error" are given.
5345 PH/43 $value is now also set for the "else" part of a ${run expansion.
5347 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
5348 being worked on, but at least Exim now implements the latest version to
5351 PH/45 In a pipe transport, although a timeout while waiting for the pipe
5352 process to complete was treated as a delivery failure, a timeout while
5353 writing the message to the pipe was logged, but erroneously treated as a
5354 successful delivery. Such timeouts include transport filter timeouts. For
5355 consistency with the overall process timeout, these timeouts are now
5356 treated as errors, giving rise to delivery failures by default. However,
5357 there is now a new Boolean option for the pipe transport called
5358 timeout_defer, which, if set TRUE, converts the failures into defers for
5359 both kinds of timeout. A transport filter timeout is now identified in
5362 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
5363 systems where "make" and "gmake" are different, calling "gmake" at top
5364 level broke things. I've arranged for the value of $(MAKE) to be passed
5365 from the Makefile to this script so that it can call the same version of
5369 A note about Exim versions 4.44 and 4.50
5370 ----------------------------------------
5372 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
5373 changes of various kinds. As a consequence, a big documentation update was
5374 needed. This delayed the release for rather longer than seemed good, especially
5375 in the light of a couple of (minor) security issues. Therefore, the changes
5376 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
5377 release. So 4.44 and 4.50 are in effect two different branches that both start
5380 I have left the 4.50 change log unchanged; it contains all the changes since
5381 4.43. The change log for 4.44 is below; many of its items are identical to
5382 those for 4.50. This seems to be the most sensible way to preserve the
5383 historical information.
5389 1. Minor wording change to the doc/README.SIEVE file.
5391 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
5392 computation of the current number of files was incorrect.
5394 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5395 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5398 4. Give more explanation in the error message when the command for a transport
5399 filter fails to execute.
5401 5. There are several places where Exim runs a non-Exim command in a
5402 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5403 was being done only for the command run by the queryprogram router. It is
5404 now done for all such subprocesses. The other cases are: ${run, transport
5405 filters, and the commands run by the lmtp and pipe transports.
5407 6. Added CONFIGURE_GROUP build-time option.
5409 7. Some older OS have a limit of 256 on the maximum number of file
5410 descriptors. Exim was using setrlimit() to set 1000 as a large value
5411 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5412 systems. I've change it so that if it can't get 1000, it tries for 256.
5414 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5415 was an oversight, and furthermore, ever since the addition of extra
5416 controls (e.g. 4.43/32), the checks on when to allow different forms of
5417 "control" were broken. There should now be diagnostics for all cases when a
5418 control that does not make sense is encountered.
5420 9. Added the /retain_sender option to "control=submission".
5422 10. $recipients is now available in the predata ACL (oversight).
5424 11. Tidy the search cache before the fork to do a delivery from a message
5425 received from the command line. Otherwise the child will trigger a lookup
5426 failure and thereby defer the delivery if it tries to use (for example) a
5427 cached ldap connection that the parent has called unbind on.
5429 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5430 of $address_data from the recipient verification was clobbered by the
5431 sender verification.
5433 13. The value of address_data from a sender verification is now available in
5434 $sender_address_data in subsequent conditions in the ACL statement.
5436 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
5438 15. Added a new option "connect=<time>" to callout options, to set a different
5441 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5442 was its contents. (It was OK if the option was not defined at all.)
5444 17. A "Completed" log line is now written for messages that are removed from
5445 the spool by the -Mrm option.
5447 18. New variables $sender_verify_failure and $recipient_verify_failure contain
5448 information about exactly what failed.
5450 19. Added -dd to debug only the daemon process.
5452 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
5453 handles timeouts, both on the server side and network timeouts. Renamed the
5454 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
5456 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
5457 It is now set to "smtps".
5459 22. $host_address is now set to the target address during the checking of
5460 ignore_target_hosts.
5462 23. When checking ignore_target_hosts for an ipliteral router, no host name was
5463 being passed; this would have caused $sender_host_name to have been used if
5464 matching the list had actually called for a host name (not very likely,
5465 since this list is usually IP addresses). A host name is now passed as
5468 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5469 code that specifies a non-restarting handler (typically sigaction() in
5470 modern systems) in an attempt to fix a rare and obscure crash bug.
5472 25. Narrowed the window for a race in the daemon that could cause it to ignore
5473 SIGCHLD signals. This is not a major problem, because they are used only to
5474 wake it up if nothing else does.
5476 26. A malformed maildirsize file could cause Exim to calculate negative values
5477 for the mailbox size or file count. Odd effects could occur as a result.
5478 The maildirsize information is now recalculated if the size or filecount
5481 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5482 support for a long time. Removed HAVE_SYS_VFS_H.
5484 28. Installed the latest version of exipick from John Jetmore.
5486 29. In an address list, if the pattern was not a regular expression, an empty
5487 subject address (from a bounce message) matched only if the pattern was an
5488 empty string. Non-empty patterns were not even tested. This was the wrong
5489 because it is perfectly reasonable to use an empty address as part of a
5490 database query. An empty address is now tested by patterns that are
5491 lookups. However, all the other forms of pattern expect the subject to
5492 contain a local part and a domain, and therefore, for them, an empty
5493 address still always fails if the pattern is not itself empty.
5495 30. Exim went into a mad DNS loop when attempting to do a callout where the
5496 host was specified on an smtp transport, and looking it up yielded more
5497 than one IP address.
5499 31. Re-factored the code for checking spool and log partition space into a
5500 function that finds that data and another that does the check. The former
5501 is then used to implement four new variables: $spool_space, $log_space,
5502 $spool_inodes, and $log_inodes.
5504 32. The RFC2047 encoding function was originally intended for short strings
5505 such as real names; it was not keeping to the 75-character limit for
5506 encoded words that the RFC imposes. It now respects the limit, and
5507 generates multiple encoded words if necessary. To be on the safe side, I
5508 have increased the buffer size for the ${rfc2047: expansion operator from
5511 33. It is now permitted to omit both strings after an "if" condition; if the
5512 condition is true, the result is "true". As before, when the second string
5513 is omitted, a false condition yields an empty string. This makes it less
5514 cumbersome to write custom ACL and router conditions.
5516 34. Failure to deliver a bounce message always caused it to be frozen, even if
5517 there was an errors_to setting on the router. The errors_to setting is now
5520 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5521 canonical form (fully expanded) before being placed in
5522 $sender_host_address.
5524 36. The table in the code that translates DNS record types into text (T_A to
5525 "A" for instance) was missing entries for NS and CNAME. It is just possible
5526 that this could have caused confusion if both these types were looked up
5527 for the same domain, because the text type is used as part of Exim's
5528 per-process caching. But the chance of anyone hitting this buglet seems
5531 37. The dnsdb lookup has been extended in a number of ways.
5533 (1) There is a new type, "zns", which walks up the domain tree until it
5534 finds some nameserver records. It should be used with care.
5536 (2) There is a new type, "mxh", which is like "mx" except that it returns
5537 just the host names, not the priorities.
5539 (3) It is now possible to give a list of domains (or IP addresses) to be
5540 looked up. The behaviour when one of the lookups defers can be
5541 controlled by a keyword.
5543 (4) It is now possible to specify the separator character for use when
5544 multiple records are returned.
5546 38. The dnslists ACL condition has been extended: it is now possible to supply
5547 a list of IP addresses and/or domains to be looked up in a particular DNS
5550 39. Added log_selector=+queue_time_overall.
5552 40. When running the queue in the test harness, wait just a tad after forking a
5553 delivery process, to get repeatability of debugging output.
5555 41. Include certificate and key file names in error message when GnuTLS fails
5556 to set them up, because the GnuTLS error message doesn't include the name
5557 of the failing file when there is a problem reading it.
5559 42. Allow both -bf and -bF in the same test run.
5561 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
5563 44. The "Exiscan patch" is now merged into the mainline Exim source.
5565 45. Sometimes the final signoff response after QUIT could fail to get
5566 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5567 < 0 before doing a fflush(). This bug looks as though it goes back to the
5568 introduction of TLS in release 3.20, but "sometimes" must have been rare
5569 because the tests only now provoked it.
5571 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
5572 (this can affect the format of dates).
5574 47. exim_tidydb, when checking for the continued existence of a message for
5575 which it has found a message-specific retry record, was not finding
5576 messages that were in split spool directories. Consequently, it was
5577 deleting retry records that should have stayed in existence.
5579 48. Steve fixed some bugs in eximstats.
5581 49. The SPA authentication driver was not abandoning authentication and moving
5582 on to the next authenticator when an expansion was forced to fail,
5583 contradicting the general specification for all authenticators. Instead it
5584 was generating a temporary error. It now behaves as specified.
5586 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
5587 (the order specifies the preference for clients). The order is now AES256,
5588 AES128, 3DES, ARCFOUR128.
5590 51. Small patch to Sieve code - explicitly set From: when generating an
5593 52. Exim crashed if a remote delivery caused a very long error message to be
5594 recorded - for instance if somebody sent an entire SpamAssassin report back
5595 as a large number of 550 error lines. This bug was coincidentally fixed by
5596 increasing the size of one of Exim's internal buffers (big_buffer) that
5597 happened as part of the Exiscan merge. However, to be on the safe side, I
5598 have made the code more robust (and fixed the comments that describe what
5601 53. Now that there can be additional text after "Completed" in log lines (if
5602 the queue_time_overall log selector is set), a one-byte patch to exigrep
5603 was needed to allow it to recognize "Completed" as not the last thing in
5606 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
5607 patch that reportedly fixes this has been added. I am not expert enough to
5608 create a test for it. This is what the patch creator wrote:
5610 "I found a little strange behaviour of ldap code when working with
5611 Windows 2003 AD Domain, where users was placed in more than one
5612 Organization Units. When I tried to give exim partial DN, the exit code
5613 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
5614 But simultaneously result of request was absolutely normal ldap result,
5615 so I produce this patch..."
5617 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
5618 REFERENCE, so I have modified the code to exclude the patch when that macro
5621 55. Some experimental protocols are using DNS PTR records for new purposes. The
5622 keys for these records are domain names, not reversed IP addresses. The
5623 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5624 leaves it alone. Component reversal etc. now happens only for IP addresses.
5627 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
5629 57. Double the size of the debug message buffer (to 2048) so that more of very
5630 long debug lines gets shown.
5632 58. The exicyclog utility now does better if the number of log files to keep
5633 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
5635 59. Two changes related to the smtp_active_hostname option:
5637 (1) $smtp_active_hostname is now available as a variable.
5638 (2) The default for smtp_banner uses $smtp_active_hostname instead
5639 of $primary_hostname.
5641 60. The host_aton() function is supposed to be passed a string that is known
5642 to be a valid IP address. However, in the case of IPv6 addresses, it was
5643 not checking this. This is a hostage to fortune. Exim now panics and dies
5644 if the condition is not met. A case was found where this could be provoked
5645 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5646 components; fortuitously, this particular loophole had already been fixed
5647 by change 4.50/55 above.
5649 If there are any other similar loopholes, the new check in host_aton()
5650 itself should stop them being exploited. The report I received stated that
5651 data on the command line could provoke the exploit when Exim was running as
5652 exim, but did not say which command line option was involved. All I could
5653 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5654 running as the user.
5657 61. There was a buffer overflow vulnerability in the SPA authentication code
5658 (which came originally from the Samba project). I have added a test to the
5659 spa_base64_to_bits() function which I hope fixes it.
5662 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
5663 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
5665 63. The daemon start-up calls getloadavg() while still root for those OS that
5666 need the first call to be done as root, but it missed one case: when
5667 deliver_queue_load_max is set with deliver_drop_privilege. This is
5668 necessary for the benefit of the queue runner, because there is no re-exec
5669 when deliver_drop_privilege is set.
5671 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
5672 This has been fixed.
5674 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5675 was in use, was not putting the data itself into the right store pool;
5676 consequently, it could be overwritten for a subsequent message in the same
5677 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5680 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
5682 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
5683 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
5684 the function were treating the return as a boolean value, which happened to
5685 work because 0=false and not-0=true, but is not correct code.
5687 68. The host_aton() function was not handling scoped IPv6 addresses (those
5688 with, for example, "%eth0" on the end) correctly.
5690 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
5691 negated items (that is, ~something) in unsigned ints. Some compilers
5692 apparently mutter when there is no cast.
5694 70. If an address verification called from an ACL failed, and did not produce a
5695 user-specific message (i.e. there was only a "system" message), nothing was
5696 put in $acl_verify_message. In this situation, it now puts the system
5699 71. Change 4.23/11 added synchronization checking at the start of an SMTP
5700 session; change 4.31/43 added the unwanted input to the log line - except
5701 that it did not do this in the start of session case. It now does.
5703 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
5704 This is wrong and can cause the other end to generate a synchronization
5705 error if it is another Exim or anything else that does the synchronization
5706 check. A QUIT command is no longer sent after a timeout.
5708 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
5709 during host lookups.
5711 74. The defer_ok option of callout verification was not working if it was used
5712 when verifying addresses in header lines, that is, for this case:
5714 verify = header_sender/callout=defer_ok
5716 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
5717 those file descriptors could be used for SMTP connections. If anything
5718 wrote to stderr (the example that came up was "warn" in embedded Perl), it
5719 could be sent to the SMTP client, causing chaos. The daemon now opens
5720 stdin, stdout, and stderr to /dev/null when it puts itself into the
5723 76. Arrange for output from Perl's "warn" command to be written to Exim's main
5724 log by default. The user can override this with suitable Perl magic.
5726 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
5727 the log message when discard triggers, was not working for the DATA ACL or
5728 for the non-SMTP ACL.
5730 78. Error message wording change in sieve.c.
5732 79. If smtp_accept_max_per_host was set, the number of connections could be
5733 restricted to fewer than expected, because the daemon was trying to set up
5734 a new connection before checking whether the processes handling previous
5735 connections had finished. The check for completed processes is now done
5736 earlier. On busy systems, this bug wouldn't be noticed because something
5737 else would have woken the daemon, and it would have reaped the completed
5740 80. If a message was submitted locally by a user whose login name contained one
5741 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
5742 It caused a spool format error. I have fixed the spool reading code. A
5743 related problem was that the "from" clause in the Received: line became
5744 illegal because of the space(s). It is now covered by ${quote_local_part.
5746 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
5749 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
5751 83. Patch from Sieve maintainer to fix unterminated string problem in
5752 "vacation" handling.
5754 84. Some minor changes to the Linux configuration files to help with other
5755 OS variants using glibc.
5757 85. One more patch for Sieve to update vacation handling to latest spec.
5760 ----------------------------------------------------
5761 See the note above about the 4.44 and 4.50 releases.
5762 ----------------------------------------------------
5768 1. Change 4.43/35 introduced a bug that caused file counts to be
5769 incorrectly computed when quota_filecount was set in an appendfile
5772 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5773 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5776 3. Give more explanation in the error message when the command for a transport
5777 filter fails to execute.
5779 4. There are several places where Exim runs a non-Exim command in a
5780 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5781 was being done only for the command run by the queryprogram router. It is
5782 now done for all such subprocesses. The other cases are: ${run, transport
5783 filters, and the commands run by the lmtp and pipe transports.
5785 5. Some older OS have a limit of 256 on the maximum number of file
5786 descriptors. Exim was using setrlimit() to set 1000 as a large value
5787 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5788 systems. I've change it so that if it can't get 1000, it tries for 256.
5790 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5791 was an oversight, and furthermore, ever since the addition of extra
5792 controls (e.g. 4.43/32), the checks on when to allow different forms of
5793 "control" were broken. There should now be diagnostics for all cases when a
5794 control that does not make sense is encountered.
5796 7. $recipients is now available in the predata ACL (oversight).
5798 8. Tidy the search cache before the fork to do a delivery from a message
5799 received from the command line. Otherwise the child will trigger a lookup
5800 failure and thereby defer the delivery if it tries to use (for example) a
5801 cached ldap connection that the parent has called unbind on.
5803 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5804 of $address_data from the recipient verification was clobbered by the
5805 sender verification.
5807 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5808 was its contents. (It was OK if the option was not defined at all.)
5810 11. A "Completed" log line is now written for messages that are removed from
5811 the spool by the -Mrm option.
5813 12. $host_address is now set to the target address during the checking of
5814 ignore_target_hosts.
5816 13. When checking ignore_target_hosts for an ipliteral router, no host name was
5817 being passed; this would have caused $sender_host_name to have been used if
5818 matching the list had actually called for a host name (not very likely,
5819 since this list is usually IP addresses). A host name is now passed as
5822 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5823 code that specifies a non-restarting handler (typically sigaction() in
5824 modern systems) in an attempt to fix a rare and obscure crash bug.
5826 15. Narrowed the window for a race in the daemon that could cause it to ignore
5827 SIGCHLD signals. This is not a major problem, because they are used only to
5828 wake it up if nothing else does.
5830 16. A malformed maildirsize file could cause Exim to calculate negative values
5831 for the mailbox size or file count. Odd effects could occur as a result.
5832 The maildirsize information is now recalculated if the size or filecount
5835 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5836 support for a long time. Removed HAVE_SYS_VFS_H.
5838 18. Updated exipick to current release from John Jetmore.
5840 19. Allow an empty sender to be matched against a lookup in an address list.
5841 Previously the only cases considered were a regular expression, or an
5844 20. Exim went into a mad DNS lookup loop when doing a callout where the
5845 host was specified on the transport, if the DNS lookup yielded more than
5848 21. The RFC2047 encoding function was originally intended for short strings
5849 such as real names; it was not keeping to the 75-character limit for
5850 encoded words that the RFC imposes. It now respects the limit, and
5851 generates multiple encoded words if necessary. To be on the safe side, I
5852 have increased the buffer size for the ${rfc2047: expansion operator from
5855 22. Failure to deliver a bounce message always caused it to be frozen, even if
5856 there was an errors_to setting on the router. The errors_to setting is now
5859 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5860 canonical form (fully expanded) before being placed in
5861 $sender_host_address.
5863 24. Updated eximstats to version 1.33
5865 25. Include certificate and key file names in error message when GnuTLS fails
5866 to set them up, because the GnuTLS error message doesn't include the name
5867 of the failing file when there is a problem reading it.
5869 26. Expand error message when OpenSSL has problems setting up cert/key files.
5872 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5873 (this can affect the format of dates).
5875 28. exim_tidydb, when checking for the continued existence of a message for
5876 which it has found a message-specific retry record, was not finding
5877 messages that were in split spool directories. Consequently, it was
5878 deleting retry records that should have stayed in existence.
5880 29. eximstats updated to version 1.35
5881 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5882 1.35 - bugfix such that pie charts by volume are generated correctly
5884 30. The SPA authentication driver was not abandoning authentication and moving
5885 on to the next authenticator when an expansion was forced to fail,
5886 contradicting the general specification for all authenticators. Instead it
5887 was generating a temporary error. It now behaves as specified.
5889 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5890 (the order specifies the preference for clients). The order is now AES256,
5891 AES128, 3DES, ARCFOUR128.
5893 31. Small patch to Sieve code - explicitly set From: when generating an
5896 32. Exim crashed if a remote delivery caused a very long error message to be
5897 recorded - for instance if somebody sent an entire SpamAssassin report back
5898 as a large number of 550 error lines. This bug was coincidentally fixed by
5899 increasing the size of one of Exim's internal buffers (big_buffer) that
5900 happened as part of the Exiscan merge. However, to be on the safe side, I
5901 have made the code more robust (and fixed the comments that describe what
5904 33. Some experimental protocols are using DNS PTR records for new purposes. The
5905 keys for these records are domain names, not reversed IP addresses. The
5906 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5907 leaves it alone. Component reversal etc. now happens only for IP addresses.
5910 34. The host_aton() function is supposed to be passed a string that is known
5911 to be a valid IP address. However, in the case of IPv6 addresses, it was
5912 not checking this. This is a hostage to fortune. Exim now panics and dies
5913 if the condition is not met. A case was found where this could be provoked
5914 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5915 components; fortuitously, this particular loophole had already been fixed
5916 by change 4.50/55 or 4.44/33 above.
5918 If there are any other similar loopholes, the new check in host_aton()
5919 itself should stop them being exploited. The report I received stated that
5920 data on the command line could provoke the exploit when Exim was running as
5921 exim, but did not say which command line option was involved. All I could
5922 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5923 running as the user.
5926 35. There was a buffer overflow vulnerability in the SPA authentication code
5927 (which came originally from the Samba project). I have added a test to the
5928 spa_base64_to_bits() function which I hope fixes it.
5931 36. The daemon start-up calls getloadavg() while still root for those OS that
5932 need the first call to be done as root, but it missed one case: when
5933 deliver_queue_load_max is set with deliver_drop_privilege. This is
5934 necessary for the benefit of the queue runner, because there is no re-exec
5935 when deliver_drop_privilege is set.
5937 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5938 was in use, was not putting the data itself into the right store pool;
5939 consequently, it could be overwritten for a subsequent message in the same
5940 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5943 38. Sometimes the final signoff response after QUIT could fail to get
5944 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5945 < 0 before doing a fflush(). This bug looks as though it goes back to the
5946 introduction of TLS in release 3.20, but "sometimes" must have been rare
5947 because the tests only now provoked it.
5953 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5954 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5955 Now it returns an integer. A number of calls were still expecting a T/F
5956 return. Fortuitously, in all cases, the tests worked in OK situations,
5957 which is the norm. However, things would have gone wrong on any write
5958 failures on the smtp file descriptor. This function is used when sending
5959 messages over SMTP and also when doing verify callouts.
5961 2. When Exim is called to do synchronous delivery of a locally submitted
5962 message (the -odf or -odi options), it no longer closes stderr before doing
5965 3. Implemented the mua_wrapper option.
5967 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5969 5. Implemented the functions header_remove(), header_testname(),
5970 header_add_at_position(), and receive_remove_recipient(), and exported them
5973 6. If an ACL "warn" statement specified the addition of headers, Exim already
5974 inserted X-ACL-Warn: at the start if there was no header name. However, it
5975 was not making this test for the second and subsequent header lines if
5976 there were newlines in the string. This meant that an invalid header could
5977 be inserted if Exim was badly configured.
5979 7. Allow an ACL "warn" statement to add header lines at the start or after all
5980 the Received: headers, as well as at the end.
5982 8. Added the rcpt_4xx retry error code.
5984 9. Added postmaster_mailfrom=xxx to callout verification option.
5986 10. Added mailfrom=xxxx to the callout verification option, for verify=
5989 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5990 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5992 12. Inserted some casts to stop certain compilers warning when using pointer
5993 differences as field lengths or precisions in printf-type calls (mostly
5994 affecting debugging statements).
5996 13. Added optional readline() support for -be (dynamically loaded).
5998 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5999 same clock tick as a message's arrival, so that its received time was the
6000 same as the "first fail" time on the retry record, and that message
6001 remained on the queue past the ultimate address timeout, every queue runner
6002 would try a delivery (because it was past the ultimate address timeout) but
6003 after another failure, the ultimate address timeout, which should have then
6004 bounced the address, did not kick in. This was a "< instead of <=" error;
6005 in most cases the first failure would have been in the next clock tick
6006 after the received time, and all would be well.
6008 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
6009 being recognized when the domain list was tested by the match_domain
6010 condition in an expansion string.
6012 16. Added the ${str2b64: operator.
6014 17. Exim was always calling setrlimit() to set a large limit for the number of
6015 processes, without checking whether the existing limit was already
6016 adequate. (It did check for the limit on file descriptors.) Furthermore,
6017 errors from getrlimit() and setrlimit() were being ignored. Now they are
6018 logged to the main and panic logs, but Exim does carry on, to try to do its
6019 job under whatever limits there are.
6021 18. Imported PCRE 5.0.
6023 19. Trivial typo in log message " temporarily refused connection" (the leading
6026 20. If the log selector return_path_on_delivery was set and an address was
6027 redirected to /dev/null, the delivery process crashed because it assumed
6028 that a return path would always be set for a "successful" delivery. In this
6029 case, the whole delivery is bypassed as an optimization, and therefore no
6032 21. Internal re-arrangement: the function for sending a challenge and reading
6033 a response while authentication was assuming a zero-terminated challenge
6034 string. It's now changed to take a pointer and a length, to allow for
6035 binary data in such strings.
6037 22. Added the cyrus_sasl authenticator (code supplied by MBM).
6039 23. Exim was not respecting finduser_retries when seeking the login of the
6040 uid under which it was called; it was always trying 10 times. (The default
6041 setting of finduser_retries is zero.) Also, it was sleeping after the final
6042 failure, which is pointless.
6044 24. Implemented tls_on_connect_ports.
6046 25. Implemented acl_smtp_predata.
6048 26. If the domain in control=submission is set empty, Exim assumes that the
6049 authenticated id is a complete email address when it generates From: or
6050 Sender: header lines.
6052 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
6053 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
6054 chown and chgrp in /bin and hostname in /usr/bin.
6056 28. Exim was keeping the "process log" file open after each use, just as it
6057 does for the main log. This opens the possibility of it remaining open for
6058 long periods when the USR1 signal hits a daemon. Occasional processlog
6059 errors were reported, that could have been caused by this. Anyway, it seems
6060 much more sensible not to leave this file open at all, so that is what now
6063 29. The long-running daemon process does not normally write to the log once it
6064 has entered its main loop, and it closes the log before doing so. This is
6065 so that log files can straightforwardly be renamed and moved. However,
6066 there are a couple of unusual error situations where the daemon does write
6067 log entries, and I had neglected to close the log afterwards.
6069 30. The text of an SMTP error response that was received during a remote
6070 delivery was being truncated at 512 bytes. This is too short for some of
6071 the long messages that one sometimes sees. I've increased the limit to
6074 31. It is now possible to make retry rules that apply only when a message has a
6075 specific sender, in particular, an empty sender.
6077 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
6078 it possible to be selective about when SMTP synchronization is enforced.
6080 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
6082 32. Implemented hosts_connection_nolog.
6084 33. Added an ACL for QUIT.
6086 34. Setting "delay_warning=" to disable warnings was not working; it gave a
6089 35. Added mailbox_size and mailbox_filecount to appendfile.
6091 36. Added control = no_multiline_responses to ACLs.
6093 37. There was a bug in the logic of the code that waits for the clock to tick
6094 in the case where the clock went backwards by a substantial amount such
6095 that the microsecond fraction of "now" was more than the microsecond
6096 fraction of "then" (but the whole seconds number was less).
6098 38. Added support for the libradius Radius client library this is found on
6099 FreeBSD (previously only the radiusclient library was supported).
6105 1. When certain lookups returned multiple values in the form name=value, the
6106 quoting of the values was not always being done properly. Specifically:
6107 (a) If the value started with a double quote, but contained no whitespace,
6109 (b) If the value contained whitespace other than a space character (i.e.
6110 tabs or newlines or carriage returns) it was not quoted.
6111 This fix has been applied to the mysql and pgsql lookups by writing a
6112 separate quoting function and calling it from the lookup code. The fix
6113 should probably also be applied to nisplus, ibase and oracle lookups, but
6114 since I cannot test any of those, I have not disturbed their existing code.
6116 2. A hit in the callout cache for a specific address caused a log line with no
6117 reason for rejecting RCPT. Now it says "Previous (cached) callout
6118 verification failure".
6120 3. There was an off-by-one bug in the queryprogram router. An over-long
6121 return line was truncated at 256 instead of 255 characters, thereby
6122 overflowing its buffer with the terminating zero. As well as fixing this, I
6123 have increased the buffer size to 1024 (and made a note to document this).
6125 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
6126 when Exim is waiting for an SMTP response from a remote server, Exim
6127 restarts its select() call on the socket, thereby resetting its timeout.
6128 This is not a problem when such interrupts are rare. Somebody set up a cron
6129 job to run exiwhat every 2 minutes, which is less than the normal select()
6130 timeout (5 or 10 minutes). This meant that the select() timeout never
6131 kicked in because it was always reset. I have fixed this by comparing the
6132 time when an interrupt arrives with the time at the start of the first call
6133 to select(). If more time than the timeout has elapsed, the interrupt is
6134 treated as a timeout.
6136 5. Some internal re-factoring in preparation for the addition of Sieve
6137 extensions (by MH). In particular, the "personal" test is moved to a
6138 separate function, and given an option for scanning Cc: and Bcc: (which is
6139 not set for Exim filters).
6141 6. When Exim created an email address using the login of the caller as the
6142 local part (e.g. when creating a From: or Sender: header line), it was not
6143 quoting the local part when it contained special characters such as @.
6145 7. Installed new OpenBSD configuration files.
6147 8. Reworded some messages for syntax errors in "and" and "or" conditions to
6148 try to make them clearer.
6150 9. Callout options, other than the timeout value, were being ignored when
6151 verifying sender addresses in header lines. For example, when using
6153 verify = header_sender/callout=no_cache
6155 the cache was (incorrectly) being used.
6157 10. Added a missing instance of ${EXE} to the exim_install script; this affects
6158 only the Cygwin environment.
6160 11. When return_path_on_delivery was set as a log selector, if different remote
6161 addresses in the same message used different return paths and parallel
6162 remote delivery occurred, the wrong values would sometimes be logged.
6163 (Whenever a remote delivery process finished, the return path value from
6164 the most recently started remote delivery process was logged.)
6166 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
6167 lines when AUTH and/or TLS are in use. This is the "received protocol"
6168 field. Exim used to use "asmtp" for authenticated SMTP, without any
6169 indication (in the protocol name) for TLS use. Now it follows the RFC and
6170 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
6171 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
6172 names appear in log lines as well as in Received: header lines.
6174 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
6175 extensions, and comparison tests, and to fix some bugs.
6177 14. Changes to the "personal" filter test:
6179 (1) The test was buggy in that it was just doing the equivalent of
6180 "contains" tests on header lines. For example, if a user's address was
6181 anne@some.where, the "personal" test would incorrectly be true for
6183 To: susanne@some.where
6185 This test is now done by extracting each address from the header in turn,
6186 and checking the entire address. Other tests that are part of "personal"
6187 are now done using regular expressions (for example, to check local parts
6188 of addresses in From: header lines).
6190 (2) The list of non-personal local parts in From: addresses has been
6191 extended to include "listserv", "majordomo", "*-request", and "owner-*",
6192 taken from the Sieve specification recommendations.
6194 (3) If the message contains any header line starting with "List-" it is
6195 treated as non-personal.
6197 (4) The test for "circular" in the Subject: header line has been removed
6198 because it now seems ill-conceived.
6200 15. Minor typos in src/EDITME comments corrected.
6202 16. Installed latest exipick from John Jetmore.
6204 17. If headers_add on a router specified a text string that was too long for
6205 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
6206 of string_sprintf() is now avoided.
6208 18. $message_body_size was not set (it was always zero) when running the DATA
6209 ACL and the local_scan() function.
6211 19. For the "mail" command in an Exim filter, no default was being set for
6212 the once_repeat time, causing a random time value to be used if "once" was
6213 specified. (If the value happened to be <= 0, no repeat happened.) The
6214 default is now 0s, meaning "never repeat". The "vacation" command was OK
6215 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
6216 (I found it when inspecting the code).
6218 20. There is now an overall timeout for performing a callout verification. It
6219 defaults to 4 times the callout timeout, which applies to individual SMTP
6220 commands during the callout. The overall timeout applies when there is more
6221 than one host that can be tried. The timeout is checked before trying the
6222 next host. This prevents very long delays if there are a large number of
6223 hosts and all are timing out (e.g. when the network connections are timing
6224 out). The value of the overall timeout can be changed by specifying an
6225 additional sub-option for "callout", called "maxwait". For example:
6227 verify = sender/callout=5s,maxwait=20s
6229 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
6230 to the end before writing, but this should make it even safer).
6232 22. Exim was forgetting that it had advertised PIPELINING for the second and
6233 subsequent messages on an SMTP connection. It was also not resetting its
6234 memory on STARTTLS and an internal HELO.
6236 23. When Exim logs an SMTP synchronization error within a session, it now
6237 records whether PIPELINING has been advertised or not.
6239 24. Added 3 instances of "(long int)" casts to time_t variables that were being
6240 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
6241 rather than long int.
6243 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
6245 26. Added the never_mail option to autoreply.
6251 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
6252 crash if the getsockname() call failed; this can happen if a connection is
6253 closed very soon after it is established. The problem was simply in the
6254 order in which certain operations were done, causing Exim to try to write
6255 to the SMTP stream before it had set up the file descriptor. The bug has
6256 been fixed by making things happen in the correct order.
6262 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
6263 before the connection was closed, thus losing the rejection response.
6265 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
6266 some early Solaris releases, but causes trouble in current releases where
6267 socklen_t is defined.
6269 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
6272 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
6275 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
6276 that releasing the top of it at the end releases what was used for sub-
6277 expansions (unless the block got too big). However, discard this block if
6278 the first thing is a variable or header, so that we can use its block when
6279 it is dynamic (useful for very large $message_headers, for example).
6281 6. Lookups now cache *every* query, not just the most recent. A new, separate
6282 store pool is used for this. It can be recovered when all lookup caches are
6283 flushed. Lookups now release memory at the end of their result strings.
6284 This has involved some general refactoring of the lookup sources.
6286 7. Some code has been added to the store_xxx() functions to reduce the amount
6287 of flapping under certain conditions.
6289 8. log_incoming_interface used to affect only the <= reception log lines. Now
6290 it causes the local interface and port to be added to several more SMTP log
6291 lines, for example "SMTP connection from", and rejection lines.
6293 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
6295 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
6297 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
6298 could be overwritten at the end of the current message (or the start of a
6299 new message if it was set in a HELO ACL). The value is now preserved for
6300 the duration of the SMTP connection.
6302 12. If a transport had a headers_rewrite setting, and a matching header line
6303 contained an unqualified address, that address was qualified, even if it
6304 did not match any rewriting rules. The underlying bug was that the values
6305 of the flags that permit the existence of unqualified sender and recipient
6306 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
6307 non-local messages, and by -bnq for local messages) were not being
6308 preserved with the message after it was received.
6310 13. When Exim was logging an SMTP synchronization error, it could sometimes log
6311 "next input=" as part of the text comprising the host identity instead of
6312 the correct text. The code was using the same buffer for two different
6313 strings. However, depending on which order the printing function evaluated
6314 its arguments, the bug did not always show up. Under Linux, for example, my
6315 test suite worked just fine.
6317 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
6318 doesn't work with some older versions of Perl. It has been changed to "my",
6319 which in any case is probably the better facility to use.
6321 15. A really picky compiler found some instances of statements for creating
6322 error messages that either had too many or two few arguments for the format
6325 16. The size of the buffer for calls to the DNS resolver has been increased
6326 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
6327 for addresses that have a lot of PTR records. This alleviates a problem; it
6328 does not fully solve it.
6330 17. A dnsdb lookup for PTR records that receives more data than will fit in the
6331 buffer now truncates the list and logs the incident, which is the same
6332 action as happens when Exim is looking up a host name and its aliases.
6333 Previously in this situation something unpredictable would happen;
6334 sometimes it was "internal error: store_reset failed".
6336 18. If a server dropped the connection unexpectedly when an Exim client was
6337 using GnuTLS and trying to read a response, the client delivery process
6338 crashed while trying to generate an error log message.
6340 19. If a "warn" verb in an ACL added multiple headers to a message in a single
6341 string, for example:
6343 warn message = H1: something\nH2: something
6345 the text was added as a single header line from Exim's point of view
6346 though it ended up OK in the delivered message. However, searching for the
6347 second and subsequent header lines using $h_h2: did not work. This has been
6348 fixed. Similarly, if a system filter added multiple headers in this way,
6349 the routers could not see them.
6351 20. Expanded the error message when iplsearch is called with an invalid key to
6352 suggest using net-iplsearch in a host list.
6354 21. When running tests using -bh, any delays imposed by "delay" modifiers in
6355 ACLs are no longer actually imposed (and a message to that effect is
6358 22. If a "gecos" field in a passwd entry contained escaped characters, in
6359 particular, if it contained a \" sequence, Exim got it wrong when building
6360 a From: or a Sender: header from that name. A second bug also caused
6361 incorrect handling when an unquoted " was present following a character
6362 that needed quoting.
6364 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
6365 was not being matched caselessly.
6367 24. Arranged for all hyphens in the exim.8 source to be escaped with
6370 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
6371 properly. Recipient callout cache records were still being keyed to include
6372 the sender, even when use_sender was set false. This led to far more
6373 callouts that were necessary. The sender is no longer included in the key
6374 when use_sender is false.
6376 26. Added "control = submission" modifier to ACLs.
6378 27. Added the ${base62d: operator to decode base 62 numbers.
6380 28. dnsdb lookups can now access SRV records.
6382 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
6383 the configuration file.
6385 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
6386 (-v) mode. This makes the output for a verbose queue run more intelligible.
6388 31. Added a use_postmaster feature to recipient callouts.
6390 32. Added the $body_zerocount variable, containing the number of binary zero
6391 bytes in the message body.
6393 33. The time of last modification of the "new" subdirectory is now used as the
6394 "mailbox time last read" when there is a quota error for a maildir
6397 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
6399 35. Added +ignore_unknown as a special item in host lists.
6401 36. Code for decoding IPv6 addresses in host lists is now included, even if
6402 IPv6 support is not being compiled. This fixes a bug in which an IPv6
6403 address was recognized as an IP address, but was then not correctly decoded
6404 into binary, causing unexpected and incorrect effects when compared with
6411 1. Very minor rewording of debugging text in manualroute to say "list of
6412 hosts" instead of "hostlist".
6414 2. If verify=header_syntax was set, and a header line with an unqualified
6415 address (no domain) and a large number of spaces between the end of the
6416 name and the colon was received, the reception process suffered a buffer
6417 overflow, and (when I tested it) crashed. This was caused by some obsolete
6418 code that should have been removed. The fix is to remove it!
6420 3. When running in the test harness, delay a bit after writing a bounce
6421 message to get a bit more predictability in the log output.
6423 4. Added a call to search_tidyup() just before forking a reception process. In
6424 theory, someone could use a lookup in the expansion of smtp_accept_max_
6425 per_host which, without the tidyup, could leave open a database connection.
6427 5. Added the variables $recipient_data and $sender_data which get set from a
6428 lookup success in an ACL "recipients" or "senders" condition, or a router
6429 "senders" option, similar to $domain_data and $local_part_data.
6431 6. Moved the writing of debug_print from before to after the "senders" test
6434 7. Change 4.31/66 (moving the time when the Received: is generated) caused
6435 problems for message scanning, either using a data ACL, or using
6436 local_scan() because the Received: header was not generated till after they
6437 were called (in order to set the time as the time of reception completion).
6438 I have revised the way this works. The header is now generated after the
6439 body is received, but before the ACL or local_scan() are called. After they
6440 are run, the timestamp in the header is updated.
6446 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
6447 before starting a queue runner without re-exec. This happened only when
6448 deliver_drop_privilege was set or when the Exim user was set to root. The
6449 effect of the bug was that timeouts during subsequent deliveries caused
6450 crashes instead of being properly handled. The handler is now left at its
6451 default (and expected) setting.
6453 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
6454 message, again when deliver_drop_privilege is set or Exim is run as root.
6455 The bug described in (1) was not present in this case, but the tidying up
6456 of the other signals was missing. I have made the two cases consistent.
6458 3. The ignore_target_hosts setting on a manualroute router was being ignored
6459 for hosts that were looked up using the /MX notation.
6461 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
6464 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
6465 operated on the sender address. After changing the $sender_address to <>
6466 for the sender address verify, Exim was re-instated it as the original
6467 (before rewriting) address, but remembering that it had rewritten it, so it
6468 wasn't rewriting it again. This bug also had the effect of breaking the
6469 sender address verification caching when the sender address was rewritten.
6471 6. The ignore_target_hosts option was being ignored by the ipliteral router.
6472 This has been changed so that if the ip literal address matches
6473 ignore_target_hosts, the router declines.
6475 7. Added expansion conditions match_domain, match_address, and match_local_
6476 part (NOT match_host).
6478 8. The placeholder for the Received: header didn't have a length field set.
6480 9. Added code to Exim itself and to exim_lock to test for a specific race
6481 condition that could lead to file corruption when using MBX delivery. The
6482 issue is with the lockfile that is created in /tmp. If this file is removed
6483 after a process has opened it but before that process has acquired a lock,
6484 there is the potential for a second process to recreate the file and also
6485 acquire a lock. This could lead to two Exim processes writing to the file
6486 at the same time. The added code performs the same test as UW imapd; it
6487 checks after acquiring the lock that its file descriptor still refers to
6488 the same named file.
6490 10. The buffer for building added header lines was of fixed size, 8192 bytes.
6491 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
6494 11. Added the smtp_active_hostname option. If used, this will typically be made
6495 to depend on the incoming interface address. Because $interface_address is
6496 not set up until the daemon has forked a reception process, error responses
6497 that can happen earlier (such as "too many connections") no longer contain
6500 12. If an expansion in a condition on a "warn" statement fails because a lookup
6501 defers, the "warn" statement is abandoned, and the next ACL statement is
6502 processed. Previously this caused the whole ACL to be aborted.
6504 13. Added the iplsearch lookup type.
6506 14. Added ident_timeout as a log selector.
6508 15. Added tls_certificate_verified as a log selector.
6510 16. Added a global option tls_require_ciphers (compare the smtp transport
6511 option of the same name). This controls incoming TLS connections.
6513 17. I finally figured out how to make tls_require_ciphers do a similar thing
6514 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
6515 before starting the TLS session.
6517 18. Tabs are now shown as \t in -bP output.
6519 19. If the log selector return_path_on_delivery was set, Exim crashed when
6520 bouncing a message because it had too many Received: header lines.
6522 20. If two routers both had headers_remove settings, and the first one included
6523 a superfluous trailing colon, the final name in the first list and the
6524 first name in the second list were incorrectly joined into one item (with a
6525 colon in the middle).
6531 1. Added -C and -D options to the exinext utility, mainly to make it easier
6532 to include in the automated testing, but these could be helpful when
6533 multiple configurations are in use.
6535 2. The exinext utility was not formatting the output nicely when there was
6536 an alternate port involved in the retry record key, nor when there was a
6537 message id as well (for retries that were specific to a specific message
6538 and a specific host). It was also confused by IPv6 addresses, because of
6539 the additional colons they contain. I have fixed the IPv4 problem, and
6540 patched it up to do a reasonable job for IPv6.
6542 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
6543 delivery, the log line now contains "pipelined" if PIPELINING was used.
6545 4. An SMTP transport process used to panic and die if the bind() call to set
6546 an explicit outgoing interface failed. This has been changed; it is now
6547 treated in the same way as a connect() failure.
6549 5. A reference to $sender_host_name in the part of a conditional expansion
6550 that was being skipped was still causing a DNS lookup. This no longer
6553 6. The def: expansion condition was not recognizing references to header lines
6554 that used bh_ and bheader_.
6556 7. Added the _cache feature to named lists.
6558 8. The code for checking quota_filecount in the appendfile transport was
6559 allowing one more file than it should have been.
6561 9. For compatibility with Sendmail, the command line option
6569 and sets the incoming protocol and host name (for trusted callers). The
6570 host name and its colon can be omitted when only the protocol is to be set.
6571 Note the Exim already has two private options, -pd and -ps, that refer to
6572 embedded Perl. It is therefore impossible to set a protocol value of "d" or
6573 "s", but I don't think that's a major issue.
6575 10. A number of refactoring changes to the code, none of which should affect
6578 (a) The number of logging options was getting close to filling up the
6579 32-bit word that was used as a bit map. I have split them into two classes:
6580 those that are passed in the argument to log_write(), and those that are
6581 only ever tested independently outside of that function. These are now in
6582 separate 32-bit words, so there is plenty of room for expansion again.
6583 There is no change in the user interface or the logging behaviour.
6585 (b) When building, for example, log lines, the code previously used a
6586 macro that called string_cat() twice, in order to add two strings. This is
6587 not really sufficiently general. Furthermore, there was one instance where
6588 it was actually wrong because one of the argument was used twice, and in
6589 one call a function was used. (As it happened, calling the function twice
6590 did not affect the overall behaviour.) The macro has been replaced by a
6591 function that can join an arbitrary number of extra strings onto a growing
6594 (c) The code for expansion conditions now uses a table and a binary chop
6595 instead of a serial search (which was left over from when there were very
6596 few conditions). Also, it now recognizes conditions like "pam" even when
6597 the relevant support is not compiled in: a suitably worded error message is
6598 given if an attempt is made to use such a condition.
6600 11. Added ${time_interval:xxxxx}.
6602 12. A bug was causing one of the ddress fields not to be passed back correctly
6603 from remote delivery subprocesses. The field in question was not being
6604 subsequently used, so this caused to problems in practice.
6606 13. Added new log selectors queue_time and deliver_time.
6608 14. Might have fixed a bug in maildirsizefile handling that threw up
6609 "unexpected character" debug warnings, and recalculated the data
6610 unnecessarily. In any case, I expanded the warning message to give more
6613 15. Added the message "Restricted characters in address" to the statements in
6614 the default ACL that block characters like @ and % in local parts.
6616 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
6617 Three changes have been made:
6619 (a) There was a serious bug; a negative response to MAIL caused the whole
6620 recipient domain to be cached as invalid, thereby blocking all messages
6621 to all local parts at the same domain, from all senders. This bug has
6622 been fixed. The domain is no longer cached after a negative response to
6623 MAIL if the sender used is not empty.
6625 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
6628 (c) A new callout option, "use_sender" has been added for people who want
6629 the modified behaviour.
6635 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
6638 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
6639 indeed breaks things for older releases.
6641 3. Added additional logging to the case where there is a problem reading data
6642 from a filter that is running in a subprocess using a pipe, in order to
6643 try to track down a specific problem.
6645 4. Testing facility fudge: when running in the test harness and attempting
6646 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
6647 getting "No route to host". Convert this to a timeout.
6649 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
6652 6. Some OS don't have socklen_t but use size_t instead. This affects the
6653 fifth argument of getsockopt() amongst other things. This is now
6654 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
6655 can be set for individual OS. I have set it for SunOS5, OSF1, and
6656 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
6657 some earlier ones do not.
6659 7. Change 4.30/15 was not doing the test caselessly.
6661 8. The standard form for an IPv6 address literal was being rejected by address
6662 parsing in, for example, MAIL and RCPT commands. An example of this kind of
6663 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6664 this, as well as the form without the "IPv6" on the front (but only when
6665 address literals are enabled, of course).
6667 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
6669 10. Exim crashed if a message with an empty sender address specified by -f
6670 encountered a router with an errors_to setting. This could be provoked only
6671 by a command such as
6675 where an empty string was supplied; "<>" did not hit this bug.
6677 11. Installed PCRE release 4.5.
6679 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
6680 remained set. It is now erased.
6682 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
6683 times from message ids (which are base 36 rather than the normal 62).
6685 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
6686 were being counted as actual protocol errors, and logged if the log
6687 selector +smtp_protocol_error was set. One cannot be perfect in this test,
6688 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
6689 and DATA following a set of rejected RCPTs do not count as protocol errors.
6690 In other words, Exim assumes they were pipelined, though this may not
6691 actually be the case. Of course, in all cases the client gets an
6692 appropriate error code.
6694 15. If a lookup fails in an ACL condition, a message about the failure may
6695 be available; it is used if testing the ACL cannot continue, because most
6696 such messages specify what the cause of the deferral is. However, some
6697 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
6698 that caused an old message to be retained and used if a later statement
6699 caused a defer, replacing the real cause of the deferral.
6701 16. If an IP address had so many PTR records that the DNS lookup buffer
6702 was not large enough to hold them, Exim could crash while trying to process
6703 the truncated data. It now detects and logs this case.
6705 17. Further to 4.21/58, another change has been made: if (and only if) the
6706 first line of a message (the first header line) ends with CRLF, a bare LF
6707 in a subsequent header line has a space inserted after it, so as not to
6708 terminate the header.
6710 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
6711 unnecessarily, used atoi() instead of strtol(), and didn't check the
6712 termination when getting file sizes from file names by regex.
6714 19. Completely re-implemented the support for maildirsize files, in the light
6715 of a number of problems with the previous contributed implementation
6716 (4.30/29). In particular:
6718 . If the quota is zero, the maildirsize file is maintained, but no quota is
6721 . If the maildir directory does not exist, it is created before any attempt
6722 to write a maildirsize file.
6724 . The quota value in the file is just a cache; if the quota is changed in
6725 the transport, the new value overrides.
6727 . A regular expression is available for excluding directories from the
6730 20. The autoreply transport checks the characters in options that define the
6731 message's headers; it allows continued headers, but it was checking with
6732 isspace() after an embedded newline instead of explicitly looking for a
6735 21. If all the "regular" hosts to which an address was routed had passed their
6736 expiry times, and had not reached their retry times, the address was
6737 bounced, even if fallback hosts were defined. Now Exim should go on to try
6740 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
6741 equivalent code in the SMTP transport. Some hosts send humungous responses
6742 to HELO/EHLO, more than 1024 it seems.
6744 23. Refactoring: code in filter.c used (void *) for "any old type" but this
6745 gives compiler warnings in some environments. I've now done it "properly",
6748 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
6749 (because of problems with the built-in one) was declared to return uschar *
6750 instead of char *, causing compiler failure.
6752 25. Fixed a file descriptor leak when processing alias/forward files.
6754 26. Fixed a minor format string issue in dbfn.c.
6756 27. Typo in exim.c: ("dmbnz" for "dbmnz").
6758 28. If a filter file refered to $h_xxx or $message_headers, and the headers
6759 contained RFC 2047 "words", Exim's memory could, under certain conditions,
6762 29. When a sender address is verified, it is cached, to save repeating the test
6763 when there is more than one recipient in a message. However, when the
6764 verification involves a callout, it is possible for different callout
6765 options to be set for different recipients. It is too complicated to keep
6766 track of this in the cache, so now Exim always runs a verification when a
6767 callout is required, relying on the callout cache for the optimization.
6768 The overhead is duplication of the address routing, but this should not be
6771 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
6772 to be verified with callout=postmaster, and the main callout worked but the
6773 postmaster check failed, the verification correctly failed. However, if a
6774 subsequent RCPT command asked for sender verification *without* the
6775 postmaster check, incorrect caching caused this verification also to fail,
6778 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
6779 it was not caching the DNS options (qualify_single, search_parents) that
6780 were used when the lookup failed. A subsequent lookup with different
6781 options therefore always gave the same answer, though there were cases
6782 where it should not have. (Example: a "domains = !$mx_any" option on a
6783 dnslookup router: the "domains" option is always processed without any
6784 widening, but the router might have qualify_single set.) Now Exim uses the
6785 cached value only when the same options are set.
6787 32. Added John Jetmore's "exipick" utility to the distribution.
6789 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
6790 than a timeout (e.g. a certificate is required, and is not provided), an
6791 Exim server now closes the connection immediately. Previously it waited for
6792 the client to close - but if the client is SSL, it seems that they each
6793 wait for each other, leading to a delay before one of them times out.
6795 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
6796 maintained 0.8.x compatibility because I don't think many are using it, and
6797 it is clearly obsolete.
6799 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
6802 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
6803 client certificate was expired. A simple patch fixes this, though I don't
6804 understand the full logic of why the verify callback is called multiple
6807 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
6808 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
6809 which causes problems with some clients (such as the Certicom SSL Plus
6810 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
6811 disables the coutermeasure allowing Eudora to connect."
6813 38. Exim was not checking that a write() to a log file succeeded. This could
6814 lead to Bad Things if a log got too big, in particular if it hit a file
6815 size limit. Exim now panics and dies if it cannot write to a log file, just
6816 as it does if it cannot open a log file.
6818 39. Modified OS/Makefile-Linux so that it now contains
6820 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
6822 The two -D definitions ensure that Exim is compiled with large file
6823 support, which makes it possible to handle log files that are bigger than
6826 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
6827 instance) a domain was checked against a named list that involved a lookup,
6828 causing $domain_data to be set, then another domain was checked against the
6829 same list, then the first domain was re-checked, the value of $domain_data
6830 after the final check could be wrong. In particular, if the second check
6831 failed, it could be set empty. This bug probably also applied to
6834 41. The strip_trailing_dot option was not being applied to the address given
6835 with the -f command-line option.
6837 42. The code for reading a message's header from the spool was incrementing
6838 $received_count, but never initializing it. This meant that the value was
6839 incorrect (doubled) while delivering a message in the same process in which
6840 it was received. In the most common configuration of Exim, this never
6841 happens - a fresh exec is done - but it can happen when
6842 deliver_drop_privilege is set.
6844 43. When Exim logs an SMTP synchronization error - client data sent too soon -
6845 it now includes up to 150 characters of the unexpected data in the log
6848 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6849 and building data strings. The size of both of these buffers was 10 000
6850 bytes - far larger than anybody would *ever* want, thought I. Needless to
6851 say, somebody hit the limit. I have increased the maximum line length to
6852 20 000 and the maximum data length of concatenated lines to 100 000. I have
6853 also fixed two bugs, because there was no checking on these buffers. Tsk,
6854 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6855 buffer is too small.
6857 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6858 lsearch lookups. Now it does.
6860 46. When parsing a route_list item in a manualroute router, a fixed-length
6861 buffer was used for the list of hosts. I made this 1024 bytes long,
6862 thinking that nobody would ever have a list of hosts that long. Wrong.
6863 Somebody had a whole pile of complicated expansion conditions, and the
6864 string was silently truncated, leading to an expansion error. It turns out
6865 that it is easier to change to an unlimited length (owing to other changes
6866 that have happened since this code was originally written) than to build
6867 structure for giving a limitation error. The length of the item that
6868 expands into the list of hosts is now unlimited.
6870 47. The lsearch lookup could not handle data where the length of text line was
6871 more than 4095 characters. Such lines were truncated, leading to shortened
6872 data being returned. It should now handle lines of any length.
6874 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6875 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6878 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6879 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6880 no longer happens in comment lines. A list of replacements is now placed
6881 at the head of all of the source files, except those whose only change is
6882 to replace PERL_COMMAND in the very first #! line.
6884 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6885 messages on the queue, with a bottom-up merge sort, using code contributed
6886 by Michael Haardt. This should make operations like -bp somewhat faster on
6887 large queues. It won't affect queue runners, except when queue_run_in_order
6890 51. Installed eximstats 1.31 in the distribution.
6892 52. Added support for SRV lookups to the dnslookup router.
6894 53. If an ACL referred to $message_body or $message_body_end, the value was not
6895 reset for any messages that followed in the same SMTP session.
6897 54. The store-handling optimization for building very long strings was not
6898 differentiating between the different store pools. I don't think this
6899 actually made any difference in practice, but I've tidied it.
6901 55. While running the routers to verify a sender address, $sender_address
6902 was still set to the sender address. This is wrong, because when routing to
6903 send a bounce to the sender, it would be empty. Therefore, I have changed
6904 it so that, while verifying a sender address, $sender_address is set to <>.
6905 (There is no change to what happens when verifying a recipient address.)
6907 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6908 target A or AAAA records (if not already returned) without resetting the
6909 qualify_single or search_parents options of the DNS resolver. These are
6910 inappropriate in this case because the targets of MX and SRV records must
6911 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6912 target that, when qualified, matched something in the local domain. These
6913 two options are now turned off when doing these lookups.
6915 57. It seems that at least some releases of Reiserfs (which does not have the
6916 concept of a fixed number of inodes) returns zero and not -1 for the
6917 number of available inodes. This interacted badly with check_spool_inodes,
6918 which assumed that -1 was the "no such thing" setting. What I have done is
6919 to check that the total number of inodes is greater than zero before doing
6920 the test of how many are available.
6922 58. When a "warn" ACL statement has a log_message modifier, the message is
6923 remembered, and not repeated. This is to avoid a lot of repetition when a
6924 message has many recipients that cause the same warning to be written.
6925 However, Exim was preserving the list of already written lines for an
6926 entire SMTP session, which doesn't seem right. The memory is now reset if a
6927 new message is started.
6929 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6930 debugging output unless log_rewrite was also set.
6932 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6933 of (int)(handle) when we know that handle contains (void *)(-1).
6935 61. The Exim daemon panic-logs an error return when it closes the incoming
6936 connection. However "connection reset by peer" seems to be common, and
6937 isn't really an error worthy of noting specially, so that particular error
6940 62. When Exim is trying to find all the local interfaces, it used to panic and
6941 die if the ioctl to get the interface flags failed. However, it seems that
6942 on at least one OS (Solaris 9) it is possible to have an interface that is
6943 included in the list of interfaces, but for which you get a failure error
6944 for this call. This happens when the interface is not "plumbed" into a
6945 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6946 failure of the "get flags" call assumes that the interface is down.
6948 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6949 makes life easier for people who are doing arithmetic on fields extracted
6950 from dates, where you often get leading zeros that should not be
6951 interpreted as octal.
6953 64. Added qualify_domain to the redirect router, to override the global
6956 65. If a pathologically long header line contained very many addresses (the
6957 report of this problem mentioned 10 000) and each of them was rewritten,
6958 Exim could use up a very large amount of memory. (It kept on making new
6959 copies of the header line as it rewrote, and never released the old ones.)
6960 At the expense of a bit more processing, the header rewriting function has
6961 been changed so that it no longer eats memory in this way.
6963 66. The generation of the Received: header has been moved from the time that a
6964 message starts to be received, to the time that it finishes. The timestamp
6965 in the Received: header should now be very close to that of the <= log
6966 line. There are two side-effects of this change:
6968 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6969 logged header lines no longer include the local Received: line, because
6970 it has not yet been created. The same applies to a copy of the message
6971 that is returned to a non-SMTP sender when a message is rejected.
6973 (b) When a filter file is tested using -bf, no additional Received: header
6974 is added to the test message. After some thought, I decided that this
6977 This change does not affect the value of $received_for. It is still set
6978 after address rewriting, but before local_scan() is called.
6980 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6982 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6983 gave an unhelpful panic error message, and a defer error. I have managed to
6984 change this behaviour so that it now rejects any supplied certificate,
6985 which seems right, as the list of acceptable certificates is empty.
6987 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6988 gave an unhelpful defer error. I have not managed to make this reject any
6989 supplied certificates, but the error message it gives is "no certificate
6990 supplied", which is not helpful.
6992 70. exigrep's output now also includes lines that are not associated with any
6993 message, but which match the given pattern. Implemented by a patch from
6994 Martin Sluka, which also tidied up the Perl a bit.
6996 71. Recipient callout verification, like sender verification, was using <> in
6997 the MAIL FROM command. This isn't really the right thing, since the actual
6998 sender may affect whether the remote host accepts the recipient or not. I
6999 have changed it to use the actual sender in the callout; this means that
7000 the cache record is now keyed on a recipient/sender pair, not just the
7001 recipient address. There doesn't seem to be a real danger of callout loops,
7002 since a callout by the remote host to check the sender would use <>.
7003 [SEE ABOVE: changed after hitting problems.]
7005 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
7006 temporary errors. However, in the case of such a code being given after
7007 the end of a data transmission (i.e. after ".") Exim was failing to write
7008 a retry record for the message. (Yes, there was some broken host that was
7009 actually sending 8xx at this point.)
7011 73. An unknown lookup type in a host list could cause Exim to panic-die when
7012 the list was checked. (An example that provoked this was putting <; in the
7013 middle of a list instead of at the start.) If this happened during a DATA
7014 ACL check, a -D file could be left lying around. This kind of configuration
7015 error no longer causes Exim to die; instead it causes a defer error. The
7016 incident is still logged to the main and panic logs.
7018 74. Buglet left over from Exim 3 conversion. The message "too many messages
7019 in one connection" was written to the rejectlog but not the mainlog, except
7020 when address rewriting (yes!) was being logged.
7022 75. Added write_rejectlog option.
7024 76. When a system filter was run not as root (that is, when system_filter_user
7025 was set), the values of the $n variables were not being returned to the
7026 main process; thus, they were not subsequently available in the $sn
7029 77. Added +return_path_on_delivery log selector.
7031 78. A connection timeout was being treated differently from recipients deferred
7032 when testing hosts_max_try with a message that was older than the host's
7033 retry timeout. (The host should not be counted, thus allowing all hosts to
7034 be tried at least once before bouncing.) This may have been the cause of an
7035 occasionally reported bug whereby a message would remain on the queue
7036 longer than the retry timeout, but would be bounced if a delivery was
7037 forced. I say "may" because I never totally pinned down the problem;
7038 setting up timeout/retry tests is difficult. See also the next item.
7040 79. The ultimate address timeout was not being applied to errors that involved
7041 a combination of host plus message (for example, a timeout on a MAIL
7042 command). When an address resolved to a number of possible hosts, and they
7043 were not all tried for each delivery (e.g. because of hosts_max_try), a
7044 message could remain on the queue longer than the retry timeout.
7046 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
7049 81. Fixed an obscure SMTP outgoing bug which required at least the following
7050 conditions: (a) there was another message waiting for the same server;
7051 (b) the server returned 5xx to all RCPT commands in the first message so
7052 that the message was not completed; (c) the server dropped the connection
7053 or gave a negative response to the RSET that Exim sends to abort the
7054 transaction. The observed case was a dropped connection after DATA that had
7055 been sent in pipelining mode. That is, the server had advertised PIPELINING
7056 but was not implementing it correctly. The effect of the bug was incorrect
7057 behaviour, such as trying another host, and this could lead to a crash.
7063 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
7064 and daemon.c were passed as pointers to ints; they should have been
7065 pointers to socklen_t variables (which are typically unsigned ints).
7067 2. Some signed/unsigned type warnings in the os.c file for Linux have been
7070 3. Fixed a really odd bug that affected only the testing scheme; patching a
7071 certain fixed string in the binary changed the value of another string that
7072 happened to be identical to the end of the original first string.
7074 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
7075 name", it returns that address as the IP address. On some operating
7076 systems (e.g. Solaris), it also passes back the IP address string as the
7077 "host name". However, on others (e.g. Linux), it passes back an empty
7078 string. Exim wasn't checking for this, and was changing the host name to an
7079 empty string, assuming it had been canonicalized.
7081 5. Although rare, it is permitted to have more than one PTR record for a given
7082 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
7083 all the names associated with an address, because they do in Solaris.
7084 However, it seems that they do not in Linux for data that comes from the
7085 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
7086 I found this out when I moved to a new Linux workstation and tried to run
7087 the Exim test suite.
7089 To get round this problem I have changed the code so that it now does its
7090 own call to the DNS to look up PTR records when searching for a host name.
7091 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
7092 addresses that are only in /etc/hosts are still found.
7094 This behaviour is, however, controlled by an option called host_lookup_
7095 order, which defaults to "bydns:byaddr". If people want to use the other
7096 order, or indeed, just use one or the other means of lookup, they can
7097 specify it in this variable.
7099 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
7100 some operating systems, this comes back from gethostbyaddr() as an empty
7101 string, and this is what Exim used to test for. However, it seems that in
7102 other systems, "." is yielded. Exim now tests for this case too.
7104 7. The values of check_spool_space and check_log_space are now held internally
7105 as a number of kilobytes instead of an absolute number of bytes. If a
7106 numbers is specified without 'K' or 'M', it is rounded up to the nearest
7107 kilobyte. This means that much larger values can be stored.
7109 8. Exim monitor: an attempt to get the action menu when not actually pointing
7110 at a message produces an empty menu entitled "No message selected". This
7111 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
7112 no entries in it ("Shell widget menu has zero width and/or height"). So I
7113 have added a single, blank menu entry in this case.
7115 9. Added ${quote_local_part.
7117 10. MIME decoding is now applied to the contents of Subject: header lines when
7120 11. Now that a reference to $sender_host_address automatically causes a reverse
7121 lookup to occur if necessary (4.13/18), there is no need to arrange for a
7122 host lookup before query-style lookups in lists that might use this
7123 variable. This has therefore been abolished, and the "net-" prefix is no
7124 longer necessary for query-style lookups.
7126 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
7127 have been a typo for LFLAGS, so it has been changed.
7129 13. The install script calls Exim with "-C /dev/null" in order to find the
7130 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
7131 to be output. However, since Exim outputs its version number before the
7132 error, it didn't break the script. It just looked ugly. I fixed this by
7133 always allowing "-C /dev/null" if the caller is root.
7135 14. Ignore overlarge ACL variable number when reading spool file - insurance
7136 against a later release with more variables having written the file.
7138 15. The standard form for an IPv6 address literal was being rejected by EHLO.
7139 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
7140 this, as well as the form without the "IPv6" on the front.
7142 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
7143 OS/Makefile-Darwin file.
7145 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
7146 only to LDAP libraries that do not have LDAP_OPT_DEREF.
7148 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
7149 variable. However, some OS use ints for inodes. I've added cast to long int
7150 to get rid of the compiler warning.
7152 19. I had forgotten to lock out "/../" in configuration file names when
7153 ALT_CONFIG_PREFIX was set.
7155 20. Routers used for verification do not need to specify transports. However,
7156 if such a router generated a host list, and callout was configured, Exim
7157 crashed, because it could not find a port number from the (non-existent)
7158 transport. It now assumes port 25 in this circumstance.
7160 21. Added the -t option to exigrep.
7162 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
7163 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
7164 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
7165 them is set, LOOKUP_LSEARCH is forced.
7167 23. "exim -bV" now outputs a list of lookups that are included in the binary.
7169 24. Added sender and host information to the "rejected by local_scan()" log
7170 line; previously there was no indication of these.
7172 25. Added .include_if_exists.
7174 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
7175 Linux. It turns out that not all file systems support this. Apparently some
7176 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
7177 do it.) To cope with this, the error EINVAL, which means that sync-ing is
7178 not supported on the file descriptor, is now ignored when Exim is trying to
7179 sync a directory. This applies only to Linux.
7181 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
7183 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
7184 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
7187 29. Added support for maildirsize files from supplied patch (modified a bit).
7189 30. The use of :fail: followed by an empty string could lead Exim to respond to
7190 sender verification failures with (e.g.):
7192 550 Verification failed for <xxx>
7193 550 Sender verify failed
7195 where the first response line was missing the '-' that indicates it is not
7196 the final line of the response.
7198 31. The loop for finding the name of the user that called Exim had a hardwired
7199 limit of 10; it now uses the value of finduser_retries, which is used for
7200 all other user lookups.
7202 32. Added $received_count variable, available in data and not_smtp ACLs, and at
7205 33. Exim was neglecting to zero errno before one call of strtol() when
7206 expanding a string and expecting an integer value. On some systems this
7207 resulted in spurious "integer overflow" errors. Also, it was casting the
7208 result into an int without checking.
7210 34. Testing for a connection timeout using "timeout_connect" in the retry rules
7211 did not work. The code looks as if it has *never* worked, though it appears
7212 to have been documented since at least release 1.62. I have made it work.
7214 35. The "timeout_DNS" error in retry rules, also documented since at least
7215 1.62, also never worked. As it isn't clear exactly what this means, and
7216 clearly it isn't a major issue, I have abolished the feature by treating it
7217 as "timeout", and writing a warning to the main and panic logs.
7219 36. The display of retry rules for -brt wasn't always showing the error code
7222 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
7223 timeout_connect_A, timeout_connect_MX.
7225 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
7226 to the empty sender.
7228 39. The daemon was not analysing the content of -oX till after it had closed
7229 stderr and disconnected from the controlling terminal. This meant that any
7230 syntax errors were only noted on the panic log, and the return code from
7231 the command was 0. By re-arranging the code a little, I've made the
7232 decoding happen first, so such errors now appear on stderr, and the return
7233 code is 1. However, the actual setting up of the sockets still happens in
7234 the disconnected process, so errors there are still only recorded on the
7237 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
7238 connections (as happens on some IP stacks) was logged at start up time as
7239 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
7240 it from "IPv6 and IPv4", which means that two separate sockets are being
7243 41. The debug output for gethostbyname2() or getipnodebyname() failures now
7244 says whether AF_INET or AF_INET6 was passed as an argument.
7246 42. Exiwhat output was messed up when time zones were included in log
7249 43. Exiwhat now gives more information about the daemon's listening ports,
7250 and whether -tls-on-connect was used.
7252 44. The "port" option of the smtp transport is now expanded.
7254 45. A "message" modifier in a "warn" statement in a non-message ACL was being
7255 silently ignored. Now an error message is written to the main and panic
7258 46. There's a new ACL modifier called "logwrite" which writes to a log file
7259 as soon as it is encountered.
7261 47. Added $local_user_uid and $local_user_gid at routing time.
7263 48. Exim crashed when trying to verify a sender address that was being
7266 49. Exim was recognizing only a space character after ".include". It now also
7267 recognizes a tab character.
7269 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
7270 extracting the relevant information from the specification. The man page no
7271 longer contains scrambled data for the -d option, and I've added a section
7272 at the front about calling Exim under different names.
7274 51. Added "extra_headers" argument to the "mail" command in filter files.
7276 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
7279 53. Installed eximstats 1.29.
7281 54. Added transport_filter_timeout as a generic transport option.
7283 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
7284 Cc: header lines. This was required by RFC 822, but it not required by RFC
7287 56. Exim used to add From:, Date:, and Message-Id: header lines to any
7288 incoming messages that did not have them. Now it does so only if the
7289 message originates locally, that is, if there is no associated remote host
7290 address. When Resent- header lines are present, this applies to the Resent-
7291 lines rather than the non-Resent- lines.
7293 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
7294 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
7296 58. Messages for configuration errors now include the name of the main
7297 configuration file - useful now that there may be more than one file in a
7298 list (.included file names were always shown).
7300 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
7301 for those rare installations that do not start the daemon as root or run it
7302 setuid root. I've cut out the call to initgroups() if the daemon is not
7305 60. The Exim user and group can now be bound into the binary as text strings
7306 that are looked up at the start of Exim's processing.
7308 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
7310 62. Added $mailstore_basename variable.
7312 63. Installed patch to sieve.c from Michael Haardt.
7314 64. When Exim failed to open the panic log after failing to open the main log,
7315 the original message it was trying to log was written to stderr and debug
7316 output, but if they were not available (the usual case in production), it
7317 was lost. Now it is written to syslog before the two lines that record the
7318 failures to open the logs.
7320 65. Users' Exim filters run in subprocesses under the user's uid. It is
7321 possible for a "deliver" command or an alias in a "personal" command to
7322 provoke an address rewrite. If logging of address rewriting is configured,
7323 this fails because the process is not running as root or exim. There may be
7324 a better way of dealing with this, but for the moment (because 4.30 needs
7325 to be released), I have disabled address rewrite logging when running a
7326 filter in a non-root, non-exim process.
7332 1. The buildconfig auxiliary program wasn't quoting the value set for
7333 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
7334 not defined. This bug was masked in 4.22 by the effect that was fixed in
7337 2. Some messages that were rejected after a message id was allocated were
7338 shown as "incomplete" by exigrep. It no longer does this for messages that
7339 are rejected by local_scan() or the DATA or non-SMTP ACLs.
7341 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
7342 have allow_domain_literals set, the ID did not get logged in the <= line.
7343 Domain literals are now always recognized in Message-ID: header lines.
7345 4. The first argument for a ${extract expansion item is the key name or field
7346 number. Leading and trailing spaces in this item were not being ignored,
7347 causing some misleading effects.
7349 5. When deliver_drop_privilege was set, single queue runner processes started
7350 manually (i.e. by the command "exim -q") or by the daemon (which uses the
7351 same command in the process it spins off) were not dropping privilege.
7353 6. When the daemon running as "exim" started a queue runner, it always
7354 re-executed Exim in the spun-off process. This is a waste of effort when
7355 deliver_drop_privilege is set. The new process now just calls the
7356 queue-runner function directly.
7362 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
7365 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
7366 ignored. Though the use of -oP was forcing the writing of a pid file, it
7367 was always written to the default place.
7369 3. If the message "no IP address found for host xxxx" is generated during
7370 incoming verification, it is now followed by identification of the incoming
7371 connection (so you can more easily find what provoked it).
7373 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
7375 5. Added some features to "harden" Exim a bit more against certain attacks:
7377 (a) There is now a build-time option called FIXED_NEVER_USERS that can
7378 be put in Local/Makefile. This is like the never_users runtime option,
7379 but it cannot be overridden. The default setting is "root".
7381 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
7382 prefix string with which any file named in a -C command line option
7385 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
7386 is retained for -C and -D only if the caller of Exim is root. Without
7387 it, the exim user may also use -C and -D and retain privilege.
7389 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
7390 command line option is disabled.
7392 6. Macro names set by the -D option must start with an upper case letter, just
7393 like macro names defined in the configuration file.
7395 7. Added "dereference=" facility to LDAP.
7397 8. Two instances of the typo "uknown" in the source files are fixed.
7399 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
7400 the Configure-Makefile script screwed up while processing it.
7402 10. Incorporated PCRE 4.4.
7404 11. The SMTP synchronization check was not operating right at the start of an
7405 SMTP session. For example, it could not catch a HELO sent before the client
7406 waited for the greeting. There is now a check for outstanding input at the
7407 point when the greeting is written. Because of the duplex, asynchronous
7408 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
7409 way, but not yet received, when the check is performed.
7411 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
7412 on TCP/IP sockets, because this apparently causes some broken clients to
7415 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
7416 unchanged) from the Cygwin maintainer.
7418 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
7419 was supported instead of testing for mbx. Effectively a typo.
7421 15. The spa authenticator server code was not checking that the input it
7422 received was valid base64.
7424 16. The debug output line for the "set" modifier in ACLs was not showing the
7425 name of the variable that was being set.
7427 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
7429 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
7430 lookup on its own. Something else was needed to trigger the lookup. For
7431 example, a match in host_lookup or the need for a host name in a host list.
7432 Now, if $sender_host_name is referenced and the host name has not yet been
7433 looked up, a lookup is performed. If the lookup fails, the variable remains
7434 empty, and $host_lookup_failed is set to "1".
7436 19. Added "eqi" as a case-independent comparison operator.
7438 20. The saslauthd authentication condition could segfault if neither service
7439 nor realm was specified.
7441 21. If an overflowing value such as "2048M" was set for message_size_limit, the
7442 error message that was logged was misleading, and incoming SMTP
7443 connections were dropped. The message is now more accurate, and temporary
7444 errors are given to SMTP connections.
7446 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
7447 (except RSET) with a 421 error, until QUIT is received. However, it was
7448 failing to send a response to QUIT.
7450 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
7451 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
7452 ACL is now run after the helo_try_verify_hosts code.
7454 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
7455 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
7456 case-independent, so other case variants are also recognized). Apparently
7457 some systems use these upper case variants.
7459 25. If more than two messages were waiting for the same host, and a transport
7460 filter was specified for the transport, Exim sent two messages over the
7461 same TCP/IP connection, and then failed with "socket operation on non-
7462 socket" when it tried to send the third.
7464 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
7466 27. The extern definition of crypt16() in expand.c was not being excluded when
7467 the OS had its own crypt16() function.
7469 28. Added bounce_return_body as a new option, and bounce_return_size_limit
7470 as a preferred synonym for return_size_limit, both as an option and as an
7473 29. Added LIBS=-liconv to OS/Makefile-OSF1.
7475 30. Changed the default configuration ACL to relax the local part checking rule
7476 for addresses that are not in any local domains. For these addresses,
7477 slashes and pipe symbols are allowed within local parts, but the sequence
7478 /../ is explicitly forbidden.
7480 31. SPA server authentication was not clearing the challenge buffer before
7483 32. log_message in a "warn" ACL statement was writing to the reject log as
7484 well as to the main log, which contradicts the documentation and doesn't
7485 seem right (because no rejection is happening). So I have stopped it.
7487 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
7488 However, I am unable to do any testing of this.
7490 34. Fixed an infelicity in the appendfile transport. When checking directories
7491 for a mailbox, to see if any needed to be created, it was accidentally
7492 using path names with one or more superfluous leading slashes; tracing
7493 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
7495 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
7496 contents are added to the log line that is written for every discarded
7497 recipient. (Previously a log_message setting was ignored.)
7499 36. The ${quote: operator now quotes the string if it is empty.
7501 37. The install script runs exim in order to find its version number. If for
7502 some reason other than non-existence or emptiness, which it checks, it
7503 could not run './exim', it was installing it with an empty version number,
7504 i.e. as "exim-". This error state is now caught, and the installation is
7507 38. An argument was missing from the function that creates an error message
7508 when Exim fails to connect to the socket for saslauthd authentication.
7509 This could cause Exim to crash, or give a corrupted message.
7511 39. Added isip, isip4, and isip6 to ${if conditions.
7513 40. The ACL variables $acl_xx are now saved with the message, and can be
7514 accessed later in routers, transports, and filters.
7516 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
7517 strings in the file are not string-expanded.
7519 42. If a MAIL command specified a SIZE value that was too large to fit into an
7520 int variable, the check against message_size_limit failed. Such values are
7521 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
7522 day this will have to be increased, but I don't think I want to be around
7523 when emails are that large.
7530 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
7531 iconv() is not standard in FreeBSD.
7533 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
7534 IPv6 enabled. The observed symptom was a segmentation fault on return from
7535 the function os_common_find_running_interfaces() in src/os.c.
7537 3. In the check_special_case() function in daemon.c I had used "errno" as an
7538 argument name, which causes warnings on some systems. This was basically a
7539 typo, since it was named "eno" in the comments!
7541 4. The code that waits for the clock to tick (at a resolution of some fraction
7542 of a second) so as to ensure message-id uniqueness was always waiting for
7543 at least one whole tick, when it could have waited for less. [This is
7544 almost certainly not relevant at current processor speeds, where it is
7545 unlikely to ever wait at all. But we try to future-proof.]
7547 5. The function that sleeps for a time interval that includes fractions of a
7548 second contained a race. It did not block SIGALRM between setting the
7549 timer, and suspending (a couple of lines later). If the interval was short
7550 and the sigsuspend() was delayed until after it had expired, the suspension
7551 never ended. On busy systems this could lead to processes getting stuck for
7554 6. Some uncommon configurations may cause a lookup to happen in a queue runner
7555 process, before it forks any delivery processes. The open lookup caching
7556 mechanism meant that the open file or database connection was passed into
7557 the delivery process. The problem was that delivery processes always tidy
7558 up cached lookup data. This could cause a problem for the next delivery
7559 process started by the queue runner, because the external queue runner
7560 process does not know about the closure. So the next delivery process
7561 still has data in the lookup cache. In the case of a file lookup, there was
7562 no problem because closing a file descriptor in a subprocess doesn't affect
7563 the parent. However, if the lookup was caching a connection to a database,
7564 the connection was closed, and the second delivery process was likely to
7565 see errors such as "PGSQL: query failed: server closed the connection
7566 unexpectedly". The problem has been fixed by closing all cached lookups
7567 in a queue runner before running a delivery process.
7569 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
7570 seem to have the "const" qualifier which it has on other OS. I've
7573 8. Change 4.21/2 was too strict. It is only if there are two authenticators
7574 *of the same type* (client or server) with the same public name that an
7575 error should be diagnosed.
7577 9. When Exim looked up a host name for an IP address, but failed to find the
7578 original IP address when looking up the host name (a safety check), it
7579 output the message "<ip address> does not match any IP for NULL", which was
7580 confusing, to say the least. The bug was that the host name should have
7581 appeared instead of "NULL".
7583 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
7584 user that is built into the binary, and the -C or -D options is used, root
7585 privilege is dropped before the configuration file is read. In addition,
7586 logging is switched to stderr instead of the normal log files. If the
7587 configuration then re-defines the Exim user, the unprivileged environment
7588 is probably not what is expected, so Exim logs a panic warning message (but
7591 However, if deliver_drop_privilege is set, the unprivileged state may well
7592 be exactly what is intended, so the warning has been cut out in that case,
7593 and Exim is allowed to try to write to its normal log files.
7599 1. smtp_return_error_details was not giving details for temporary sender
7600 or receiver verification errors.
7602 2. Diagnose a configuration error if two authenticators have the same public
7605 3. Exim used not to create the message log file for a message until the first
7606 delivery attempt. This could be confusing when incoming messages were held
7607 for policy or load reasons. The message log file is now created at the time
7608 the message is received, and an initial "Received" line is written to it.
7610 4. The automatically generated man page for command line options had a minor
7611 bug that caused no ill effects; however, a more serious problem was that
7612 the procedure for building the man page automatically didn't always
7613 operate. Consequently, release 4.20 contains an out-of-date version. This
7614 shouldn't happen again.
7616 5. When building Exim with embedded Perl support, the script that builds the
7617 Makefile was calling 'perl' to find its compile-time parameters, ignoring
7618 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
7620 6. The freeze_tell option was not being used for messages that were frozen on
7621 arrival, either by an ACL or by local_scan().
7623 7. Added the smtp_incomplete_transaction log selector.
7625 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
7626 was accepting AUTH without a new EHLO.
7628 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
7629 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
7632 10. Logging of TCP/IP connections (when configured) now happens in the main
7633 daemon process instead of the child process, so that the TCP/IP connection
7634 count is more accurate (but it can never be perfect).
7636 11. The use of "drop" in a nested ACL was not being handled correctly in the
7637 outer ACL. Now, if condition failure induced by the nested "drop" causes
7638 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
7639 or "require"), the connection is dropped.
7641 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
7642 that yield "discard" can now be used with an "accept" or a "discard" verb,
7643 but an error is generated for any others (because I can't see a useful way
7644 to define what should happen).
7646 13. When an ACL is read dynamically from a file (or anywhere else), the lines
7647 are now processed in the same way as lines in the Exim configuration file.
7648 In particular, continuation lines are supported.
7650 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
7652 15. Added -ti meaning -t -i.
7654 16. Check for letters, digits, hyphens, and dots in the names of dnslist
7655 domains, and warn by logging if others are found.
7657 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
7658 returned from GIFCONF when Exim is trying to find the list of interfaces on
7659 a host. The code in os.c has been modified to copy each ifreq to an aligned
7660 structure in all cases.
7662 Also, in some cases, the returned ifreq's were being copied to a 'struct
7663 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
7664 means the last couple of bytes of an IPv6 address could be chopped if the
7665 ifreq contained only a normal sockaddr (14 bytes storage).
7667 18. Named domain lists were not supported in the hosts_treat_as_local option.
7668 An entry such as +xxxx was not recognized, and was treated as a literal
7671 19. Ensure that header lines added by a DATA ACL are included in the reject log
7672 if the ACL subsequently rejects the message.
7674 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
7675 MD5 (which is deprecated).
7677 21. When testing a filter file using -bf, Exim was writing a message when it
7678 took the sender from a "From " line in the message, but it was not doing so
7679 when it took $return_path from a Return-Path: header line. It now does.
7681 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
7682 with a valid header line field name (a series of printing characters
7683 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
7685 23. Changed "disc" in the source to "disk" to conform to the documentation and
7686 the book and for uniformity.
7688 24. Ignore Sendmail's -Ooption=value command line item.
7690 25. When execve() failed while trying to run a command in a pipe transport,
7691 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
7692 could be confused with a return value of 69 from the command itself. This
7693 has been changed to 127, the value the shell returns if it is asked to run
7694 a non-existent command. The wording for the related log line suggests a
7695 non-existent command as the problem.
7697 26. If received_header_text expands to an empty string, do not add a Received:
7698 header line to the message. (Well, it adds a token one on the spool, but
7699 marks it "old" so that it doesn't get used or transmitted.)
7701 27. Installed eximstats 1.28 (addition of -nt option).
7703 28. There was no check for failure on the call to getsockname() in the daemon
7704 code. This can fail if there is a shortage of resources on the system, with
7705 ENOMEM, for example. A temporary error is now given on failure.
7707 29. Contrary to the C standard, it seems that in some environments, the
7708 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
7709 program. Exim now does this explicitly; it affects the formatting of
7710 timestamps using strftime().
7712 30. If exiqsumm was given junk data, it threw up some uninitialized variable
7713 complaints. I've now initialized all the variables, to avoid this.
7715 32. Header lines added by a system filter were not being "seen" during
7716 transport-time rewrites.
7718 33. The info_callback() function passed to OpenSSL is set up with type void
7719 (*)(SSL *, int, int), as described somewhere. However, when calling the
7720 function (actually a macro) that sets it up, the type void(*)() is
7721 expected. I've put in a cast to prevent warnings from picky compilers.
7723 34. If a DNS black list lookup found a CNAME record, but there were no A
7724 records associated with the domain it pointed at, Exim crashed.
7726 35. If a DNS black list lookup returned more than one A record, Exim ignored
7727 all but the first. It now scans all returned addresses if a particular IP
7728 value is being sought. In this situation, the contents of the
7729 $dnslist_value variable are a list of all the addresses, separated by a
7732 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
7733 to accept a host name and all its aliases if the forward lookup for any of
7734 them yielded the IP address of the incoming connection. Now it accepts only
7735 those names whose forward lookup yields the correct IP address. Any other
7736 names are discarded. This closes a loophole whereby a rogue DNS
7737 administrator could create reverse DNS records to break through a
7738 wildcarded host restriction in an ACL.
7740 37. If a user filter or a system filter that ran in a subprocess used any of
7741 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
7742 the wrong values were passed to the pipe command ($thisaddress had the
7743 value of $0, $0 had the value of $1, etc). This bug was introduced by
7744 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
7746 38. Improved the line breaking for long SMTP error messages from ACLs.
7747 Previously, if there was no break point between 40 and 75 characters, Exim
7748 left the rest of the message alone. Two changes have been made: (a) I've
7749 reduced the minimum length to 35 characters; (b) if it can't find a break
7750 point between 35 and 75 characters, it looks ahead and uses the first one
7751 that it finds. This may give the occasional overlong line, but at least the
7752 remaining text gets split now.
7754 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
7755 file descriptors might be low, and that setting 1000 would always raise it.
7756 It turns out that in some environments, the limit is already over 1000 and
7757 that lowering it causes trouble. So now Exim takes care not to decrease it.
7759 40. When delivering a message, the value of $return_path is set to $sender_
7760 address at the start of routing (routers may change the value). By an
7761 oversight, this default was not being set up when an address was tested by
7762 -bt or -bv, which affected the outcome if any router or filter referred to
7765 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
7766 log only when "message" or "log_message" are set. However, if one of the
7767 conditions was an address verification, or a call to a nested ACL, the
7768 messages generated by the underlying test were being passed through. This
7769 no longer happens. The underlying message is available in $acl_verify_
7770 message for both "message" and "log_message" expansions, so it can be
7771 passed through if needed.
7773 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
7774 new expansion $bh_ to give the encoded byte string without charset
7775 translation. Translation happens only if iconv() is available; HAVE_ICONV
7776 indicates this at build time. HEADERS_CHARSET gives the charset to
7777 translate to; headers_charset can change it in the configuration, and
7778 "headers charset" can change it in an individual filter file.
7780 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
7781 that creates RFC 2047 encoded "words" labels them as that charset instead
7782 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
7783 expansion operator; (ii) when Exim creates a From: line for a local
7784 message; (iii) when a header line is rewritten to include a "phrase" part.
7786 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
7787 buggy, causing it to skip the first lines of messages whose message ID
7788 ended in 'D'. This would not have bitten before Exim release 4.14, because
7789 message IDs were unlikely to end in 'D' before then. The effect was to have
7790 incorrect size information for certain domains.
7792 45. #include "config.h" was missing at the start of the crypt16.c module. This
7793 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
7796 46. If there was a timeout during a "random" callout check, Exim treated it as
7797 a failure of the random address, and carried on sending RSET and the real
7798 address. If the delay was just some slowness somewhere, the response to the
7799 original RCPT would be taken as a response to RSET and so on, causing
7800 mayhem of various kinds.
7802 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
7803 when I implemented it. It didn't allow for the fact that some option values
7804 may legitimately be negative (e.g. size_addition), and it didn't even do
7805 the right test for positive values.
7807 48. Domain names in DNS records are case-independent. Exim always looks them up
7808 in lower case. Some resolvers return domain names in exactly the case they
7809 appear in the zone file, that is, they may contain uppercase letters. Not
7810 all resolvers do this - some return always lower case. Exim was treating a
7811 change of case by a resolver as a change of domain, similar to a widening
7812 of a domain abbreviation. This triggered its re-routing code and so it was
7813 trying to route what was effectively the same domain again. This normally
7814 caused routing to fail (because the router wouldn't handle the domain
7815 twice). Now Exim checks for this case specially, and just changes the
7816 casing of the domain that it ultimately uses when it transmits the message
7819 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
7822 50. If a filter generated a file delivery with a non-absolute name (possible if
7823 no home directory exists for the router), the forbid_file option was not
7826 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
7827 the existing equality matching.
7829 52. Exim was using ints instead of ino_t variables in some places where it was
7830 dealing with inode numbers.
7832 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
7833 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
7834 called TMPDIR, and if it finds it is different, it changes its value.
7836 54. The smtp_printf() function is now made available to local_scan() so
7837 additional output lines can be written before returning. There is also an
7838 smtp_fflush() function to enable the detection of a dropped connection.
7839 The variables smtp_input and smtp_batched_input are exported to
7842 55. Changed the default runtime configuration: the message "Unknown user"
7843 has been removed from the ACL, and instead placed on the localuser router,
7844 using the cannot_route_message feature. This means that any verification
7845 failures that generate their own messages won't get overridden. Similarly,
7846 the "Unrouteable address" message that was in the ACL for unverifiable
7847 relay addresses has also been removed.
7849 56. Added hosts_avoid_esmtp to the smtp transport.
7851 57. The exicyclog script was not checking for the esoteric option
7852 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7853 will work only if exicyclog is run under the appropriate euid.
7855 58. Following a discussion on the list, the rules by which Exim recognises line
7856 endings on incoming messages have been changed. The -dropcr and drop_cr
7857 options are now no-ops, retained only for backwards compatibility. The
7858 following line terminators are recognized: LF CRLF CR. However, special
7859 processing applies to CR:
7861 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7862 nor a local message in the state where . is a terminator.
7864 (ii) If a bare CR is encountered in a header line, an extra space is added
7865 after the line terminator so as not to end the header. The reasoning
7866 behind this is that bare CRs in header lines are most likely either
7867 to be mistakes, or people trying to play silly games.
7869 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7870 was being incorrectly given as 18 bytes larger than it should have been.
7871 This is a VOB (very old bug).
7873 60. This may never have affected anything current, but just in case it has:
7874 When the local host is found other than at the start of a list of hosts,
7875 the local host, those with the same MX, and any that follow, are discarded.
7876 When the list in question was part of a longer list of hosts, the following
7877 hosts (not currently being processed) were also being discarded. This no
7878 longer happens. I'm not sure if this situation could ever has previously
7881 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7884 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7885 header. This is something that is recommended in a new Internet Draft, and
7886 is something that is documented as being done by Sendmail. There are two
7887 possible values. For messages generated by the autoreply transport, Exim
7890 Auto-Submitted: auto-replied
7892 whereas for all other generated messages (e.g. bounces) it adds
7894 Auto-Submitted: auto-generated
7896 63. The "personal" condition in filters now includes a test for the
7897 Auto-Submitted: header. If it contains the string "auto-" the message it
7898 not considered personal.
7900 64. Added rcpt_include_affixes as a generic transport option.
7902 65. Added queue_only_override (default true).
7904 66. Added the syslog_duplication option.
7906 67. If what should have been the first header line of a message consisted of
7907 a space followed by a colon, Exim was mis-interpreting it as a header line.
7908 It isn't of course - it is syntactically invalid and should therefore be
7909 treated as the start of the message body. The misbehaviour could have
7910 caused a number of strange effects, including loss of data in subsequent
7911 header lines, and spool format errors.
7913 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7914 client host had authenticated. This control can now be exercised by an ACL
7915 for more flexibility.
7917 69. By default, callouts do not happen when testing with -bh. There is now a
7918 variant, -bhc, which does actually run the callout code, including
7919 consulting and updating the callout cache.
7921 70. Added support for saslauthd authentication, courtesy of Alexander
7924 71. If statvfs() failed on the spool or log directories while checking their
7925 size for availability, Exim confusingly gave the error "space shortage".
7926 Furthermore, in debugging mode it crashed with a floating point exception.
7927 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7928 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7929 serious problem, Exim now writes to the main and panic logs when this
7930 happens, with details of the failure. It then refuses to accept the
7931 incoming message, giving the message "spool directory problem" or "log
7932 directory problem" with a 421 code for SMTP messages.
7934 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7935 0, 1, and 2 exist, because some OS complain for execs without them (see
7936 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7937 descriptors. However, the code omitted to check that the open succeeded,
7938 causing mysterious errors if for some reason the permissions on /dev/null
7939 got screwed. Now Exim writes a message to the main and panic logs, and
7940 bombs out if it can't open /dev/null.
7942 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7943 interact so that it is all more flexible. It is supposed to remain
7944 backwards compatible. Also added extra_local_interfaces.
7946 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7947 to bomb out with an assertion failure - to the client this appears as a
7948 connection drop. This problem occurs in the part of the code that was taken
7949 from the Samba project. Fortunately, the assertion is in a very simple
7950 function, so I have fixed this by reproducing the function inline in the
7951 one place where it is called, and arranging for authentication to fail
7952 instead of killing the process with assert().
7954 75. The SPA client code was not working when the server requested OEM rather
7955 than Unicode encoding.
7957 76. Added code to make require_files with a specific uid setting more usable in
7958 the case where statting the file as root fails - usually a non-root-mounted
7959 NFS file system. When this happens and the failure is EACCES, Exim now
7960 forks a subprocess and does the per-uid checking as the relevant uid.
7962 77. Added process_log_path.
7964 78. If log_file_path was not explicitly set, a setting of check_log_space or
7965 check_log_inodes was ignored.
7967 79. If a space check for the spool or log partitions fails, the incident is now
7968 logged. Of course, in the latter case the data may get lost...
7970 80. Added the %p formatting code to string_format() so that it can be used to
7971 print addresses in debug_print(). Adjusted all the address printing in the
7972 debugging in store.c to use %p rather than %d.
7974 81. There was a concern that a line of code in smtp_in.c could overflow a
7975 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7976 initially expressed, the concern was not well-founded, because trailing
7977 spaces are removed early. However, if the trailing spaces were followed by
7978 a NULL, they did not get removed, so the overflow was possible. Two fixes
7981 (a) I re-wrote the offending code in a cleaner fashion.
7982 (b) If an incoming SMTP command contains a NULL character, it is rejected
7985 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7986 runs initgroups(), so that if the Exim user is in any additional groups,
7987 they will be used during message reception.
7993 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob