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).
207 JH/01 Avoid costly startup code when not strictly needed. This reduces time
208 for some exim process initialisations. It does mean that the logging
209 of TLS configuration problems is only done for the daemon startup.
211 JH/02 Early-pipelining support code is now included unless disabled in Makefile.
213 JH/03 DKIM verification defaults no long accept sha1 hashes, to conform to
214 RFC 8301. They can still be enabled, using the dkim_verify_hashes main
217 JH/04 Support CHUNKING from an smtp transport using a transport_filter, when
218 DKIM signing is being done. Previously a transport_filter would always
219 disable CHUNKING, falling back to traditional DATA.
221 JH/05 Regard command-line recipients as tainted.
223 JH/06 Bug 340: Remove the daemon pid file on exit, when due to SIGTERM.
225 JH/07 Bug 2489: Fix crash in the "pam" expansion condition. It seems that the
226 PAM library frees one of the arguments given to it, despite the
227 documentation. Therefore a plain malloc must be used.
229 JH/08 Bug 2491: Use tainted buffers for the transport smtp context. Previously
230 on-stack buffers were used, resulting in a taint trap when DSN information
231 copied from a received message was written into the buffer.
233 JH/09 Bug 2493: Harden ARC verify against Outlook, whick has been seen to mix
234 the ordering of its ARC headers. This caused a crash.
236 JH/10 Bug 2492: Use tainted memory for retry record when needed. Previously when
237 a new record was being constructed with information from the peer, a trap
240 JH/11 Bug 2494: Unset the default for dmarc_tld_file. Previously a naiive
241 installation would get error messages from DMARC verify, when it hit the
242 nonexistent file indicated by the default. Distros wanting DMARC enabled
243 should both provide the file and set the option.
244 Also enforce no DMARC verification for command-line sourced messages.
246 JH/12 Fix an uninitialised flag in early-pipelining. Previously connections
247 could, depending on the platform, hang at the STARTTLS response.
249 JH/13 Bug 2498: Reset a counter used for ARC verify before handling another
250 message on a connection. Previously if one message had ARC headers and
251 the following one did not, a crash could result when adding an
252 Authentication-Results: header.
254 JH/14 Bug 2500: Rewind some of the common-coding in string handling between the
255 Exim main code and Exim-related utities. The introduction of taint
256 tracking also did many adjustments to string handling. Since then, eximon
257 frequently terminated with an assert failure.
259 JH/15 When PIPELINING, synch after every hundred or so RCPT commands sent and
260 check for 452 responses. This slightly helps the inefficieny of doing
261 a large alias-expansion into a recipient-limited target. The max_rcpt
262 transport option still applies (and at the current default, will override
263 the new feature). The check is done for either cause of synch, and forces
264 a fast-retry of all 452'd recipients using a new MAIL FROM on the same
265 connection. The new facility is not tunable at this time.
267 JH/16 Fix the variables set by the gsasl authenticator. Previously a pointer to
268 library live data was being used, so the results became garbage. Make
269 copies while it is still usable.
271 JH/17 Logging: when the deliver_time selector ise set, include the DT= field
272 on delivery deferred (==) and failed (**) lines (if a delivery was
273 attemtped). Previously it was only on completion (=>) lines.
275 JH/18 Authentication: the gsasl driver not provides the $authN variables in time
276 for the expansion of the server_scram_iter and server_scram_salt options.
278 WB/01 SPF: DNS lookups for the obsolete SPF RR type done by the libspf2 library
279 are now specifically given a NO_DATA response without hitting the system
280 resolver. The library goes on to do the now-standard TXT lookup.
281 Use of dnsdb lookups is not affected.
283 JH/19 Bug 2507: Modules: on handling a dynamic-module (lookups) open failure,
284 only retrieve the errormessage once. Previously two calls to dlerror()
285 were used, and the second one (for mainlog/paniclog) retrieved null
288 JH/20 Taint checking: disallow use of tainted data for
289 - the appendfile transport file and directory options
290 - the pipe transport command
291 - the autoreply transport file, log and once options
292 - file names used by the redirect router (including filter files)
294 - paths used by single-key lookups
295 Previously this was permitted.
297 JH/21 Bug 2501: Fix init call in the heimdal authenticator. Previously it
298 adjusted the size of a major service buffer; this failed because the
299 buffer was in use at the time. Change to a compile-time increase in the
300 buffer size, when this authenticator is compiled into exim.
302 JH/22 Taint-checking: move to safe-mode taint checking on all platforms. The
303 previous fast-mode was untenable in the face of glibs using mmap to
304 support larger malloc requests.
306 PP/01 Update the openssl_options possible values through OpenSSL 1.1.1c.
307 New values supported, if defined on system where compiled:
308 allow_no_dhe_kex, cryptopro_tlsext_bug, enable_middlebox_compat,
309 no_anti_replay, no_encrypt_then_mac, prioritize_chacha, tlsext_padding
311 JH/23 Performance improvement in the initial phase of a two-pass queue run. By
312 running a limited number of proceses in parallel, a benefit is gained. The
313 amount varies with the platform hardware and load. The use of the option
314 queue_run_in_order means we cannot do this, as ordering becomes
317 JH/24 Bug 2524: fix the cyrus_sasl auth driver gssapi usage. A previous fix
318 had introduced a string-copy (for ensuring NUL-termination) which was not
319 appropriate for that case, which can include embedded NUL bytes in the
320 block of data. Investigation showed the copy to actually be needless, the
321 data being length-specified.
323 JH/25 Fix use of concurrent TLS connections under GnuTLS. When a callout was
324 done during a receiving connection, and both used TLS, global info was
325 used rather than per-connection info for tracking the state of data
326 queued for transmission. This could result in a connection hang.
328 JH/26 Fix use of the SIZE parameter on MAIL commands, on continued connections.
329 Previously, when delivering serveral messages down a single connection
330 only the first would provide a SIZE. This was due to the size information
331 not being properly tracked.
333 JH/27 Bug 2530: When operating in a timezone with sub-minute offset, such as
334 TAI (at 37 seconds currently), pretend to be in UTC for time-related
335 expansion and logging. Previously, spurious values such as a future
336 minute could be seen.
338 JH/28 Bug 2533: Fix expansion of ${tr } item. When called in some situations
339 it could crash from a null-deref. This could also affect the
340 ${addresses: } operator and ${readsock } item.
342 JH/29 Bug 2537: Fix $mime_part_count. When a single connection had a non-mime
343 message following a mime one, the variable was not reset.
345 JH/30 When an pipelined-connect fails at the first response, assume incorrect
346 cached capability (perhaps the peer reneged?) and immediately retry in
349 JH/31 Fix spurious detection of timeout while writing to transport filter.
351 JH/32 Bug 2541: Fix segfault on bad cmdline -f (sender) argument. Previously
352 an attempt to copy the string was made before checking it.
354 JH/33 Fix the dsearch lookup to return an untainted result. Previously the
355 taint of the lookup key was maintained; we now regard the presence in the
356 filesystem as sufficient validation.
358 JH/34 Fix the readsocket expansion to not segfault when an empty "options"
359 argument is supplied.
361 JH/35 The dsearch lookup now requires that the directory is an absolute path.
362 Previously this was not checked, and nonempty relative paths made an
363 access under Exim's current working directory.
365 JH/36 Bug 2554: Fix msg:defer event for the hosts_max_try_hardlimit case.
366 Previously no event was raised.
368 JH/37 Bug 2552: Fix the check on spool space during reception to use the SIZE
369 parameter supplied by the sender MAIL FROM command. Previously it was
370 ignored, and only the check_spool_space option value for the required
373 JH/38 Fix $dkim_key_length. This should, after a DKIM verification, present
374 the size of the signing public-key. Previously it was instead giving
375 the size of the signature hash.
377 JH/39 DKIM verification: the RFC 8301 restriction on sizes of RSA keys is now
378 the default. See the (new) dkim_verify_min_keysizes option.
380 JH/40 Fix a memory-handling bug: when a connection carried multiple messages
381 and an ACL use a lookup for checking either the local_part or domain,
382 stale data could be accessed. Ensure that variable references are
383 dropped between messages.
385 JH/41 Bug 2571: Fix SPA authenticator. Running as a server, an offset supplied
386 by the client was not checked as pointing within response data before
387 being used. A malicious client could thus cause an out-of-bounds read and
388 possibly gain authentication. Fix by adding the check.
390 JH/42 Internationalisation: change the default for downconversion in the smtp
391 transport to be "if needed". Previously it was "as previously set" for
392 the message, which usually meant "if needed" for message-submission but
393 "no" for everything else. However, MTAs have been seen using SMTPUTF8
394 even when the envelope addresses did not need it, resulting in forwarding
395 failures to non-supporting MTAs. A downconvert in such cases will be
396 a no-op on the addresses, merely dropping the use of SMTPUTF8 by the
397 transport. The change does mean that addresses needing conversion will
398 be converted when previously a delivery failure would occur.
400 JH/43 Fix possible long line in DSN. Previously when a very long SMTP error
401 response was received it would be used unchecked in a fail-DSN, violating
402 standards on line-length limits. Truncate if needed.
404 HS/01 Remove parameters of the link to www.open-spf.org. The linked form
405 doesn't work. (Additionally add a new main config option to configure the
412 JH/01 OpenSSL: With debug enabled output keying information sufficient, server
413 side, to decode a TLS 1.3 packet capture.
415 JH/02 OpenSSL: Suppress the sending of (stateful) TLS1.3 session tickets.
416 Previously the default library behaviour applied, sending two, each in
419 JH/03 Debug output for ACL now gives the config file name and line number for
422 JH/04 The default received_header_text now uses the RFC 8314 tls cipher clause.
424 JH/05 DKIM: ensure that dkim_domain elements are lowercased before use.
426 JH/06 Fix buggy handling of autoreply bounce_return_size_limit, and a possible
427 buffer overrun for (non-chunking) other transports.
429 JH/07 GnuTLS: Our use of late (post-handshake) certificate verification, under
430 TLS1.3, means that a server rejecting a client certificate is not visible
431 to the client until the first read of encrypted data (typically the
432 response to EHLO). Add detection for that case and treat it as a failed
433 TLS connection attempt, so that the normal retry-in-clear can work (if
434 suitably configured).
436 JB/01 Bug 2375: fix expansions of 822 addresses having comments in local-part
437 and/or domain. Found and fixed by Jason Betts.
439 JH/08 Add hardening against SRV & TLSA lookups the hit CNAMEs (a nonvalid
440 configuration). If a CNAME target was not a wellformed name pattern, a
443 JH/09 Logging: Fix initial listening-on line for multiple ports for an IP when
444 the OS reports them interleaved with other addresses.
446 JH/10 OpenSSL: Fix aggregation of messages. Previously, when PIPELINING was
447 used both for input and for a verify callout, both encrypted, SMTP
448 responses being sent by the server could be lost. This resulted in
449 dropped connections and sometimes bounces generated by a peer sending
452 JH/11 Harden plaintext authenticator against a badly misconfigured client-send
453 string. Previously it was possible to cause undefined behaviour in a
454 library routine (usually a crash). Found by "zerons".
456 JH/12 Bug 2384: fix "-bP smtp_receive_timeout". Previously it returned no
459 JH/13 Bug 2386: Fix builds with Dane under LibreSSL 2.9.0 onward. Some old
460 API was removed, so update to use the newer ones.
462 JH/14 Bug 1891: Close the log file if receiving a non-smtp message, without
463 any timeout set, is taking a long time. Previously we would hang on to a
464 rotated logfile "forever" if the input was arriving with long gaps
465 (a previous attempt to fix addressed lack, for a long time, of initial
468 HS/01 Bug 2390: Use message_id for tempfile creation to avoid races in a
469 shared (NFS) environment. The length of the tempfile name is now
470 4 + 16 ("hdr.$message_exim_id") which might break on file
471 systems which restrict the file name length to lower values.
474 HS/02 Bug 2390: Use message_id for tempfile creation to avoid races in a
475 shared (NFS) environment.
477 HS/03 Bug 2392: exigrep does case sensitive *option* processing (as it
478 did for all versions <4.90). Notably -M, -m, --invert, -I may be
481 JH/15 Use unsigned when creating bitmasks in macros, to avoid build errors
482 on some platforms for bit 31.
484 JH/16 GnuTLS: rework ciphersuite strings under recent library versions. Thanks
485 to changes apparently associated with TLS1.3 handling some of the APIs
486 previously used were either nonfunctional or inappropriate. Strings
487 like TLS1.3:ECDHE_SECP256R1__RSA_PSS_RSAE_SHA256__AES_256_GCM__AEAD:256
488 and TLS1.2:ECDHE_SECP256R1__RSA_SHA256__AES_128_CBC__SHA256:128 replace
489 the previous TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256 .
490 This affects log line X= elements, the $tls_{in,out}_cipher variables,
491 and the use of specific cipher names in the encrypted= ACL condition.
493 JH/17 OpenSSL: the default openssl_options now disables ssl_v3.
495 JH/18 GnuTLS: fix $tls_out_ocsp under hosts_request_ocsp. Previously the
496 verification result was not updated unless hosts_require_ocsp applied.
498 JH/19 Bug 2398: fix listing of a named-queue. Previously, even with the option
499 queue_list_requires_admin set to false, non-admin users were denied the
502 JH/20 Bug 2389: fix server advertising of usable certificates, under GnuTLS in
503 directory-of-certs mode. Previously they were advertised despite the
506 JH/21 The smtp transport option "hosts_noproxy_tls" is now unset by default.
507 A single TCP connection by a client will now hold a TLS connection open
508 for multiple message deliveries, by default. Previously the default was to
511 JH/22 The smtp transport option "hosts_try_dane" now enables all hosts by
512 default. If built with the facility, DANE will be used. The facility
513 SUPPORT_DANE is now enabled in the prototype build Makefile "EDITME".
515 JH/23 The build default is now for TLS to be included; the SUPPORT_TLS define
516 is replaced with DISABLE_TLS. Either USE_GNUTLS or (the new) USE_OPENSSL
517 must be defined and you must still, unless you define DISABLE_TLS, manage
518 the the include-dir and library-file requirements that go with that
519 choice. Non-TLS builds are still supported.
521 JH/24 Fix duplicated logging of peer name/address, on a transport connection-
524 JH/25 The smtp transport option "hosts_try_fastopen" now enables all hosts by
525 default. If the platform supports and has the facility enabled, it will
526 be requested on all coneections.
528 JH/26 The PIPE_CONNECT facility is promoted from experimental status and is now
529 controlled by the build-time option SUPPORT_PIPE_CONNECT.
531 PP/01 Unbreak heimdal_gssapi, broken in 4.92.
533 JH/27 Bug 2404: Use the main-section configuration option "dsn_from" for
534 success-DSN messages. Previously the From: header was always the default
535 one for these; the option was ignored.
537 JH/28 Fix the timeout on smtp response to apply to the whole response.
538 Previously it was reset for every read, so a teergrubing peer sending
539 single bytes within the time limit could extend the connection for a
540 long time. Credit to Qualsys Security Advisory Team for the discovery.
542 JH/29 Fix DSN Final-Recipient: field. Previously it was the post-routing
543 delivery address, which leaked information of the results of local
544 forwarding. Change to the original envelope recipient address, per
547 JH/30 Bug 2411: Fix DSN generation when RFC 3461 failure notification is
548 requested. Previously not bounce was generated and a log entry of
549 error ignored was made.
551 JH/31 Avoid re-expansion in ${sort } expansion. (CVE-2019-13917)
553 JH/32 Introduce a general tainting mechanism for values read from the input
554 channel, and values derived from them. Refuse to expand any tainted
555 values, to catch one form of exploit.
557 JH/33 Bug 2413: Fix dkim_strict option. Previously the expansion result
558 was unused and the unexpanded text used for the test. Found and
559 fixed by Ruben Jenster.
561 JH/34 Fix crash after TLS shutdown. When the TCP/SMTP channel was left open,
562 an attempt to use a TLS library read routine dereffed a nul pointer,
565 JH/35 Bug 2409: filter out-of-spec chars from callout response before using
566 them in our smtp response.
568 JH/36 Have the general router option retry_use_local_part default to true when
569 any of the restrictive preconditions are set (to anything). Previously it
570 was only for check_local user. The change removes one item of manual
571 configuration which is required for proper retries when a remote router
572 handles a subset of addresses for a domain.
574 JH/37 Appendfile: when evaluating quota use (non-quota_size_regex) take the file
575 link count into consideration.
577 HS/04 Fix handling of very log lines in -H files. If a -<key> <value> line
578 caused the extension of big_buffer, the following lines were ignored.
580 JH/38 Bug 1395: Teach the DNS negative-cache about TTL value from the SOA in
581 accordance with RFC 2308. Previously there was no expiry, so a longlived
582 receive process (eg. due to ACL delays) versus a short SOA value could
585 HS/05 Handle trailing backslash gracefully. (CVE-2019-15846)
587 JH/39 Promote DMARC support to mainline.
589 JH/40 Bug 2452: Add a References: header to DSNs.
591 JH/41 With GnuTLS 3.6.0 (and later) do not attempt to manage Diffie-Hellman
592 parameters. The relevant library call is documented as "Deprecated: This
593 function is unnecessary and discouraged on GnuTLS 3.6.0 or later. Since
594 3.6.0, DH parameters are negotiated following RFC7919."
596 HS/06 Change the default of dnssec_request_domains to "*"
598 JH/42 Bug 2545: Fix CHUNKING for all RCPT commands rejected. Previously we
599 carried on and emitted a BDAT command, even when PIPELINING was not
602 JH/43 Bug 2465: Fix taint-handling in dsearch lookup. Previously a nontainted
603 buffer was used for the filename, resulting in a trap when tainted
604 arguments (eg. $domain) were used.
606 JH/44 With OpenSSL 1.1.1 (onwards) disable renegotiation for TLS1.2 and below;
607 recommended to avoid a possible server-load attack. The feature can be
608 re-enabled via the openssl_options main cofiguration option.
610 JH/45 local_scan API: documented the current smtp_printf() call. This changed
611 for version 4.90 - adding a "more data" boolean to the arguments.
612 Bumped the ABI version number also, this having been missed previously;
613 release versions 4.90 to 4.92.3 inclusive were effectively broken in
614 respect of usage of smtp_printf() by either local_scan code or libraries
615 accessed via the ${dlfunc } expansion item. Both will need coding
616 adjustment for any calls to smtp_printf() to match the new function
617 signature; a FALSE value for the new argument is always safe.
619 JH/46 FreeBSD: fix use of the sendfile() syscall. The shim was not updating
620 the file-offset (which the Linux syscall does, and exim expects); this
621 resulted in an indefinite loop.
623 JH/47 ARC: fix crash in signing, triggered when a configuration error failed
624 to do ARC verification. The Authentication-Results: header line added
625 by the configuration then had no ARC item.
631 JH/01 Remove code calling the customisable local_scan function, unless a new
632 definition "HAVE_LOCAL_SCAN=yes" is present in the Local/Makefile.
634 JH/02 Bug 1007: Avoid doing logging from signal-handlers, as that can result in
635 non-signal-safe functions being used.
637 JH/03 Bug 2269: When presented with a received message having a stupidly large
638 number of DKIM-Signature headers, disable DKIM verification to avoid
639 a resource-consumption attack. The limit is set at twenty.
641 JH/04 Add variables $arc_domains, $arc_oldest_pass for ARC verify. Fix the
642 report of oldest_pass in ${authres } in consequence, and separate out
643 some descriptions of reasons for verification fail.
645 JH/05 Bug 2273: Cutthrough delivery left a window where the received messsage
646 files in the spool were present and unlocked. A queue-runner could spot
647 them, resulting in a duplicate delivery. Fix that by doing the unlock
648 after the unlink. Investigation by Tim Stewart. Take the opportunity to
649 add more error-checking on spoolfile handling while that code is being
652 PP/01 Refuse to open a spool data file (*-D) if it's a symlink.
653 No known attacks, no CVE, this is defensive hardening.
655 JH/06 Bug 2275: The MIME ACL unlocked the received message files early, and
656 a queue-runner could start a delivery while other operations were ongoing.
657 Cutthrough delivery was a common victim, resulting in duplicate delivery.
658 Found and investigated by Tim Stewart. Fix by using the open message data
659 file handle rather than opening another, and not locally closing it (which
660 releases a lock) for that case, while creating the temporary .eml format
661 file for the MIME ACL. Also applies to "regex" and "spam" ACL conditions.
663 JH/07 Bug 177: Make a random-recipient callout success visible in ACL, by setting
664 $sender_verify_failure/$recipient_verify_failure to "random".
666 JH/08 When generating a selfsigned cert, use serial number 1 since zero is not
669 JH/09 Bug 2274: Fix logging of cmdline args when starting in an unlinked cwd.
670 Previously this would segfault.
672 JH/10 Fix ARC signing for case when DKIM signing failed. Previously this would
675 JH/11 Bug 2264: Exim now only follows CNAME chains one step by default. We'd
676 like zero, since the resolver should be doing this for us, But we need one
677 as a CNAME but no MX presence gets the CNAME returned; we need to check
678 that doesn't point to an MX to declare it "no MX returned" rather than
679 "error, loop". A new main option is added so the older capability of
680 following some limited number of chain links is maintained.
682 JH/12 Add client-ip info to non-pass iprev ${authres } lines.
684 JH/13 For receent Openssl versions (1.1 onward) use modern generic protocol
685 methods. These should support TLS 1.3; they arrived with TLS 1.3 and the
686 now-deprecated earlier definitions used only specified the range up to TLS
687 1.2 (in the older-version library docs).
689 JH/14 Bug 2284: Fix DKIM signing for body lines starting with a pair of dots.
691 JH/15 Rework TLS client-side context management. Stop using a global, and
692 explicitly pass a context around. This enables future use of TLS for
693 connections to service-daemons (eg. malware scanning) while a client smtp
694 connection is using TLS; with cutthrough connections this is quite likely.
696 JH/16 Fix ARC verification to do AS checks in reverse order.
698 JH/17 Support a "tls" option on the ${readsocket } expansion item.
700 JH/18 Bug 2287: Fix the protocol name (eg utf8esmtp) for multiple messages
701 using the SMTPUTF8 option on their MAIL FROM commands, in one connection.
702 Previously the "utf8" would be re-prepended for every additional message.
704 JH/19 Reject MAIL FROM commands with SMTPUTF8 when the facility was not advertised.
705 Previously thery were accepted, resulting in issues when attempting to
706 forward messages to a non-supporting MTA.
708 PP/02 Let -n work with printing macros too, not just options.
710 JH/20 Bug 2296: Fix cutthrough for >1 address redirection. Previously only
711 one parent address was copied, and bogus data was used at delivery-logging
712 time. Either a crash (after delivery) or bogus log data could result.
713 Discovery and analysis by Tim Stewart.
715 PP/03 Make ${utf8clean:} expansion operator detect incomplete final character.
716 Previously if the string ended mid-character, we did not insert the
717 promised '?' replacement.
719 PP/04 Documentation: current string operators work on bytes, not codepoints.
721 JH/21 Change as many as possible of the global flags into one-bit bitfields; these
722 should pack well giving a smaller memory footprint so better caching and
723 therefore performance. Group the declarations where this can't be done so
724 that the byte-sized flag variables are not interspersed among pointer
725 variables, giving a better chance of good packing by the compiler.
727 JH/22 Bug 1896: Fix the envelope from for DMARC forensic reports to be possibly
728 non-null, to avoid issues with sites running BATV. Previously reports were
729 sent with an empty envelope sender so looked like bounces.
731 JH/23 Bug 2318: Fix the noerror command within filters. It wasn't working.
732 The ignore_error flag wasn't being returned from the filter subprocess so
733 was not set for later routers. Investigation and fix by Matthias Kurz.
735 JH/24 Bug 2310: Raise a msg:fail:internal event for each undelivered recipient,
736 and a msg:complete for the whole, when a message is manually removed using
737 -Mrm. Developement by Matthias Kurz, hacked on by JH.
739 JH/25 Avoid fixed-size buffers for pathnames in DB access. This required using
740 a "Gnu special" function, asprintf() in the DB utility binary builds; I
741 hope that is portable enough.
743 JH/26 Bug 2311: Fix DANE-TA verification under GnuTLS. Previously it was also
744 requiring a known-CA anchor certificate; make it now rely entirely on the
745 TLSA as an anchor. Checking the name on the leaf cert against the name
746 on the A-record for the host is still done for TA (but not for EE mode).
748 JH/27 Fix logging of proxy address. Previously, a pointless "PRX=[]:0" would be
749 included in delivery lines for non-proxied connections, when compiled with
750 SUPPORT_SOCKS and running with proxy logging enabled.
752 JH/28 Bug 2314: Fire msg:fail:delivery event even when error is being ignored.
753 Developement by Matthias Kurz, tweaked by JH. While in that bit of code,
754 move the existing event to fire before the normal logging of message
755 failure so that custom logging is bracketed by normal logging.
757 JH/29 Bug 2322: A "fail" command in a non-system filter (file) now fires the
758 msg:fail:internal event. Developement by Matthias Kurz.
760 JH/30 Bug 2329: Increase buffer size used for dns lookup from 2k, which was
761 far too small for todays use of crypto signatures stored there. Go all
762 the way to the max DNS message size of 64kB, even though this might be
763 overmuch for IOT constrained device use.
765 JH/31 Fix a bad use of a copy function, which could be used to pointlessly
766 copy a string over itself. The library routine is documented as not
767 supporting overlapping copies, and on MacOS it actually raised a SIGABRT.
769 JH/32 For main options check_spool_space and check_inode_space, where the
770 platform supports 64b integers, support more than the previous 2^31 kB
771 (i.e. more than 2 TB). Accept E, P and T multipliers in addition to
772 the previous G, M, k.
774 JH/33 Bug 2338: Fix the cyrus-sasl authenticator to fill in the
775 $authenticated_fail_id variable on authentication failure. Previously
778 JH/34 Increase RSA keysize of autogen selfsign cert from 1024 to 2048. RHEL 8.0
779 OpenSSL didn't want to use such a weak key. Do for GnuTLS also, and for
780 more-modern GnuTLS move from GNUTLS_SEC_PARAM_LOW to
781 GNUTLS_SEC_PARAM_MEDIUM.
783 JH/35 OpenSSL: fail the handshake when SNI processing hits a problem, server
784 side. Previously we would continue as if no SNI had been received.
786 JH/36 Harden the handling of string-lists. When a list consisted of a sole
787 "<" character, which should be a list-separator specification, we walked
788 off past the nul-terimation.
790 JH/37 Bug 2341: Send "message delayed" warning MDNs (restricted to external
791 causes) even when the retry time is not yet met. Previously they were
792 not, meaning that when (say) an account was over-quota and temp-rejecting,
793 and multiple senders' messages were queued, only one sender would get
794 notified on each configured delay_warning cycle.
796 JH/38 Bug 2351: Log failures to extract envelope addresses from message headers.
798 JH/39 OpenSSL: clear the error stack after an SSL_accept(). With anon-auth
799 cipher-suites, an error can be left on the stack even for a succeeding
800 accept; this results in impossible error messages when a later operation
803 AM/01 Bug 2359: GnuTLS: repeat lowlevel read and write operations while they
804 return error codes indicating retry. Under TLS1.3 this becomes required.
806 JH/40 Fix the feature-cache refresh for EXPERIMENTAL_PIPE_CONNECT. Previously
807 it only wrote the new authenticators, resulting in a lack of tracking of
808 peer changes of ESMTP extensions until the next cache flush.
810 JH/41 Fix the loop reading a message header line to check for integer overflow,
811 and more-often against header_maxsize. Previously a crafted message could
812 induce a crash of the recive process; now the message is cleanly rejected.
814 JH/42 Bug 2366: Fix the behaviour of the dkim_verify_signers option. It had
815 been totally disabled for all of 4.91. Discovery and fix by "Mad Alex".
821 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
822 When redis_servers is set to a list of > 1 element, and the Redis servers
823 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
824 case of MOVED into a DEFER case instead, thus moving the query onto the
825 next server in the list. For a cluster of N elements, all N servers must
826 be defined in redis_servers.
828 GF/02 Catch and remove uninitialized value warning in exiqsumm
829 Check for existence of @ARGV before looking at $ARGV[0]
831 JH/01 Replace the store_release() internal interface with store_newblock(),
832 which internalises the check required to safely use the old one, plus
833 the allocate and data copy operations duplicated in both (!) of the
834 extant use locations.
836 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
837 modifier. This matches the restriction on the commandline.
839 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
840 Previously only the last row was returned.
842 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
843 we assumed that tags in the header were well-formed, and parsed the
844 element content after inspecting only the first char of the tag.
845 Assumptions at that stage could crash the receive process on malformed
848 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
849 While running the DKIM ACL we operate on the Permanent memory pool so that
850 variables created with "set" persist to the DATA ACL. Also (at any time)
851 DNS lookups that fail create cache records using the Permanent pool. But
852 expansions release any allocations made on the current pool - so a dnsdb
853 lookup expansion done in the DKIM ACL releases the memory used for the
854 DNS negative-cache, and bad things result. Solution is to switch to the
855 Main pool for expansions.
856 While we're in that code, add checks on the DNS cache during store_reset,
857 active in the testsuite.
858 Problem spotted, and debugging aided, by Wolfgang Breyha.
860 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
861 When none of the hosts presented to a transport match an already-open
862 connection, close it and proceed with the list. Previously we would
863 queue the message. Spotted by Lena with Yahoo, probably involving
866 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
867 Previously a spurious "250 OK id=" response was appended to the proper
870 JH/08 The "support for" informational output now, which built with Content
871 Scanning support, has a line for the malware scanner interfaces compiled
872 in. Interface can be individually included or not at build time.
874 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
875 by the template makefile "src/EDITME". The "STREAM" support for an older
876 ClamAV interface method is removed.
878 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
879 rows affected is given instead).
881 JH/11 The runtime Berkeley DB library version is now additionally output by
882 "exim -d -bV". Previously only the compile-time version was shown.
884 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
885 SMTP connection. Previously, when one had more recipients than the
886 first, an abortive onward connection was made. Move to full support for
887 multiple onward connections in sequence, handling cutthrough connection
888 for all multi-message initiating connections.
890 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
891 routers. Previously, a multi-recipient message would fail to match the
892 onward-connection opened for the first recipient, and cause its closure.
894 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
895 a timeout on read on a GnuTLS initiating connection, resulting in the
896 initiating connection being dropped. This mattered most when the callout
897 was marked defer_ok. Fix to keep the two timeout-detection methods
900 JH/15 Relax results from ACL control request to enable cutthrough, in
901 unsupported situations, from error to silently (except under debug)
902 ignoring. This covers use with PRDR, frozen messages, queue-only and
905 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
907 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
908 metadata, resulting in a crash in free().
910 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
911 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
912 Broken also in d185889f4, with init system revamp.
914 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
915 Previously we abruptly closed the connection after reading a malware-
916 found indication; now we go on to read the "scan ok" response line,
919 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
920 ACL. Previously, a crash would result.
922 JH/19 Speed up macro lookups during configuration file read, by skipping non-
923 macro text after a replacement (previously it was only once per line) and
924 by skipping builtin macros when searching for an uppercase lead character.
926 JH/20 DANE support moved from Experimental to mainline. The Makefile control
927 for the build is renamed.
929 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
930 was allocated for every new TLS startup, meaning one per message. Fix
931 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
933 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
934 reported the original. Fix to report (as far as possible) the ACL
935 result replacing the original.
937 JH/23 Fix memory leak during multi-message connections using STARTTLS under
938 OpenSSL. Certificate information is loaded for every new TLS startup,
939 and the resources needed to be freed.
941 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
943 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
946 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
947 DATA response info to the (existing) per-recipient response info for
948 the "C=" log element. It can have useful tracking info from the
949 destination system. Patch from Simon Arlott.
951 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
952 length value. Previously this would segfault.
954 HS/02 Support Avast multiline protoocol, this allows passing flags to
955 newer versions of the scanner.
957 JH/28 Ensure that variables possibly set during message acceptance are marked
958 dead before release of memory in the daemon loop. This stops complaints
959 about them when the debug_store option is enabled. Discovered specifically
960 for sender_rate_period, but applies to a whole set of variables.
961 Do the same for the queue-runner and queue-list loops, for variables set
962 from spool message files. Do the same for the SMTP per-message loop, for
963 certain variables indirectly set in ACL operations.
965 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
966 as a multi-recipient message from a mailinglist manager). The coding had
967 an arbitrary cutoff number of characters while checking for more input;
968 enforced by writing a NUL into the buffer. This corrupted long / fast
969 input. The problem was exposed more widely when more pipelineing of SMTP
970 responses was introduced, and one Exim system was feeding another.
971 The symptom is log complaints of SMTP syntax error (NUL chars) on the
972 receiving system, and refused recipients seen by the sending system
973 (propating to people being dropped from mailing lists).
974 Discovered and pinpointed by David Carter.
976 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
977 replaced by the ${authresults } expansion.
979 JH/31 Bug 2257: Fix pipe transport to not use a socket-only syscall.
981 HS/03 Set a handler for SIGTERM and call exit(3) if running as PID 1. This
982 allows proper process termination in container environments.
984 JH/32 Bug 2258: Fix spool_wireformat in combination with LMTP transport.
985 Previously the "final dot" had a newline after it; ensure it is CR,LF.
987 JH/33 SPF: remove support for the "spf" ACL condition outcome values "err_temp"
988 and "err_perm", deprecated since 4.83 when the RFC-defined words
989 "temperror" and "permerror" were introduced.
991 JH/34 Re-introduce enforcement of no cutthrough delivery on transports having
992 transport-filters or DKIM-signing. The restriction was lost in the
993 consolidation of verify-callout and delivery SMTP handling.
994 Extend the restriction to also cover ARC-signing.
996 JH/35 Cutthrough: for a final-dot response timeout (and nonunderstood responses)
997 in defer=pass mode supply a 450 to the initiator. Previously the message
1000 PP/02 DANE: add dane_require_tls_ciphers SMTP Transport option; if unset,
1001 tls_require_ciphers is used as before.
1003 HS/03 Malware Avast: Better match the Avast multiline protocol. Add
1004 "pass_unscanned". Only tmpfails from the scanner are written to
1005 the paniclog, as they may require admin intervention (permission
1006 denied, license issues). Other scanner errors (like decompression
1007 bombs) do not cause a paniclog entry.
1009 JH/36 Fix reinitialisation of DKIM logging variable between messages.
1010 Previously it was possible to log spurious information in receive log
1013 JH/37 Bug 2255: Revert the disable of the OpenSSL session caching. This
1014 triggered odd behaviour from Outlook Express clients.
1016 PP/03 Add util/renew-opendmarc-tlds.sh script for safe renewal of public
1019 JH/38 DKIM: accept Ed25519 pubkeys in SubjectPublicKeyInfo-wrapped form,
1020 since the IETF WG has not yet settled on that versus the original
1021 "bare" representation.
1023 JH/39 Fix syslog logging for syslog_timestamp=no and log_selector +millisec.
1024 Previously the millisecond value corrupted the output.
1025 Fix also for syslog_pid=no and log_selector +pid, for which the pid
1026 corrupted the output.
1032 JH/01 Rework error string handling in TLS interface so that the caller in
1033 more cases is responsible for logging. This permits library-sourced
1034 string to be attached to addresses during delivery, and collapses
1035 pairs of long lines into single ones.
1037 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
1038 during configuration. Wildcards are allowed and expanded.
1040 JH/02 Rework error string handling in DKIM to pass more info back to callers.
1041 This permits better logging.
1043 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
1044 do not close it down and have the child transport start it up again on
1045 the passed-on TCP connection. Instead, proxy the child (and any
1046 subsequent ones) for TLS via a unix-domain socket channel. Logging is
1047 affected: the continued delivery log lines do not have any DNSSEC, TLS
1048 Certificate or OCSP information. TLS cipher information is still logged.
1050 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
1051 identical IP addresses on different listening ports. Will also affect
1054 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
1055 add noisy ifdef guards to special-case this sillyness.
1056 Patch from Bernd Kuhls.
1058 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
1059 than 255 are no longer allowed.
1061 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
1062 Disable the session-cache too, which might reduce our load. Since we
1063 currrectly use a new context for every connection, both as server and
1064 client, there is no benefit for these.
1065 GnuTLS appears to not support tickets server-side by default (we don't
1066 call gnutls_session_ticket_enable_server()) but client side is enabled
1067 by default on recent versions (3.1.3 +) unless the PFS priority string
1070 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
1071 <https://reproducible-builds.org/specs/source-date-epoch/>.
1073 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
1074 the check for any unsuccessful recipients did not notice the limit, and
1075 erroneously found still-pending ones.
1077 JH/08 Pipeline CHUNKING command and data together, on kernels that support
1078 MSG_MORE. Only in-clear (not on TLS connections).
1080 JH/09 Avoid using a temporary file during transport using dkim. Unless a
1081 transport-filter is involved we can buffer the headers in memory for
1082 creating the signature, and read the spool data file once for the
1083 signature and again for transmission.
1085 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
1086 4.77 as the kernel support then wasn't solid, having issues in 64bit
1087 mode. Now, it's been long enough. Add support for FreeBSD also.
1089 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
1090 case where the routing stage had gathered several addresses to send to
1091 a host before calling the transport for the first, we previously failed
1092 to close down TLS in the old transport process before passing the TCP
1093 connection to the new process. The new one sent a STARTTLS command
1094 which naturally failed, giving a failed delivery and bloating the retry
1095 database. Investigation and fix prototype from Wolfgang Breyha.
1097 JH/12 Fix check on SMTP command input synchronisation. Previously there were
1098 false-negatives in the check that the sender had not preempted a response
1099 or prompt from Exim (running as a server), due to that code's lack of
1100 awareness of the SMTP input buffering.
1102 PP/04 Add commandline_checks_require_admin option.
1103 Exim drops privileges sanely, various checks such as -be aren't a
1104 security problem, as long as you trust local users with access to their
1105 own account. When invoked by services which pass untrusted data to
1106 Exim, this might be an issue. Set this option in main configuration
1107 AND make fixes to the calling application, such as using `--` to stop
1110 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
1111 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
1112 responses to those, into a single TLS record each way (this usually means
1113 a single packet). As a side issue, smtp_enforce_sync now works on TLS
1116 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
1117 affects you only if you're dancing at the edge of the param size limits.
1118 If you are, and this message makes sense to you, then: raise the
1119 configured limit or use OpenSSL 1.1. Nothing we can do for older
1122 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
1123 cmdline element to get the documented default one. Previously it was
1126 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
1127 are unsuable for non-message-specific reasons - eg. connection timeout,
1128 banner-time rejection.
1130 JH/16 Fix logging of delivery remote port, when specified by router, under
1133 PP/06 Repair manualroute's ability to take options in any order, even if one
1134 is the name of a transport.
1137 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
1139 JH/17 Change the list-building routines interface to use the expanding-string
1140 triplet model, for better allocation and copying behaviour.
1142 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
1143 Previously it was constructed the first time a possibly-matching string
1144 was met in the configuration file input during startup; now it is done
1147 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
1148 compatible one, to avoid the (poorly documented) possibility of a config
1149 file in the working directory redirecting the DB files, possibly correpting
1150 some existing file. CVE-2017-10140 assigned for BDB.
1152 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
1153 cache-hot. Previously, although the result was properly cached, the
1154 initial verify call returned a defer.
1156 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
1157 the main verify for receipient in uncached-mode.
1159 JH/22 Retire historical build files to an "unsupported" subdir. These are
1160 defined as "ones for which we have no current evidence of testing".
1162 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
1163 if present. Previously it was ignored.
1165 JH/24 Start using specified-initialisers in C structure init coding. This is
1166 a C99 feature (it's 2017, so now considered safe).
1168 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
1169 if was a fixed-sized field and bitmask ops via macros; it is now more
1172 PP/07 GitHub PR 56: Apply MariaDB build fix.
1173 Patch provided by Jaroslav Škarvada.
1175 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
1176 during Coverity cleanups [4.87 JH/47]
1177 Diagnosis and fix provided by Michael Fischer v. Mollard.
1179 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
1180 the right size to place the terminating semicolon on its own folded
1181 line, the header hash was calculated to an incorrect value thanks to
1182 the (relaxed) space the fold became.
1184 HS/02 Fix Bug 2130: large writes from the transport subprocess were chunked
1185 and confused the parent.
1187 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
1188 which could crash as a result. This could lead to undeliverable messages.
1190 JH/28 Logging: "next input sent too soon" now shows where input was truncated
1193 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
1194 matters on fast-turnover and PID-randomising systems, which were getting
1195 out-of-order delivery.
1197 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
1198 a possibly-overlapping copy. The symptom was that "Remote host closed
1199 connection in response to HELO" was logged instead of the actual 4xx
1202 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
1203 Previously only that bufferd was discarded, resulting in SYMTP command
1206 JH/32 DKIM: when a message has multiple signatures matching an identity given
1207 in dkim_verify_signers, run the dkim acl once for each. Previously only
1208 one run was done. Bug 2189.
1210 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
1211 "panic the current process" to "deliberately defer". The panic log is
1212 still written with the problem list name; the mail and reject logs now
1213 get a temp-reject line for the message that was being handled, saying
1214 something like "domains check lookup or other defer". The SMTP 451
1215 message is still "Temporary local problem".
1217 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
1218 A crafted sequence of BDAT commands could result in in-use memory beeing
1219 freed. CVE-2017-16943.
1221 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
1222 from SMTP input. Previously it was always done; now only done for DATA
1223 and not BDAT commands. CVE-2017-16944.
1225 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
1226 to the message (such as an overlong header line). Previously this was
1227 not done and we did not exit BDAT mode. Followon from the previous item
1228 though a different problem.
1234 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
1235 than -2003 did; needs libidn2 in addition to libidn.
1237 JH/02 The path option on a pipe transport is now expanded before use.
1239 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
1240 Patch provided by "Björn", documentation fix added too.
1242 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
1243 missing a wire-to-host endian conversion.
1245 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
1246 close after a BDAT command line could be taken as a following command,
1247 giving a synch failure. Fix by only checking for synch immediately
1248 before acknowledging the chunk.
1250 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
1251 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
1252 macro. Patches provided by Josh Soref.
1254 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
1255 Previously we did not; the RFC seems ambiguous and VRFY is not listed
1256 by IANA as a service extension. However, John Klensin suggests that we
1259 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
1260 the dkim code may be unix-mode line endings rather than smtp wire-format
1261 CRLF, so prepend a CR to any bare LF.
1263 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
1264 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
1266 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
1267 the first were themselves being wrongly included in the feed into dkim
1268 processing; with most chunk sizes in use this resulted in an incorrect
1269 body hash calculated value.
1271 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
1272 DKIM signature block, for verification. Although advised against by
1273 standards it is specifically not ruled illegal.
1275 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
1277 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
1278 missing a body hash (the bh= tag).
1280 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
1281 It seems that HAProxy sends the Proxy Protocol information in clear and
1282 only then does a TLS startup, so do the same.
1284 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
1285 TCP connections (such as for Spamd) unless the daemon successfully set
1286 Fast Open mode on its listening sockets. This fixes breakage seen on
1287 too-old kernels or those not configured for Fast Open, at the cost of
1288 requiring both directions being enabled for TFO, and TFO never being used
1289 by non-daemon-related Exim processes.
1291 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
1292 endings, at least on the first header line. Try to canonify any that get
1293 past that check, despite the cost.
1295 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
1296 now limited to an arbitrary five deep, while parsing addresses with the
1297 strip_excess_angle_brackets option enabled.
1299 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
1300 instead leave the unprompted TLS handshake in socket buffer for the
1301 TLS library to consume.
1303 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
1305 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
1307 JH/16 Drop variables when they go out of scope. Memory management drops a whole
1308 region in one operation, for speed, and this leaves assigned pointers
1309 dangling. Add checks run only under the testsuite which checks all
1310 variables at a store-reset and panics on a dangling pointer; add code
1311 explicitly nulling out all the variables discovered. Fixes one known
1312 bug: a transport crash, where a dangling pointer for $sending_ip_address
1313 originally assigned in a verify callout, is re-used.
1315 PP/06 Drop '.' from @INC in various Perl scripts.
1317 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
1319 PP/08 Reduce a number of compilation warnings under clang; building with
1320 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
1321 should be warning-free.
1323 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
1325 HS/01 Fix portability problems introduced by PP/08 for platforms where
1326 realloc(NULL) is not equivalent to malloc() [SunOS et al].
1328 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
1329 chunk. This allows us to accept broken chunked messages. We need a more
1330 general solution here.
1332 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
1333 already-broken messages in the queue.
1335 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
1337 JH/19 Fix reference counting bug in routing-generated-address tracking.
1343 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
1344 supports it and a size is available (ie. the sending peer gave us one).
1346 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
1347 years of being deprecated). The replacements are the ACLs
1348 acl_smtp_mime and acl_not_smtp_mime.
1350 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
1351 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
1352 or even an in-clear connection were permitted. Now, if the host lookup
1353 was dnssec and dane was requested then the host is only used if the
1354 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
1355 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
1356 if one fails this test.
1357 This means that a poorly-configured remote DNS will make it incommunicado;
1358 but it protects against a DNS-interception attack on it.
1360 JH/04 Bug 1810: make continued-use of an open smtp transport connection
1361 non-noisy when a race steals the message being considered.
1363 JH/05 If main configuration option tls_certificate is unset, generate a
1364 self-signed certificate for inbound TLS connections.
1366 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
1367 in rewrites and routers.
1369 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
1370 and logged a warning sing 4.83; now they are a configuration file error.
1372 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
1373 (lacking @domain). Apply the same qualification processing as RCPT.
1375 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
1377 JH/10 Support ${sha256:} applied to a string (as well as the previous
1380 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
1381 a cutthrough deliver is pending, as we always want to make a connection.
1382 This also avoids re-routing the message when later placing the cutthrough
1383 connection after a verify cache hit.
1384 Do not update it with the verify result either.
1386 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
1387 when routing results in more than one destination address.
1389 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
1390 signing (which inhibits the cutthrough capability). Previously only
1391 the presence of an option was tested; now an expansion evaluating as
1392 empty is permissible (obviously it should depend only on data available
1393 when the cutthrough connection is made).
1395 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
1396 the relevant preceding SMTP command did not note the pipelining mode.
1398 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
1399 Previously they were not counted.
1401 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
1402 as one having no matching records. Previously we deferred the message
1403 that needed the lookup.
1405 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
1406 distinguished as "(=".
1408 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
1409 for missing MX records. Previously it only worked for missing A records.
1411 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
1413 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
1414 after the data-go-ahead and data-ack. Patch from Jason Betts.
1416 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
1417 even for a "none" policy. Patch from Tony Meyer.
1419 JH/22 Fix continued use of a connection for further deliveries. If a port was
1420 specified by a router, it must also match for the delivery to be
1423 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
1424 When one of the recipients of a message was unsuitable for the connection
1425 (has no matching addresses), we lost track of needing to mark it
1426 deferred. As a result mail would be lost.
1428 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
1430 JH/25 Decoding ACL controls is now done using a binary search; the source code
1431 takes up less space and should be simpler to maintain. Merge the ACL
1432 condition decode tables also, with similar effect.
1434 JH/26 Fix problem with one_time used on a redirect router which returned the
1435 parent address unchanged. A retry would see the parent address marked as
1436 delivered, so not attempt the (identical) child. As a result mail would
1439 JH/27 Fix a possible security hole, wherein a process operating with the Exim
1440 UID can gain a root shell. Credit to http://www.halfdog.net/ for
1441 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
1444 JH/28 Enable {spool,log} filesystem space and inode checks as default.
1445 Main config options check_{log,spool}_{inodes,space} are now
1446 100 inodes, 10MB unless set otherwise in the configuration.
1448 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
1449 Previously it only applied to the main-section connection policy
1452 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
1454 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
1455 by me. Added RFC7919 DH primes as an alternative.
1457 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
1458 are not in the system include path.
1460 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
1461 GnuTLS, when a session startup failed (eg because the client disconnected)
1462 Exim did stdio operations after fclose. This was exposed by a recent
1463 change which nulled out the file handle after the fclose.
1465 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
1466 signed directly by the cert-signing cert, rather than an intermediate
1467 OCSP-signing cert. This is the model used by LetsEncrypt.
1469 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
1471 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
1472 an incoming connection.
1474 HS/02 Bug 1802: Do not half-close the connection after sending a request
1477 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
1478 fallback to "prime256v1".
1480 JH/34 SECURITY: Use proper copy of DATA command in error message.
1481 Could leak key material. Remotely exploitable. CVE-2016-9963.
1487 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
1488 and 3.4.4 - once the server is enabled to respond to an OCSP request
1489 it does even when not requested, resulting in a stapling non-aware
1490 client dropping the TLS connection.
1492 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
1493 support variable-length bit vectors. No functional change.
1495 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
1496 The I= interface field on outgoing lines is now after the H= remote
1497 host field, same as incoming lines. There is a separate
1498 outgoing_interface log selector which allows you to disable the
1501 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
1502 If not running log_selector +smtp_connection the mainlog would be held
1503 open indefinitely after a "too many connections" event, including to a
1504 deleted file after a log rotate. Leave the per net connection logging
1505 leaving it open for efficiency as that will be quickly detected by the
1506 check on the next write.
1508 HS/01 Bug 1671: Fix post transport crash.
1509 Processing the wait-<transport> messages could crash the delivery
1510 process if the message IDs didn't exist for some reason. When
1511 using 'split_spool_directory=yes' the construction of the spool
1512 file name failed already, exposing the same netto behaviour.
1514 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
1515 mime_regex ACL conditions.
1517 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
1518 to DSN fail messages (bounces): remote IP, remote greeting, remote response
1519 to HELO, local diagnostic string.
1521 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
1522 log line to debug. Even when configured with a tls authenticator many
1523 client connections are expected to not authenticate in this way, so
1524 an authenticate fail is not an error.
1526 HS/02 Add the Exim version string to the process info. This way exiwhat
1527 gives some more detail about the running daemon.
1529 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
1530 matter for fast-change records such as DNSBLs.
1532 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
1533 retry record, even if constant. There may be multiple transports with
1534 different interface settings and the retry behaviour needs to be kept
1537 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
1539 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
1541 JH/10 Bug 840: fix log_defer_output option of pipe transport
1543 JH/11 Bug 830: use same host for all RCPTS of a message, even under
1544 hosts_randomize. This matters a lot when combined with mua_wrapper.
1546 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
1547 ${quote_pgsql:<string>} operator.
1549 JH/13 Bug 1708: avoid misaligned access in cached lookup.
1551 JH/14 Change header file name for freeradius-client. Relevant if compiling
1552 with Radius support; from the Gentoo tree and checked under Fedora.
1554 JH/15 Bug 1712: Introduce $prdr_requested flag variable
1556 JH/16 Bug 1714: Permit an empty string as expansion result for transport
1557 option transport_filter, meaning no filtering.
1559 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
1561 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
1562 defaults to "*" (all hosts). The variable is now available when not built
1563 with TLS, default unset, mainly to enable keeping the testsuite sane.
1564 If a server certificate is not supplied (via tls_certificate) an error is
1565 logged, and clients will find TLS connections fail on startup. Presumably
1566 they will retry in-clear.
1567 Packagers of Exim are strongly encouraged to create a server certificate
1568 at installation time.
1570 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
1571 with the $config_file variable.
1573 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
1574 in transport context, after the attempt, and per-recipient. The latter type
1575 is per host attempted. The event data is the error message, and the errno
1576 information encodes the lookup type (A vs. MX) used for the (first) host,
1577 and the trailing two digits of the smtp 4xx response.
1579 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
1580 to write to mainlog (or rejectlog, paniclog) in the window between file
1581 creation and permissions/ownership being changed. Particularly affects
1582 installations where exicyclog is run as root, rather than exim user;
1583 result is that the running daemon panics and dies.
1585 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
1587 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
1588 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
1589 "pri" and "weight". Note that the previous implicit priority given by the
1590 list order is no longer honoured.
1592 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
1593 for DKIM processing.
1595 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
1596 by defining SUPPORT_SOCKS.
1598 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
1599 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
1600 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
1601 variables are renamed to proxy_{local,external}_{address,port}.
1603 JH/27 Move Internationalisation support from Experimental to mainline, enabled
1604 for a build by defining SUPPORT_I18N
1606 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
1607 of the query string, and make ${quote_redis:} do that quoting.
1609 JH/29 Move Events support from Experimental to mainline, enabled by default
1610 and removable for a build by defining DISABLE_EVENT.
1612 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
1614 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
1615 cached by the daemon.
1617 JH/32 Move Redis support from Experimental to mainline, enabled for a build
1618 by defining LOOKUP_REDIS. The libhiredis library is required.
1620 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
1621 keys are given for lookup.
1623 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
1624 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
1625 only supported when built with TLS support. The PolarSSL SHA routines
1626 are still used when the TLS library is too old for convenient support.
1628 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
1629 openssl_options), for security. OpenSSL forces this from version 1.1.0
1630 server-side so match that on older versions.
1632 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
1633 allocation for $value could be released as the expansion processing
1634 concluded, but leaving the global pointer active for it.
1636 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
1637 and to use the domains and local_parts ACL conditions.
1639 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
1640 incorrectly not doubled on cutthrough transmission, hence seen as a
1641 body-termination at the receiving system - resulting in truncated mails.
1642 Commonly the sender saw a TCP-level error, and retransmitted the message
1643 via the normal store-and-forward channel. This could result in duplicates
1644 received - but deduplicating mailstores were liable to retain only the
1645 initial truncated version.
1647 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
1649 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
1651 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
1652 we're in there, support oversigning also; bug 1309.
1654 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
1656 HS/04 Add support for keep_environment and add_environment options.
1658 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
1659 either intentional arithmetic overflow during PRNG, or testing config-
1662 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
1663 delivery resulted in actual delivery. Cancel cutthrough before DATA
1666 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
1667 rejects a recipient: pass the reject to the originator.
1669 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
1670 Many were false-positives and ignorable, but it's worth fixing the
1673 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
1674 for the new environment-manipulation done at startup. Move the routines
1675 from being local to tls.c to being global via the os.c file.
1677 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
1678 an extract embedded as result-arg for a map, the first arg for extract
1679 is unavailable so we cannot tell if this is a numbered or keyed
1680 extraction. Accept either.
1686 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
1689 JH/02 The smtp transport option "multi_domain" is now expanded.
1691 JH/03 The smtp transport now requests PRDR by default, if the server offers
1694 JH/04 Certificate name checking on server certificates, when exim is a client,
1695 is now done by default. The transport option tls_verify_cert_hostnames
1696 can be used to disable this per-host. The build option
1697 EXPERIMENTAL_CERTNAMES is withdrawn.
1699 JH/05 The value of the tls_verify_certificates smtp transport and main options
1700 default to the word "system" to access the system default CA bundle.
1701 For GnuTLS, only version 3.0.20 or later.
1703 JH/06 Verification of the server certificate for a TLS connection is now tried
1704 (but not required) by default. The verification status is now logged by
1705 default, for both outbound TLS and client-certificate supplying inbound
1708 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
1711 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
1712 Status Notification (bounce) messages are now MIME format per RFC 3464.
1713 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
1714 under the control of the dsn_advertise_hosts option, and routers may
1715 have a dsn_lasthop option.
1717 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
1718 default, modifiable by a malware= option. The list separator for
1719 the options can now be changed in the usual way. Bug 68.
1721 JH/10 The smtp_receive_timeout main option is now expanded before use.
1723 JH/11 The incoming_interface log option now also enables logging of the
1724 local interface on delivery outgoing connections.
1726 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
1727 if the interface and destination host and port all match.
1729 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
1732 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
1733 Patch from Andrew Lewis.
1735 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
1736 now supports optional time-restrictions, weighting, and priority
1737 modifiers per server. Patch originally by <rommer@active.by>.
1739 JH/16 The spamd_address main option now supports a mixed list of local
1740 and remote servers. Remote servers can be IPv6 addresses, and
1741 specify a port-range.
1743 JH/17 Bug 68: The spamd_address main option now supports an optional
1744 timeout value per server.
1746 JH/18 Bug 1581: Router and transport options headers_add/remove can
1747 now have the list separator specified.
1749 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
1752 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
1755 JH/21 Support for the A6 type of dns record is withdrawn.
1757 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
1758 rather than the verbs used.
1760 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
1761 from 255 to 1024 chars.
1763 JH/24 Verification callouts now attempt to use TLS by default.
1765 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
1766 are generic router options now. The defaults didn't change.
1768 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
1769 Original patch from Alexander Shikoff, worked over by JH.
1771 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
1772 files if ZCAT_COMMAND is not executable.
1774 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
1776 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
1778 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
1779 Normally benign, it bites when the pair was led to by a CNAME;
1780 modern usage is to not canonicalize the domain to a CNAME target
1781 (and we were inconsistent anyway for A-only vs AAAA+A).
1783 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
1785 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
1786 when evaluating $sender_host_dnssec.
1788 JH/31 Check the HELO verification lookup for DNSSEC, adding new
1789 $sender_helo_dnssec variable.
1791 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
1793 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
1795 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
1797 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
1798 documented as working, but never had. Support all but $spam_report.
1800 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
1801 added for tls authenticator.
1803 HS/03 Add perl_taintmode main config option
1809 TL/01 When running the test suite, the README says that variables such as
1810 no_msglog_check are global and can be placed anywhere in a specific
1811 test's script, however it was observed that placement needed to be near
1812 the beginning for it to behave that way. Changed the runtest perl
1813 script to read through the entire script once to detect and set these
1814 variables, reset to the beginning of the script, and then run through
1815 the script parsing/test process like normal.
1817 TL/02 The BSD's have an arc4random API. One of the functions to induce
1818 adding randomness was arc4random_stir(), but it has been removed in
1819 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
1820 function when detected.
1822 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
1823 cause callback expansion.
1825 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
1826 syntax errors in an expansion can be treated as a string instead of
1827 logging or causing an error, due to the internal use of bool_lax
1828 instead of bool when processing it.
1830 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
1831 server certificates when making smtp deliveries.
1833 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
1835 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
1837 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
1839 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
1840 Merged patch from Sebastian Wiedenroth.
1842 JH/05 Fix results-pipe from transport process. Several recipients, combined
1843 with certificate use, exposed issues where response data items split
1844 over buffer boundaries were not parsed properly. This eventually
1845 resulted in duplicates being sent. This issue only became common enough
1846 to notice due to the introduction of connection certificate information,
1847 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1849 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1850 size buffer was used, resulting in syntax errors when an expansion
1853 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1854 version 3.3.6 or later.
1856 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1857 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1858 both become "event_action", the variables become $event_name, $event_data
1859 and $event_defer_errno. There is a new variable $verify_mode, usable in
1860 routers, transports and related events. The tls:cert event is now also
1861 raised for inbound connections, if the main configuration event_action
1864 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1865 early OCSP support, but no stapling (appears to be less than 1.0.0).
1867 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1868 server certificate names available under the smtp transport option
1869 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1872 JH/10 Time-related extraction expansions from certificates now use the main
1873 option "timezone" setting for output formatting, and are consistent
1874 between OpenSSL and GnuTLS compilations. Bug 1541.
1876 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1877 encoded parameter in the incoming message. Bug 1558.
1879 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1880 include certificate info, eximon was claiming there were spoolfile
1883 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1885 JH/14 Log delivery-related information more consistently, using the sequence
1886 "H=<name> [<ip>]" wherever possible.
1888 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1889 are problematic for Debian distribution, omit them from the release
1892 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1894 JH/16 Fix string representation of time values on 64bit time_t architectures.
1897 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1898 output list separator was used.
1903 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1904 checkers that were complaining about end of non-void function with no
1907 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1908 This was a regression introduced in 4.83 by another bugfix.
1910 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1912 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1913 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1919 TF/01 Correctly close the server side of TLS when forking for delivery.
1921 When a message was received over SMTP with TLS, Exim failed to clear up
1922 the incoming connection properly after forking off the child process to
1923 deliver the message. In some situations the subsequent outgoing
1924 delivery connection happened to have the same fd number as the incoming
1925 connection previously had. Exim would try to use TLS and fail, logging
1926 a "Bad file descriptor" error.
1928 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1929 utilities have not been installed.
1931 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1932 temporary space as the ACL may create new global variables.
1934 TL/01 LDAP support uses per connection or global context settings, depending
1935 upon the detected version of the libraries at build time.
1937 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1938 to extract and use the src ip:port in logging and expansions as if it
1939 were a direct connection from the outside internet. PPv2 support was
1940 updated based on HAProxy spec change in May 2014.
1942 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1944 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1945 Properly escape header and check for NULL return.
1947 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1950 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1952 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1953 characters in header names, implemented as a verify condition.
1954 Contributed by Michael Fischer v. Mollard.
1956 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1957 results permerror and temperror. Previous values are deprecated but
1958 still accepted. In a future release, err_perm and err_temp will be
1959 completely removed, which will be a backward incompatibility if the
1960 ACL tests for either of these two old results. Patch contributed by
1961 user bes-internal on the mailing list.
1963 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1965 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1966 selectors, in both main and reject logs.
1968 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1971 JH/07 Add malware type "sock" for talking to simple daemon.
1973 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1975 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1976 routers/transports under cutthrough routing.
1978 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1979 numbers. Touch up "bool" conditional to keep the same definition.
1981 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1983 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1985 JH/12 Expand items in router/transport headers_add or headers_remove lists
1986 individually rather than the list as a whole. Bug 1452.
1988 Required for reasonable handling of multiple headers_ options when
1989 they may be empty; requires that headers_remove items with embedded
1990 colons must have them doubled (or the list-separator changed).
1992 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1993 view the policy declared in the DMARC record. Currently, $dmarc_status
1994 is a combined value of both the record presence and the result of the
1997 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1999 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
2000 dnslookup router and the smtp transport (applying to the forward
2003 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
2004 of ldap servers used for a specific lookup. Patch provided by Heiko
2007 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
2008 New variable $lookup_dnssec_authenticated for observability.
2010 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
2011 Patch submitted by Lars Timman.
2013 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
2015 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
2016 Requires trusted mode and valid format message id, aborts otherwise.
2017 Patch contributed by Heiko Schlichting.
2019 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
2020 certextract with support for various fields. Bug 1358.
2022 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
2023 is requested by default, modifiable by smtp transport option
2026 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
2027 operate on certificate variables to give certificate fingerprints
2028 Also new ${sha256:cert_variable}.
2030 JH/23 The PRDR feature is moved from being Experimental into the mainline.
2032 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
2033 Christian Aistleitner.
2035 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
2037 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
2038 file. Patch from Wolfgang Breyha.
2040 JH/25 Expand the coverage of the delivery $host and $host_address to
2041 client authenticators run in verify callout. Bug 1476.
2043 JH/26 Port service names are now accepted for tls_on_connect_ports, to
2044 align with daemon_smtp_ports. Bug 72.
2046 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
2047 support and error reporting did not work properly.
2049 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
2050 and is readable. Patch from Andrew Colin Kissa.
2052 TL/14 Enhance documentation of ${run expansion and how it parses the
2053 commandline after expansion, particularly in the case when an
2054 unquoted variable expansion results in an empty value.
2056 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
2058 PP/02 Fix internal collision of T_APL on systems which support RFC3123
2059 by renaming away from it. Addresses GH issue 15, reported by
2062 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
2064 TL/15 SECURITY: prevent double expansion in math comparison functions
2065 (can expand unsanitized data). Not remotely exploitable.
2072 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
2074 PP/02 Make -n do something, by making it not do something.
2075 When combined with -bP, the name of an option is not output.
2077 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
2080 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
2081 $sender_host_name and config options to manage this, and basic check
2084 PP/05 DSCP support for outbound connections and control modifier for inbound.
2086 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
2087 (Only plugin which currently uses this is kerberos4, which nobody should
2088 be using, but we should make it available and other future plugins might
2089 conceivably use it, even though it would break NAT; stuff *should* be
2090 using channel bindings instead).
2092 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
2093 name; added for Sendmail compatibility; requires admin caller.
2094 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
2095 just ignore it); requires trusted caller.
2096 Also parse but ignore: -Ac -Am -X<logfile>
2099 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
2101 TL/02 Add +smtp_confirmation as a default logging option.
2103 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
2104 Patch by Magnus Holmgren from 2007-02-20.
2106 TL/04 Bugzilla 1281 - Spec typo.
2107 Bugzilla 1283 - Spec typo.
2108 Bugzilla 1290 - Spec grammar fixes.
2110 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
2112 TL/06 Add Experimental DMARC support using libopendmarc libraries.
2114 TL/07 Fix an out of order global option causing a segfault. Reported to dev
2115 mailing list by by Dmitry Isaikin.
2117 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
2119 JH/02 Support "G" suffix to numbers in ${if comparisons.
2121 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
2123 NM/01 Bugzilla 1197 - Spec typo
2124 Bugzilla 1196 - Spec examples corrections
2126 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
2128 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
2129 gnutls_enable_pkcs11, but renamed to more accurately indicate its
2132 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
2133 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
2135 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
2136 "acl {{name}{arg}...}", and optional args on acl condition
2139 JH/05 Permit multiple router/transport headers_add/remove lines.
2141 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
2143 JH/07 Avoid using a waiting database for a single-message-only transport.
2144 Performance patch from Paul Fisher. Bugzilla 1262.
2146 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
2149 JH/09 Add $headers_added variable, with content from use of ACL modifier
2150 add_header (but not yet added to the message). Bugzilla 199.
2152 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
2153 Pulled from Bugzilla 817 by Wolfgang Breyha.
2155 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
2157 (nb: this is the same fix as in Exim 4.80.1)
2159 JH/11 Add A= logging on delivery lines, and a client_set_id option on
2162 JH/12 Add optional authenticated_sender logging to A= and a log_selector
2165 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
2167 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
2168 advertise SMTP AUTH mechanism to us, instead of a generic
2169 protocol violation error. Also, make Exim more robust to bad
2170 data from the Dovecot auth socket.
2172 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
2174 When a queue runner is handling a message, Exim first routes the
2175 recipient addresses, during which it prunes them based on the retry
2176 hints database. After that it attempts to deliver the message to
2177 any remaining recipients. It then updates the hints database using
2180 So if a recipient address works intermittently, it can get repeatedly
2181 deferred at routing time. The retry hints record remains fresh so the
2182 address never reaches the final cutoff time.
2184 This is a fairly common occurrence when a user is bumping up against
2185 their storage quota. Exim had some logic in its local delivery code
2186 to deal with this. However it did not apply to per-recipient defers
2187 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
2189 This change adds a proper retry rule check during routing so that the
2190 final cutoff time is checked against the message's age. We only do
2191 this check if there is an address retry record and there is not a
2192 domain retry record; this implies that previous attempts to handle
2193 the address had the retry_use_local_parts option turned on. We use
2194 this as an approximation for the destination being like a local
2195 delivery, as in LMTP.
2197 I suspect this new check makes the old local delivery cutoff check
2198 redundant, but I have not verified this so I left the code in place.
2200 TF/02 Correct gecos expansion when From: is a prefix of the username.
2202 Test 0254 submits a message to Exim with the header
2206 When I ran the test suite under the user fanf2, Exim expanded
2207 the header to contain my full name, whereas it should have added
2208 a Resent-Sender: header. It erroneously treats any prefix of the
2209 username as equal to the username.
2211 This change corrects that bug.
2213 GF/01 DCC debug and logging tidyup
2214 Error conditions log to paniclog rather than rejectlog.
2215 Debug lines prefixed by "DCC: " to remove any ambiguity.
2217 TF/03 Avoid unnecessary rebuilds of lookup-related code.
2219 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
2220 Bug spotted by Jeremy Harris; was flawed since initial commit.
2221 Would have resulted in OCSP responses post-SNI triggering an Exim
2222 NULL dereference and crash.
2224 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
2226 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
2227 Bug detection, analysis and fix by Samuel Thibault.
2228 Bugzilla 1331, Debian bug #698092.
2230 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
2232 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
2233 Server implementation by Todd Lyons, client by JH.
2234 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
2235 config variable "prdr_enable" controls whether the server
2236 advertises the facility. If the client requests PRDR a new
2237 acl_data_smtp_prdr ACL is called once for each recipient, after
2238 the body content is received and before the acl_smtp_data ACL.
2239 The client is controlled by both of: a hosts_try_prdr option
2240 on the smtp transport, and the server advertisement.
2241 Default client logging of deliveries and rejections involving
2242 PRDR are flagged with the string "PRDR".
2244 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
2245 fclose(). Diagnosis by Todd Lyons.
2247 PP/17 Update configure.default to handle IPv6 localhost better.
2248 Patch by Alain Williams (plus minor tweaks).
2251 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
2252 This is now consistent with GnuTLS, and is now documented: the
2253 previous undocumented portable approach to treating the option as
2254 unset was to force an expansion failure. That still works, and
2255 an empty string is now equivalent.
2257 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
2258 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
2259 not performing validation itself.
2261 PP/20 Added force_command boolean option to pipe transport.
2262 Patch from Nick Koston, of cPanel Inc.
2264 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
2267 TF/04 Added udpsend ACL modifier and hexquote expansion operator
2269 PP/21 Fix eximon continuous updating with timestamped log-files.
2270 Broken in a format-string cleanup in 4.80, missed when I repaired the
2271 other false fix of the same issue.
2272 Report and fix from Heiko Schlichting.
2275 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
2276 Report from Prashanth Katuri.
2278 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
2279 It's SecureTransport, so affects any MacOS clients which use the
2280 system-integrated TLS libraries, including email clients.
2282 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
2283 using a MIME ACL for non-SMTP local injection.
2284 Report and assistance in diagnosis by Warren Baker.
2286 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
2288 JH/16 Fix comparisons for 64b. Bugzilla 1385.
2290 TL/09 Add expansion variable $authenticated_fail_id to keep track of
2291 last id that failed so it may be referenced in subsequent ACL's.
2293 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
2296 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
2297 ldap library initialization, allowing self-signed CA's to be
2298 used. Also properly sets require_cert option later in code by
2299 using NULL (global ldap config) instead of ldap handle (per
2300 session). Bug diagnosis and testing by alxgomz.
2302 TL/12 Enhanced documentation in the ratelimit.pl script provided in
2303 the src/util/ subdirectory.
2305 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
2306 renamed to Transport Post Delivery Action by Jeremy Harris, as
2309 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
2310 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
2311 redis_servers = needs to be configured which will be used by the redis
2312 lookup. Patch from Warren Baker, of The Packet Hub.
2314 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
2316 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
2317 hostname or reverse DNS when processing a host list. Used suggestions
2318 from multiple comments on this bug.
2320 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
2322 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
2323 Missed a few lines, added it to make the runtest require no keyboard
2326 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
2327 contains upper case chars. Make router use caseful_local_part.
2329 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
2330 support when GnuTLS has been built with p11-kit.
2336 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
2338 This, or similar/improved, will also be change PP/11 of 4.82.
2344 PP/01 Handle short writes when writing local log-files.
2345 In practice, only affects FreeBSD (8 onwards).
2346 Bugzilla 1053, with thanks to Dmitry Isaikin.
2348 NM/01 Bugzilla 949 - Documentation tweak
2350 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
2353 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
2355 PP/02 Implemented gsasl authenticator.
2357 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
2359 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
2360 `pkg-config foo` for cflags/libs.
2362 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
2363 with rest of GSASL and with heimdal_gssapi.
2365 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
2366 `pkg-config foo` for cflags/libs for the TLS implementation.
2368 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
2369 properties get this fed in as external SSF. A number of robustness
2370 and debugging improvements to the cyrus_sasl authenticator.
2372 PP/08 cyrus_sasl server now expands the server_realm option.
2374 PP/09 Bugzilla 1214 - Log authentication information in reject log.
2375 Patch by Jeremy Harris.
2377 PP/10 Added dbmjz lookup type.
2379 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
2381 PP/12 MAIL args handles TAB as well as SP, for better interop with
2382 non-compliant senders.
2383 Analysis and variant patch by Todd Lyons.
2385 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
2386 Bug report from Lars Müller <lars@samba.org> (via SUSE),
2387 Patch from Dirk Mueller <dmueller@suse.com>
2389 PP/13 tls_peerdn now print-escaped for spool files.
2390 Observed some $tls_peerdn in wild which contained \n, which resulted
2391 in spool file corruption.
2393 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
2394 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
2395 or write after TLS renegotiation, which otherwise led to messages
2398 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
2399 as a tracking header (ie: a signed header comes before the signature).
2400 Patch from Wolfgang Breyha.
2402 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
2403 comma-sep list; embedded commas doubled.
2405 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
2407 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
2409 Report and patch from Dmitry Banschikov.
2411 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
2412 Removed SSL_clear() after SSL_new() which led to protocol negotiation
2413 failures. We appear to now support TLS1.1+ with Exim.
2415 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
2416 lets Exim select keys and certificates based upon TLS SNI from client.
2417 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
2418 before an outbound SMTP session. New log_selector, +tls_sni.
2420 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
2421 NULL dereference. Report and patch from Alun Jones.
2423 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
2424 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
2425 Not seeing resolver debug output on NetBSD, but suspect this is a
2426 resolver implementation change.
2428 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
2429 Left warnings. Added "eximon gdb" invocation mode.
2431 PP/21 Defaulting "accept_8bitmime" to true, not false.
2433 PP/22 Added -bw for inetd wait mode support.
2435 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
2436 locate the relevant includes and libraries. Made this the default.
2438 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
2439 Bugzilla 1246, report and most of solution from Tomasz Kusy.
2441 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
2442 This may cause build issues on older platforms.
2444 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
2445 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
2446 gnutls_require_mac & gnutls_require_protocols (no longer supported).
2447 Added SNI support via GnuTLS too.
2448 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
2450 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
2452 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
2453 Applied second patch from Janne, implementing suggestion to default
2454 multiple-strings-in-record handling to match SPF spec.
2456 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
2458 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
2459 read-only, out of scope).
2460 Patch from Wolfgang Breyha, report from Stuart Northfield.
2462 PP/29 Fix three issues highlighted by clang analyser static analysis.
2463 Only crash-plausible issue would require the Cambridge-specific
2464 iplookup router and a misconfiguration.
2465 Report from Marcin Mirosław.
2467 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
2469 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
2470 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
2471 As part of this, removing so much warning spew let me fix some minor
2472 real issues in debug logging.
2474 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
2475 assignment on my part. Fixed.
2477 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
2478 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
2479 Janne Snabb (who went above and beyond: thank you).
2481 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
2482 string otherwise requires a connection and a bunch more work and it's
2483 relatively easy to get wrong. Should also expose TLS library linkage
2486 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
2487 64-bit ${eval} (JH/03).
2489 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
2490 GNU libc to support some of the 64-bit stuff, should not lead to
2491 conflicts. Defined before os.h is pulled in, so if a given platform
2492 needs to override this, it can.
2494 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
2495 protection layer was required, which is not implemented.
2496 Bugzilla 1254, patch from Wolfgang Breyha.
2498 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
2499 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
2500 tls_dhparam take prime identifiers. Also unbreak combination of
2501 OpenSSL+DH_params+TLSSNI.
2503 PP/39 Disable SSLv2 by default in OpenSSL support.
2509 PP/01 Solaris build fix for Oracle's LDAP libraries.
2510 Bugzilla 1109, patch from Stephen Usher.
2512 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
2514 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
2517 TF/02 Fix a couple more cases where we did not log the error message
2518 when unlink() failed. See also change 4.74-TF/03.
2520 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
2521 lock up or crash if it happened to be inside a call to libc when it
2522 got a SIGUSR1 from exiwhat.
2524 The SIGUSR1 handler appends the current process status to the process
2525 log which is later printed by exiwhat. It used to use the general
2526 purpose logging code to do this, but several functions it calls are
2527 not safe for signals.
2529 The new output code in the SIGUSR1 handler is specific to the process
2530 log, and simple enough that it's easy to inspect for signal safety.
2531 Removing some special cases also simplifies the general logging code.
2532 Removing the spurious timestamps from the process log simplifies
2535 TF/04 Improved ratelimit ACL condition.
2537 The /noupdate option has been deprecated in favour of /readonly which
2538 has clearer semantics. The /leaky, /strict, and /readonly update modes
2539 are mutually exclusive. The update mode is no longer included in the
2540 database key; it just determines when the database is updated. (This
2541 means that when you upgrade Exim will forget old rate measurements.)
2543 Exim now checks that the per_* options are used with an update mode that
2544 makes sense for the current ACL. For example, when Exim is processing a
2545 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
2546 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
2547 must specify per_mail/readonly. If you omit the update mode it defaults to
2548 /leaky where that makes sense (as before) or /readonly where required.
2550 The /noupdate option is now undocumented but still supported for
2551 backwards compatibility. It is equivalent to /readonly except that in
2552 ACLs where /readonly is required you may specify /leaky/noupdate or
2553 /strict/noupdate which are treated the same as /readonly.
2555 A useful new feature is the /count= option. This is a generalization
2556 of the per_byte option, so that you can measure the throughput of other
2557 aggregate values. For example, the per_byte option is now equivalent
2558 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
2560 The per_rcpt option has been generalized using the /count= mechanism
2561 (though it's more complicated than the per_byte equivalence). When it is
2562 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
2563 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
2564 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
2565 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
2566 Note that using per_rcpt with a non-readonly update mode in more than
2567 one ACL will cause the recipients to be double-counted. (The per_mail
2568 and per_byte options don't have this problem.)
2570 The handling of very low rates has changed slightly. If the computed rate
2571 is less than the event's count (usually one) then this event is the first
2572 after a long gap. In this case the rate is set to the same as this event's
2573 count, so that the first message of a spam run is counted properly.
2575 The major new feature is a mechanism for counting the rate of unique
2576 events. The new per_addr option counts the number of different
2577 recipients that someone has sent messages to in the last time period. It
2578 behaves like per_rcpt if all the recipient addresses are different, but
2579 duplicate recipient addresses do not increase the measured rate. Like
2580 the /count= option this is a general mechanism, so the per_addr option
2581 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
2582 example, measure the rate that a client uses different sender addresses
2583 with the options per_mail/unique=$sender_address. There are further
2584 details in the main documentation.
2586 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
2588 TF/06 Removed a few PCRE remnants.
2590 TF/07 Automatically extract Exim's version number from tags in the git
2591 repository when doing development or release builds.
2593 PP/02 Raise smtp_cmd_buffer_size to 16kB.
2594 Bugzilla 879. Patch from Paul Fisher.
2596 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
2597 Heavily based on revision 40f9a89a from Simon Arlott's tree.
2600 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
2602 PP/05 Variable $av_failed, true if the AV scanner deferred.
2603 Bugzilla 1078. Patch from John Horne.
2605 PP/06 Stop make process more reliably on build failure.
2606 Bugzilla 1087. Patch from Heiko Schlittermann.
2608 PP/07 Make maildir_use_size_file an _expandable_ boolean.
2609 Bugzilla 1089. Patch from Heiko Schlittermann.
2611 PP/08 Handle ${run} returning more data than OS pipe buffer size.
2612 Bugzilla 1131. Patch from Holger Weiß.
2614 PP/09 Handle IPv6 addresses with SPF.
2615 Bugzilla 860. Patch from Wolfgang Breyha.
2617 PP/10 GnuTLS: support TLS 1.2 & 1.1.
2619 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
2622 PP/11 match_* no longer expand right-hand-side by default.
2623 New compile-time build option, EXPAND_LISTMATCH_RHS.
2624 New expansion conditions, "inlist", "inlisti".
2626 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
2628 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
2630 PP/14 fix log_write() format string regression from TF/03.
2631 Bugzilla 1152. Patch from Dmitry Isaikin.
2637 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
2639 PP/02 Harmonised TLS library version reporting; only show if debugging.
2640 Layout now matches that introduced for other libraries in 4.74 PP/03.
2642 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
2644 PP/04 New "dns_use_edns0" global option.
2646 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
2649 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
2650 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
2652 TK/01 Updated PolarSSL code to 0.14.2.
2653 Bugzilla 1097. Patch from Andreas Metzler.
2655 PP/07 Catch divide-by-zero in ${eval:...}.
2656 Fixes bugzilla 1102.
2658 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
2661 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
2662 format-string attack -- SECURITY: remote arbitrary code execution.
2664 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
2665 time unintentionally subject to list matching rules, letting the header
2666 cause arbitrary Exim lookups (of items which can occur in lists, *not*
2667 arbitrary string expansion). This allowed for information disclosure.
2669 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
2670 INT_MIN/-1 -- value coerced to INT_MAX.
2676 NM/01 Workaround for PCRE version dependency in version reporting
2679 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
2680 This fixes portability to compilers other than gcc, notably
2681 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
2683 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
2684 makefiles for portability to HP-UX and POSIX correctness.
2686 PP/01 Permit LOOKUP_foo enabling on the make command-line.
2687 Also via indented variable definition in the Makefile.
2688 (Debugging by Oliver Heesakkers).
2690 PP/02 Restore caching of spamd results with expanded spamd_address.
2691 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
2693 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
2694 Improves build reliability. Fix from: Frank Elsner
2696 NM/02 Fix wide character breakage in the rfc2047 coding
2697 Fixes bug 1064. Patch from Andrey N. Oktyabrski
2699 NM/03 Allow underscore in dnslist lookups
2700 Fixes bug 1026. Patch from Graeme Fowler
2702 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
2703 Code patches from Adam Ciarcinski of NetBSD.
2705 NM/04 Fixed exiqgrep to cope with mailq missing size issue
2708 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
2709 is logged, to avoid truncation. Patch from John Horne.
2711 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
2712 Patch from Jakob Hirsch.
2714 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
2715 SQL string expansion failure details.
2716 Patch from Andrey Oktyabrski.
2718 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
2719 Patch from Simon Arlott.
2721 PP/09 New lookups functionality failed to compile on old gcc which rejects
2722 extern declarations in function scope.
2723 Patch from Oliver Fleischmann
2725 PP/10 Use sig_atomic_t for flags set from signal handlers.
2726 Check getgroups() return and improve debugging.
2727 Fixed developed for diagnosis in bug 927 (which turned out to be
2730 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
2731 Patch from Mark Zealey.
2733 PP/12 Bugzilla 1056: Improved spamd server selection.
2734 Patch from Mark Zealey.
2736 PP/13 Bugzilla 1086: Deal with maildir quota file races.
2737 Based on patch from Heiko Schlittermann.
2739 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
2740 Patch from Uwe Doering, sign-off by Michael Haardt.
2742 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
2743 variable declaration deep within a block. Bug and patch from
2746 PP/15 lookups-Makefile IRIX compatibility coercion.
2748 PP/16 Make DISABLE_DKIM build knob functional.
2750 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
2751 Patch by Simon Arlott
2753 TF/03 Fix valgrind.h portability to C89 compilers that do not support
2754 variable argument macros. Our copy now differs from upstream.
2760 TF/01 Failure to get a lock on a hints database can have serious
2761 consequences so log it to the panic log.
2763 TF/02 Log LMTP confirmation messages in the same way as SMTP,
2764 controlled using the smtp_confirmation log selector.
2766 TF/03 Include the error message when we fail to unlink a spool file.
2768 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
2769 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
2770 for maintaining out-of-tree patches for some time.
2772 PP/01 Bugzilla 139: Documentation and portability issues.
2773 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
2774 Handle per-OS dynamic-module compilation flags.
2776 PP/02 Let /dev/null have normal permissions.
2777 The 4.73 fixes were a little too stringent and complained about the
2778 permissions on /dev/null. Exempt it from some checks.
2779 Reported by Andreas M. Kirchwitz.
2781 PP/03 Report version information for many libraries, including
2782 Exim version information for dynamically loaded libraries. Created
2783 version.h, now support a version extension string for distributors
2784 who patch heavily. Dynamic module ABI change.
2786 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
2787 privilege escalation vulnerability whereby the Exim run-time user
2788 can cause root to append content of the attacker's choosing to
2791 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
2794 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
2795 If dropping privileges for untrusted macros, we disabled normal logging
2796 on the basis that it would fail; for the Exim run-time user, this is not
2797 the case, and it resulted in successful deliveries going unlogged.
2798 Fixed. Reported by Andreas Metzler.
2804 PP/01 Date: & Message-Id: revert to normally being appended to a message,
2805 only prepend for the Resent-* case. Fixes regression introduced in
2806 Exim 4.70 by NM/22 for Bugzilla 607.
2808 PP/02 Include check_rfc2047_length in configure.default because we're seeing
2809 increasing numbers of administrators be bitten by this.
2811 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
2813 PP/03 Bugzilla 994: added openssl_options main configuration option.
2815 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
2817 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
2819 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
2821 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
2822 without a peer certificate, leading to a segfault because of an
2823 assumption that peers always have certificates. Be a little more
2824 paranoid. Problem reported by Martin Tscholak.
2826 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
2827 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
2828 NB: ClamAV planning to remove STREAM in "middle of 2010".
2829 CL also introduces -bmalware, various -d+acl logging additions and
2830 more caution in buffer sizes.
2832 PP/09 Implemented reverse_ip expansion operator.
2834 PP/10 Bugzilla 937: provide a "debug" ACL control.
2836 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
2838 PP/12 Bugzilla 973: Implement --version.
2840 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2842 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2844 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2846 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2847 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2848 ignore trailing whitespace.
2850 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2852 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2855 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2856 Notification from Dr Andrew Aitchison.
2858 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2859 ExtendedDetectionInfo response format.
2860 Notification from John Horne.
2862 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2865 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2866 XSL and documented dependency on system catalogs, with examples of how
2869 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2872 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2873 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2874 configuration file which is writeable by the Exim user or group.
2876 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2877 of configuration files to cover files specified with the -C option if
2878 they are going to be used with root privileges, not just the default
2881 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2882 option (effectively making it always true).
2884 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2885 files to be used while preserving root privileges.
2887 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2888 that rogue child processes cannot use them.
2890 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2891 run-time user, instead of root.
2893 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2894 Exim run-time user without dropping privileges.
2896 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2897 result string, instead of calling string_vformat() twice with the same
2900 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2901 for other users. Others should always drop root privileges if they use
2902 -C on the command line, even for a whitelisted configure file.
2904 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2906 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2912 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2913 $data_path, and $header_path variables; fixed documentation bugs and
2916 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2917 exipick to access non-standard spools, including the "frozen" queue
2920 NM/01 Bugzilla 965: Support mysql stored procedures.
2921 Patch from Alain Williams
2923 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2925 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2926 Patch from Andreas Metzler
2928 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2929 Patch from Kirill Miazine
2931 NM/05 Bugzilla 671: Added umask to procmail example.
2933 JJ/03 installed exipick 20100323.0, fixing doc bug
2935 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2936 directory. Notification and patch from Dan Rosenberg.
2938 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2940 TK/02 Improve log output when DKIM signing operation fails.
2942 MH/01 Treat the transport option dkim_domain as a colon separated
2943 list, not as a single string, and sign the message with each element,
2944 omitting multiple occurences of the same signer.
2946 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2947 Bugzilla 985, 986. Patch by Simon Arlott
2949 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2950 Patch by Simon Arlott
2952 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2953 MBX locking. Notification from Dan Rosenberg.
2959 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2961 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2963 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2965 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2967 NM/04 Bugzilla 915: Flag broken perl installation during build.
2973 TK/01 Added patch by Johannes Berg that expands the main option
2974 "spamd_address" if it starts with a dollar sign.
2976 TK/02 Write list of recipients to X-Envelope-Sender header when building
2977 the mbox-format spool file for content scanning (suggested by Jakob
2980 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2981 (http://www.dcc-servers.net/) support via dccifd. Activated by
2982 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2984 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2985 by Mark Daniel Reidel <mr@df.eu>.
2987 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2988 When building exim an external PCRE library is now needed -
2989 PCRE is a system library on the majority of modern systems.
2990 See entry on PCRE_LIBS in EDITME file.
2992 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2993 conversation. Added nologin parameter to request.
2994 Patch contributed by Kirill Miazine.
2996 TF/01 Do not log submission mode rewrites if they do not change the address.
2998 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
3000 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
3001 log files in place. Contributed by Roberto Lima.
3003 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
3005 TF/03 Bugzilla 615: When checking the local_parts router precondition
3006 after a local_part_suffix or local_part_prefix option, Exim now
3007 does not use the address's named list lookup cache, since this
3008 contains cached lookups for the whole local part.
3010 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
3011 Robert Millan. Documentation is in experimental-spec.txt.
3013 TF/04 Bugzilla 668: Fix parallel build (make -j).
3015 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
3017 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
3018 Patch provided by Jan Srzednicki.
3020 TF/05 Leading white space used to be stripped from $spam_report which
3021 wrecked the formatting. Now it is preserved.
3023 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
3024 that they are available at delivery time.
3026 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
3028 TF/08 TLS error reporting now respects the incoming_interface and
3029 incoming_port log selectors.
3031 TF/09 Produce a more useful error message if an SMTP transport's hosts
3032 setting expands to an empty string.
3034 NM/06 Bugzilla 744: EXPN did not work under TLS.
3035 Patch provided by Phil Pennock.
3037 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
3038 Patch provided by Richard Godbee.
3040 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
3041 acl_smtp_notquit, added index entry.
3043 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
3044 Patch provided by Eugene Bujak.
3046 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
3047 accept(). Patch provided by Maxim Dounin.
3049 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
3050 Patch provided by Phil Pennock.
3052 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
3054 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
3055 Patch provided by Brad "anomie" Jorsch.
3057 NM/14 Bugzilla 622: Added timeout setting to transport filter.
3058 Patch provided by Dean Brooks.
3060 TK/05 Add native DKIM support (does not depend on external libraries).
3062 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
3063 Patch provided by Graeme Fowler.
3065 NM/16 Bugzilla 851: Documentation example syntax fix.
3067 NM/17 Changed NOTICE file to remove references to embedded PCRE.
3069 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
3072 NM/19 Bugzilla 745: TLS version reporting.
3073 Patch provided by Phil Pennock.
3075 NM/20 Bugzilla 167: bool: condition support.
3076 Patch provided by Phil Pennock.
3078 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
3079 clients. Patch provided by Phil Pennock.
3081 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
3082 Patch provided by Brad "anomie" Jorsch.
3084 NM/23 Bugzilla 687: Fix misparses in eximstats.
3085 Patch provided by Heiko Schlittermann.
3087 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
3088 Patch provided by Heiko Schlittermann.
3090 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
3091 plus update to original patch.
3093 NM/26 Bugzilla 799: Documentation correction for ratelimit.
3095 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
3096 Patch provided by David Brownlee.
3098 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
3100 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
3102 NM/30 Bugzilla 888: TLS documentation bugfixes.
3104 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
3106 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
3107 Unlike the original bugzilla I have changed all shell scripts in src tree.
3109 NM/33 Bugzilla 898: Transport filter timeout fix.
3110 Patch by Todd Rinaldo.
3112 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
3113 Patch by Serge Demonchaux.
3115 NM/35 Bugzilla 39: Base64 decode bug fixes.
3116 Patch by Jakob Hirsch.
3118 NM/36 Bugzilla 909: Correct connect() call in dcc code.
3120 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
3122 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
3124 NM/39 Bugzilla 911: Fixed MakeLinks build script.
3130 TK/01 Add preliminary DKIM support. Currently requires a forked version of
3131 ALT-N's libdkim that I have put here:
3132 http://duncanthrax.net/exim-experimental/
3134 Note to Michael Haardt: I had to rename some vars in sieve.c. They
3135 were called 'true' and it seems that C99 defines that as a reserved
3136 keyword to be used with 'bool' variable types. That means you could
3137 not include C99-style headers which use bools without triggering
3138 build errors in sieve.c.
3140 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
3141 as mailq or other aliases. Changed the --help handling significantly
3142 to do whats expected. exim_usage() emits usage/help information.
3144 SC/01 Added the -bylocaldomain option to eximstats.
3146 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
3148 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
3150 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
3156 PH/01 Another patch from the Sieve maintainer.
3158 PH/02 When an IPv6 address is converted to a string for single-key lookup
3159 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
3160 dots are used instead of colons so that keys in lsearch files need not
3161 contain colons. This was done some time before quoting was made available
3162 in lsearch files. However, iplsearch files do require colons in IPv6 keys
3163 (notated using the quote facility) so as to distinguish them from IPv4
3164 keys. This meant that lookups for IP addresses in host lists did not work
3165 for iplsearch lookups.
3167 This has been fixed by arranging for IPv6 addresses to be expressed with
3168 colons if the lookup type is iplsearch. This is not incompatible, because
3169 previously such lookups could never work.
3171 The situation is now rather anomalous, since one *can* have colons in
3172 ordinary lsearch keys. However, making the change in all cases is
3173 incompatible and would probably break a number of configurations.
3175 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
3178 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
3179 conversion specification without a maximum field width, thereby enabling
3180 a rogue spamd server to cause a buffer overflow. While nobody in their
3181 right mind would setup Exim to query an untrusted spamd server, an
3182 attacker that gains access to a server running spamd could potentially
3183 exploit this vulnerability to run arbitrary code as the Exim user.
3185 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
3186 $primary_hostname instead of what libspf2 thinks the hosts name is.
3188 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
3189 a directory entry by the name of the lookup key. Previously, if a
3190 symlink pointed to a non-existing file or a file in a directory that
3191 Exim lacked permissions to read, a lookup for a key matching that
3192 symlink would fail. Now it is enough that a matching directory entry
3193 exists, symlink or not. (Bugzilla 503.)
3195 PH/03 The body_linecount and body_zerocount variables are now exported in the
3198 PH/04 Added the $dnslist_matched variable.
3200 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
3201 This means they are set thereafter only if the connection becomes
3204 PH/06 Added the client_condition to authenticators so that some can be skipped
3205 by clients under certain conditions.
3207 PH/07 The error message for a badly-placed control=no_multiline_responses left
3208 "_responses" off the end of the name.
3210 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
3212 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
3213 (without spaces) instead of just copying the configuration text.
3215 PH/10 Added the /noupdate option to the ratelimit ACL condition.
3217 PH/11 Added $max_received_linelength.
3219 PH/12 Added +ignore_defer and +include_defer to host lists.
3221 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
3222 way in which PCRE > 7.0 is built.
3224 PH/14 Implemented queue_only_load_latch.
3226 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
3227 MAIL command. The effect was to mangle the value on 64-bit systems.
3229 PH/16 Another patch from the Sieve maintainer.
3231 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
3233 PH/18 If a system quota error occurred while trying to create the file for
3234 a maildir delivery, the message "Mailbox is full" was not appended to the
3235 bounce if the delivery eventually timed out. Change 4.67/27 below applied
3236 only to a quota excession during the actual writing of the file.
3238 PH/19 It seems that peer DN values may contain newlines (and other non-printing
3239 characters?) which causes problems in log lines. The DN values are now
3240 passed through string_printing() before being added to log lines.
3242 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
3243 and InterBase are left for another time.)
3245 PH/21 Added message_body_newlines option.
3247 PH/22 Guard against possible overflow in moan_check_errorcopy().
3249 PH/23 POSIX allows open() to be a macro; guard against that.
3251 PH/24 If the recipient of an error message contained an @ in the local part
3252 (suitably quoted, of course), incorrect values were put in $domain and
3253 $local_part during the evaluation of errors_copy.
3259 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
3260 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
3263 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
3264 issue a MAIL command.
3266 PH/02 In an ACL statement such as
3268 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
3270 if a client was not listed at all, or was listed with a value other than
3271 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
3272 the condition was not true (as it should be), so access was not denied.
3273 The bug was that the ! inversion was incorrectly passed on to the second
3274 item. This has been fixed.
3276 PH/03 Added additional dnslists conditions == and =& which are different from
3277 = and & when the dns lookup returns more than one IP address.
3279 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
3280 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
3282 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
3283 FSYNC, which compiles an option called disable_fsync that allows for
3284 bypassing fsync(). The documentation is heavily laced with warnings.
3286 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
3288 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
3289 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
3290 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
3291 including adding "make clean"; (3) Added -fPIC when compiling the test
3292 dynamically loaded module, to get rid of a warning.
3294 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
3295 message fails, move_frozen_messages = true and ignore_bounce_errors_after
3296 = 0s. The bug is otherwise harmless.
3298 PH/07 There was a bug in the dovecot authenticator such that the value of
3299 $auth1 could be overwritten, and so not correctly preserved, after a
3300 successful authentication. This usually meant that the value preserved by
3301 the server_setid option was incorrect.
3303 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
3305 PH/09 Installed PCRE release 7.0.
3307 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
3308 run for batched SMTP input. It is now run at the start of every message
3309 in the batch. While fixing this I discovered that the process information
3310 (output by running exiwhat) was not always getting set for -bs and -bS
3311 input. This is fixed, and it now also says "batched" for BSMTP.
3313 PH/11 Added control=no_pipelining.
3315 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
3316 patch, slightly modified), and move the expansion of helo_data till after
3317 the connection is made in the smtp transport (so it can use these
3320 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
3322 PH/14 Added log_selector = +pid.
3324 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
3326 PH/16 Add ${if forany and ${if forall.
3328 PH/17 Added dsn_from option to vary the From: line in DSNs.
3330 PH/18 Flush SMTP output before performing a callout, unless control =
3331 no_callout_flush is set.
3333 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
3334 was true (the default) a successful delivery failed to delete the retry
3335 item, thus causing premature timeout of the address. The bug is now
3338 PH/20 Added hosts_avoid_pipelining to the smtp transport.
3340 PH/21 Long custom messages for fakedefer and fakereject are now split up
3341 into multiline responses in the same way that messages for "deny" and
3342 other ACL rejections are.
3344 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
3345 with slight modification.
3347 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
3348 draft, changing the syntax and factoring some duplicate code".
3350 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
3351 for deliveries of the second and subsequent messages over the same SMTP
3354 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
3355 ${reduce, with only minor "tidies".
3357 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
3359 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
3360 expansion side effects.
3362 PH/27 When a message times out after an over-quota error from an Exim-imposed
3363 quota, the bounce message says "mailbox is full". This message was not
3364 being given when it was a system quota that was exceeded. It now should
3367 MH/03 Made $recipients available in local_scan(). local_scan() already has
3368 better access to the recipient list through recipients_list[], but
3369 $recipients can be useful in postmaster-provided expansion strings.
3371 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
3372 in the case of a MAIL command with additional options following the
3373 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
3374 were accidentally chopped off.
3376 PH/29 SMTP synchronization checks are implemented when a command is read -
3377 there is a check that no more input is waiting when there shouldn't be
3378 any. However, for some commands, a delay in an ACL can mean that it is
3379 some time before the response is written. In this time, more input might
3380 arrive, invalidly. So now there are extra checks after an ACL has run for
3381 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
3382 pipelining has not been advertised.
3384 PH/30 MH's patch to allow iscntrl() characters to be list separators.
3386 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
3387 returned in the SMTP response when smtp_return_error_details was false.
3388 This has been fixed.
3390 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
3391 instead of the C I/O that was originally supplied, because problems were
3392 reported on Solaris.
3394 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
3395 Exim which did not show up earlier: it was assuming that a call to
3396 SSL_CTX_set_info_callback() might give an error value. In fact, there is
3397 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
3398 was a macro that became an assignment, so it seemed to work. This has
3399 changed to a proper function call with a void return, hence the compile
3400 error. Exim's code has been fixed.
3402 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
3405 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
3407 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
3409 PH/37 If a message is not accepted after it has had an id assigned (e.g.
3410 because it turns out to be too big or there is a timeout) there is no
3411 "Completed" line in the log. When some messages of this type were
3412 selected by exigrep, they were listed as "not completed". Others were
3413 picked up by some special patterns. I have improved the selection
3414 criteria to be more general.
3416 PH/38 The host_find_failed option in the manualroute router can now be set
3417 to "ignore", to completely ignore a host whose IP address cannot be
3418 found. If all hosts are ignored, the behaviour is controlled by the new
3419 host_all_ignored option.
3421 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
3422 homing or because of multiple MX records with /mx) generated more than
3423 one IP address, and the following item turned out to be the local host,
3424 all the secondary addresses of the first item were incorrectly removed
3425 from the list, along with the local host and any following hosts (which
3426 is what is supposed to happen).
3428 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
3429 whoever called Exim into the -H file. In the case of the daemon it was
3430 behaving confusingly. When first started, it used values for whoever
3431 started the daemon, but after a SIGHUP it used the Exim user (because it
3432 calls itself on a restart). I have changed the code so that it now always
3435 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
3436 message are rejected with the same error (e.g. no authentication or bad
3437 sender address), and a DATA command is nevertheless sent (as can happen
3438 with PIPELINING or a stupid MUA), the error message that was given to the
3439 RCPT commands is included in the rejection of the DATA command. This is
3440 intended to be helpful for MUAs that show only the final error to their
3443 PH/42 Another patch from the Sieve maintainer.
3445 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
3446 Eximstats - Fixed some broken HTML links and added missing column headers
3448 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
3449 columns for Rejects, Temp Rejects, Ham, and Spam rows.
3451 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
3453 PH/43 Yet another patch from the Sieve maintainer.
3455 PH/44 I found a way to check for a TCP/IP connection going away before sending
3456 the response to the final '.' that terminates a message, but only in the
3457 case where the client has not sent further data following the '.'
3458 (unfortunately, this is allowed). However, in many cases there won't be
3459 any further data because there won't be any more messages to send. A call
3460 to select() can be used: if it shows that the input is "ready", there is
3461 either input waiting, or the socket has been closed. An attempt to read
3462 the next input character can distinguish the two cases. Previously, Exim
3463 would have sent an OK response which the client would never have see.
3464 This could lead to message repetition. This fix should cure that, at
3465 least in a lot of common cases.
3467 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
3468 advertised in response to EHLO.
3474 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
3475 fixed by 4.65/MH/01 (is this a record?) are fixed:
3477 (i) An empty string was always treated as zero by the numeric comparison
3478 operators. This behaviour has been restored.
3480 (ii) It is documented that the numeric comparison operators always treat
3481 their arguments as decimal numbers. This was broken in that numbers
3482 starting with 0 were being interpreted as octal.
3484 While fixing these problems I realized that there was another issue that
3485 hadn't been noticed. Values of message_size_limit (both the global option
3486 and the transport option) were treated as octal if they started with 0.
3487 The documentation was vague. These values are now always treated as
3488 decimal, and I will make that clear in the documentation.
3494 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
3495 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
3498 MH/01 Don't check that the operands of numeric comparison operators are
3499 integers when their expansion is in "skipping" mode (fixes bug
3500 introduced by 4.64-PH/07).
3502 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
3503 child addresses, Exim now panics and dies. Previously, because the count
3504 is held in a short int, deliveries were likely to be lost. As such a
3505 large number of recipients for a single message is ridiculous
3506 (performance will be very, very poor), I have chosen to impose a limit
3507 rather than extend the field.
3513 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
3514 leftover -K file (the existence of which was triggered by #402).
3515 While we were at it, introduced process PID as part of the -K
3516 filename. This should rule out race conditions when creating
3519 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
3520 processing considerably. Previous code took too long for large mails,
3521 triggering a timeout which in turn triggers #401.
3523 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
3524 in the DK code in transports.c. sendfile() is not really portable,
3525 hence the _LINUX specificness.
3527 TF/01 In the add_headers option to the mail command in an Exim filter,
3528 there was a bug that Exim would claim a syntax error in any
3529 header after the first one which had an odd number of characters
3532 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
3533 callout verification, Exim cached a "reject" for the entire domain. This
3534 is correct for most verifications, but it is not correct for a recipient
3535 verification with use_sender or use_postmaster set, because in that case
3536 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
3537 case of MAIL FROM:<> rejection from other early rejections (e.g.
3538 rejection of HELO). When verifying a recipient using a non-null MAIL
3539 address, the cache is ignored if it shows MAIL FROM:<> rejection.
3540 Whatever the result of the callout, the value of the domain cache is
3541 left unchanged (for any other kind of callout, getting as far as trying
3542 RCPT means that the domain itself is ok).
3544 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
3547 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
3548 manifest itself as EPIPE rather than ECONNECT. When tidying away a
3549 session, the daemon ignores ECONNECT errors and logs others; it now
3550 ignores EPIPE as well.
3552 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
3553 (quoted-printable decoding).
3555 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
3556 later the small subsequent patch to fix an introduced bug.
3558 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
3560 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
3562 PH/08 An error is now given if message_size_limit is specified negative.
3564 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
3565 to be given (somewhat) arbitrary names.
3567 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
3570 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
3571 miscellaneous code fixes
3573 PH/10 Added the log_reject_target ACL modifier to specify where to log
3576 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
3577 hostname. This is wrong, because it relates to the incoming message (and
3578 probably the interface on which it is arriving) and not to the outgoing
3579 callout (which could be using a different interface). This has been
3580 changed to use the value of the helo_data option from the smtp transport
3581 instead - this is what is used when a message is actually being sent. If
3582 there is no remote transport (possible with a router that sets up host
3583 addresses), $smtp_active_hostname is used.
3585 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
3586 tweaks were necessary in order to get it to work (see also 21 below):
3587 (a) The code assumed that strncpy() returns a negative number on buffer
3588 overflow, which isn't the case. Replaced with Exim's string_format()
3590 (b) There were several signed/unsigned issues. I just did the minimum
3591 hacking in of casts. There is scope for a larger refactoring.
3592 (c) The code used strcasecmp() which is not a standard C function.
3593 Replaced with Exim's strcmpic() function.
3594 (d) The code set only $1; it now sets $auth1 as well.
3595 (e) A simple test gave the error "authentication client didn't specify
3596 service in request". It would seem that Dovecot has changed its
3597 interface. Fortunately there's a specification; I followed it and
3598 changed what the client sends and it appears to be working now.
3600 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
3603 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
3604 address A is aliased to B and C, where B exists and C does not. Without
3605 -v the output is "A verified" because verification stops after a
3606 successful redirection if more than one address is generated. However,
3607 with -v the child addresses are also verified. Exim was outputting "A
3608 failed to verify" and then showing the successful verification for C,
3609 with its parentage. It now outputs "B failed to verify", showing B's
3610 parentage before showing the successful verification of C.
3612 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
3613 look up a TXT record in a specific list after matching in a combined
3616 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
3617 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
3618 they consult the DNS. I had assumed they would set it the way they
3619 wanted; and indeed my experiments on Linux seem to show that in some
3620 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
3621 To be on the safe side, however, I have now made the interface to
3622 host_find_byname() similar to host_find_bydns(), with an argument
3623 containing the DNS resolver options. The host_find_byname() function now
3624 sets these options at its start, just as host_find_bydns() does. The smtp
3625 transport options dns_qualify_single and dns_search_parents are passed to
3626 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
3627 of host_find_byname() use the default settings of RES_DEFNAMES
3628 (qualify_single) but not RES_DNSRCH (search_parents).
3630 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
3631 spool_read_header() do less string testing, by means of a preliminary
3632 switch on the second character of optional "-foo" lines. (This is
3633 overdue, caused by the large number of possibilities that now exist.
3634 Originally there were few.) While I was there, I also converted the
3635 str(n)cmp tests so they don't re-test the leading "-" and the first
3636 character, in the hope this might squeeze out yet more improvement.
3638 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
3639 flag allowing group syntax was set by the header_syntax check but not
3640 turned off, possible causing trouble later; (2) The flag was not being
3641 set at all for the header_verify test, causing "group"-style headers to
3642 be rejected. I have now set it in this case, and also caused header_
3643 verify to ignore an empty address taken from a group. While doing this, I
3644 came across some other cases where the code for allowing group syntax
3645 while scanning a header line wasn't quite right (mostly, not resetting
3646 the flag correctly in the right place). These bugs could have caused
3647 trouble for malformed header lines. I hope it is now all correct.
3649 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
3650 with the "reply" argument non-NULL. The code, however (which originally
3651 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
3652 but it didn't always do it. This confused somebody who was copying the
3653 code for some other use. I have removed all the tests.
3655 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
3656 feature that was used to support insecure browsers during the U.S. crypto
3657 embargo. It requires special client support, and Exim is probably the
3658 only MTA that supported it -- and would never use it because real RSA is
3659 always available. This code has been removed, because it had the bad
3660 effect of slowing Exim down by computing (never used) parameters for the
3661 RSA_EXPORT functionality.
3663 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
3664 authenticator to fail if there's a tab character in the incoming data
3665 (there should never be unless someone is messing about, as it's supposed
3666 to be base64-encoded). Also added, on Timo's advice, the "secured" option
3667 if the connection is using TLS or if the remote IP is the same as the
3668 local IP, and the "valid-client-cert option" if a client certificate has
3671 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
3672 authenticators. This can be used for authorization after authentication
3673 succeeds. (In the case of plaintext, it servers for both authentication
3676 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
3677 if any retry times were supplied.
3679 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
3680 connection, where there is no client IP address to check. In this
3681 situation, the verify now always succeeds.
3683 PH/25 Applied John Jetmore's -Mset patch.
3685 PH/26 Added -bem to be like -Mset, but loading a message from a file.
3687 PH/27 In a string expansion for a processed (not raw) header when multiple
3688 headers of the same name were present, leading whitespace was being
3689 removed from all of them, but trailing whitespace was being removed only
3690 from the last one. Now trailing whitespace is removed from each header
3691 before concatenation. Completely empty headers in a concatenation (as
3692 before) are ignored.
3694 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
3695 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
3697 PH/29 [Removed. This was a change that I later backed out, and forgot to
3698 correct the ChangeLog entry (that I had efficiently created) before
3699 committing the later change.]
3701 PH/30 Exim was sometimes attempting to deliver messages that had suffered
3702 address errors (4xx response to RCPT) over the same connection as other
3703 messages routed to the same hosts. Such deliveries are always "forced",
3704 so retry times are not inspected. This resulted in far too many retries
3705 for the affected addresses. The effect occurred only when there were more
3706 hosts than the hosts_max_try setting in the smtp transport when it had
3707 the 4xx errors. Those hosts that it had tried were not added to the list
3708 of hosts for which the message was waiting, so if all were tried, there
3709 was no problem. Two fixes have been applied:
3711 (i) If there are any address or message errors in an SMTP delivery, none
3712 of the hosts (tried or untried) are now added to the list of hosts
3713 for which the message is waiting, so the message should not be a
3714 candidate for sending over the same connection that was used for a
3715 successful delivery of some other message. This seems entirely
3716 reasonable: after all the message is NOT "waiting for some host".
3717 This is so "obvious" that I'm not sure why it wasn't done
3718 previously. Hope I haven't missed anything, but it can't do any
3719 harm, as the worst effect is to miss an optimization.
3721 (ii) If, despite (i), such a delivery is accidentally attempted, the
3722 routing retry time is respected, so at least it doesn't keep
3723 hammering the server.
3725 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
3726 in ${readsocket because some servers need this prod.
3728 PH/32 Added some extra debug output when updating a wait-xxx database.
3730 PH/33 The hint "could be header name not terminated by colon", which has been
3731 given for certain expansion errors for a long time, was not being given
3732 for the ${if def:h_colon_omitted{... case.
3734 PH/34 The spec says: "With one important exception, whenever a domain list is
3735 being scanned, $domain contains the subject domain." There was at least
3736 one case where this was not true.
3738 PH/35 The error "getsockname() failed: connection reset by peer" was being
3739 written to the panic log as well as the main log, but it isn't really
3740 panic-worthy as it just means the connection died rather early on. I have
3741 removed the panic log writing for the ECONNRESET error when getsockname()
3744 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
3745 runs only) independently of the message's sender address. This meant
3746 that, if the 4xx error was in fact related to the sender, a different
3747 message to the same recipient with a different sender could confuse
3748 things. In particular, this can happen when sending to a greylisting
3749 server, but other circumstances could also provoke similar problems.
3750 I have changed the default so that the retry time for these errors is now
3751 based a combination of the sender and recipient addresses. This change
3752 can be overridden by setting address_retry_include_sender=false in the
3755 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
3756 remote server are returned as part of bounce messages. This was not
3757 happening for LMTP over a pipe (the lmtp transport), but now it is the
3758 same for both kinds of LMTP.
3760 PH/38 Despite being documented as not happening, Exim was rewriting addresses
3761 in header lines that were in fact CNAMEs. This is no longer the case.
3763 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
3764 and queue runs started by the daemon processed all messages. This has
3765 been fixed so that -R and -S can now usefully be given with -q<time>.
3767 PH/40 Import PCRE release 6.7 (fixes some bugs).
3769 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
3771 PH/42 Give an error if -q is specified more than once.
3773 PH/43 Renamed the variables $interface_address and $interface_port as
3774 $received_ip_address and $received_port, to make it clear that these
3775 values apply to message reception, and not to the outgoing interface when
3776 a message is delivered. (The old names remain recognized, of course.)
3778 PH/44 There was no timeout on the connect() call when using a Unix domain
3779 socket in the ${readsocket expansion. There now is.
3781 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
3782 be meaningful with "accept".
3784 SC/01 Eximstats V1.43
3785 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
3787 SC/02 Eximstats V1.44
3788 Use a glob alias rather than an array ref in the generated
3789 parser. This improves both readability and performance.
3791 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
3792 Collect SpamAssassin and rejection statistics.
3793 Don't display local sender or destination tables unless
3794 there is data to show.
3795 Added average volumes into the top table text output.
3797 SC/04 Eximstats V1.46
3798 Collect data on the number of addresses (recipients)
3799 as well as the number of messages.
3801 SC/05 Eximstats V1.47
3802 Added 'Message too big' to the list of mail rejection
3803 reasons (thanks to Marco Gaiarin).
3805 SC/06 Eximstats V1.48
3806 Mainlog lines which have GMT offsets and are too short to
3807 have a flag are now skipped.
3809 SC/07 Eximstats V1.49 (Alain Williams)
3810 Added the -emptyok flag.
3812 SC/08 Eximstats V1.50
3813 Fixes for obtaining the IP address from reject messages.
3815 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
3816 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
3817 whitespace changes from 4.64-PH/27
3819 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
3822 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
3823 are found, allow negative numbers in numeric criteria)
3825 JJ/06 exipick.20061117.2, added new $message_body_missing variable
3827 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
3828 to match changes made in 4.64-PH/43
3830 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
3832 PH/47 Put in an explicit test for a DNS lookup of an address record where the
3833 "domain" is actually an IP address, and force a failure. This locks out
3834 those revolvers/nameservers that support "A-for-A" lookups, in
3835 contravention of the specifications.
3837 PH/48 When a host name was looked up from an IP address, and the subsequent
3838 forward lookup of the name timed out, the host name was left in
3839 $sender_host_name, contrary to the specification.
3841 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3842 restricted to single-key lookups, Exim was not diagnosing an error if
3843 * or *@ was used with a query-style lookup.
3845 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3847 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3848 long ago, but noone interested enough thought of it. Let's just say that
3849 the "1.1" means that there are some new functions that weren't there at
3850 some point in the past.
3852 PH/51 Error processing for expansion failure of helo_data from an smtp
3853 transport during callout processing was broken.
3855 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3856 tested/used via the -bh/-bhc/-bs options.
3858 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3859 bug, fixed in subsequent PCRE releases).
3861 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3862 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3864 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3870 SC/01 Use a glob alias rather than an array ref in eximstats generated
3871 parser. This improves both readability and performance.
3873 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3874 Don't display local sender or destination tables in eximstats unless
3875 there is data to show.
3876 Added average volumes into the eximstats top table text output.
3878 SC/03 Collect data on the number of addresses (recipients) as well
3879 as the number of messages in eximstats.
3881 TF/01 Correct an error in the documentation for the redirect router. Exim
3882 does (usually) call initgroups() when daemonizing.
3884 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3885 with consistent privilege compared to when running as a daemon.
3887 TF/03 Note in the spec that $authenticated_id is not set for local
3888 submissions from trusted users.
3890 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3891 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3893 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3894 by adding some example configuration directives to the default
3895 configuration file. A little bit of work is required to uncomment the
3896 directives and define how usernames and passwords are checked, but
3897 there is now a framework to start from.
3899 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3900 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3901 without this. I don't know how relevant this is to other LDAP libraries.
3903 PH/02 Add the verb name to the "unknown ACL verb" error.
3905 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3907 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3909 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3910 directory not expanded when it should be if an expanded home directory
3911 was set for the address (which is overridden by the transport).
3913 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3916 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3917 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3918 because it is too late at that time, and has no effect.
3920 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3921 security issue with \' (bugzilla #107). I could not use the
3922 PQescapeStringConn() function, because it needs a PGconn value as one of
3925 PH/08 When testing addresses using -bt, indicate those final addresses that
3926 are duplicates that would not cause an additional delivery. At least one
3927 person was confused, thinking that -bt output corresponded to deliveries.
3928 (Suppressing duplicates isn't a good idea as you lose the information
3929 about possibly different redirections that led to the duplicates.)
3931 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3932 systems where poll() doesn't work, in particular OS X.
3934 PH/10 Added more information to debugging output for retry time not reached.
3936 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3937 operations in malware.c.
3939 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3942 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3943 syslog_duplication set false, log lines that would normally be written
3944 both the the main log and to the reject log were not written to syslog at
3947 PH/14 In the default configuration, change the use of "message" in ACL warn
3948 statements to "add_header".
3950 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3951 not followed by a command (e.g. "seen endif").
3953 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3954 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3957 PH/17 Added extra conditions to the default value of delay_warning_condition
3961 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3962 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3963 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3966 The Auto-Submitted: and various List- headers are standardised, whereas I
3967 don't think Precedence: ever was.
3969 PH/18 Refactored debugging code in route_finduser() to show more information,
3970 in particular, the error code if getpwnam() issues one.
3972 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3973 This is apparently needed in addition to the PH/07 change above to avoid
3974 any possible encoding problems.
3976 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3977 but not after initializing Perl.
3979 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3980 output them only if debugging. By default they are written stderr,
3981 apparently, which is not desirable.
3983 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3986 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3989 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3991 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3992 authenticated or an ident call has been made. Suppress the default
3993 values for $authenticated_id and $authenticated_sender (but permit -oMai
3994 and -oMas) when testing with -bh.
3996 PH/24 Re-jigged the order of the tests in the default configuration so that the
3997 tests for valid domains and recipients precede the DNS black list and CSA
3998 tests, on the grounds that those ones are more expensive.
4000 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
4001 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
4002 This bug exists in every version of Exim that I still have, right back to
4005 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
4006 However, an attempt to turn on case-sensitivity in a regex key by
4007 including (?-i) didn't work because the subject string was already
4008 lowercased, and the effects were non-intuitive. It turns out that a
4009 one-line patch can be used to allow (?-i) to work as expected.
4015 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
4016 other effects) broke the use of negated acl sub-conditions.
4018 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
4021 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
4022 "Deny" causes Exim to reject the incoming connection with a 554 error.
4023 Unfortunately, if there is a major crisis, such as a disk failure,
4024 tcp-wrappers gives "deny", whereas what one would like would be some
4025 kind of temporary error. A kludge has been added to help with this.
4026 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
4027 554 error is used if errno is still zero or contains ENOENT (which occurs
4028 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
4031 PH/03 Add -lutil to the default FreeBSD LIBS setting.
4033 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
4034 errors. Otherwise a message that provokes a temporary error (when other
4035 messages do not) can cause a whole host to time out.
4037 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
4038 addresses were routed directly to files or pipes from a redirect router.
4039 File deliveries just didn't batch; pipe deliveries might have suffered
4042 PH/06 A failure to get a lock for a hints database would erroneously always say
4043 "Failed to get write lock", even when it was really a read lock.
4045 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
4046 of 0600. This has been changed to use the value of the lockfile_mode
4047 option (which defaults to 0600).
4049 PH/08 Applied small patch from the Sieve maintainer.
4051 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
4052 folder from quota calculations, a direct delivery into this folder messed
4053 up the contents of the maildirsize file. This was because the regex was
4054 used only to exclude .Trash (or whatever) when the size of the mailbox
4055 was calculated. There was no check that a delivery was happening into an
4056 excluded directory. This bug has been fixed by ignoring all quota
4057 processing for deliveries into excluded directories.
4059 PH/10 Added the maildirfolder_create_regex option to appendfile.
4065 PH/01 The code for finding all the local interface addresses on a FreeBSD
4066 system running IPv6 was broken. This may well have applied to all BSD
4067 systems, as well as to others that have similar system calls. The broken
4068 code found IPv4 interfaces correctly, but gave incorrect values for the
4069 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
4070 that it would not match correctly against @[] and not recognize the IPv6
4073 PH/02 The ipliteral router was not recognizing addresses of the form user@
4074 [ipv6:....] because it didn't know about the "ipv6:" prefix.
4076 PH/03 Added disable_ipv6.
4078 PH/04 Changed $reply_address to use the raw form of the headers instead of the
4079 decoded form, because it is most often used to construct To: headers
4080 lines in autoreplies, and the decoded form may well be syntactically
4081 invalid. However, $reply_address has leading white space removed, and all
4082 newlines turned into spaces so that the autoreply transport does not
4085 PH/05 If group was specified without a user on a router, and no group or user
4086 was specified on a transport, the group from the router was ignored.
4088 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
4089 for visible compile-time settings that can be used to change these
4090 numbers, for those that want even more. Backwards compatibility with old
4091 spool files has been maintained. However, going back to a previous Exim
4092 release will lost any variables that are in spool files.
4094 PH/07 Two small changes when running in the test harness: increase delay when
4095 passing a TCP/IP connection to a new process, in case the original
4096 process has to generate a bounce, and remove special handling of
4097 127.0.0.2 (sic), which is no longer necessary.
4099 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
4100 be the same on different OS.
4102 PH/09 Moved a debug statement in filter processing to avoid a race problem when
4105 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
4106 whether --show-vars was specified or not
4108 JJ/02 exipick: Added support for new ACL variable spool format introduced
4111 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
4112 syntactically invalid From: or Reply-to: line, and a filter used this to
4113 generate an autoreply, and therefore failed to obtain an address for the
4114 autoreply, Exim could try to deliver to a non-existent relative file
4115 name, causing unrelated and misleading errors. What now happens is that
4116 it logs this as a hard delivery error, but does not attempt to create a
4119 PH/11 The exinext utility has a -C option for testing purposes, but although
4120 the given file was scanned by exinext itself; it wasn't being passed on
4121 when Exim was called.
4123 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
4124 an end-of-file indication when reading a command response.
4126 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
4127 compiled. In many other places in Exim, IPv6 addresses are always
4128 recognized, so I have changed this. It also means that IPv4 domain
4129 literals of the form [IPV4:n.n.n.n] are now always recognized.
4131 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
4132 used if the router is not running as root, for example, when verifying at
4133 ACL time, or when using -bh. The debugging output from this situation was
4134 non-existent - all you got was a failure to exec. I have made two
4137 (a) Failures to set uid/gid, the current directory, or a process leader
4138 in a subprocess such as that created by queryprogram now generate
4139 suitable debugging output when -d is set.
4141 (b) The queryprogram router detects when it is not running as root,
4142 outputs suitable debugging information if -d is set, and then runs
4143 the subprocess without attempting to change uid/gid.
4145 PH/15 Minor change to Makefile for building test_host (undocumented testing
4148 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
4149 additional section of a DNS packet that returns MX or SRV records.
4150 Instead, it always explicitly searches for A/AAAA records. This avoids
4151 major problems that occur when a DNS server includes only records of one
4152 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
4153 fixed another bug: if SRV records were looked up and the corresponding
4154 address records were *not* found in the additional section, the port
4155 values from the SRV records were lost.
4157 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
4158 using the correct key (the original address) when searching the retry
4159 rules in order to find which one to use for generating the retry hint.
4161 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
4162 adding the default one. Similarly, if it contains a Reply-To: header, the
4163 errors_reply_to option, if set, is not used.
4165 PH/19 When calculating a retry time, Exim used to measure the "time since
4166 failure" by looking at the "first failed" field in the retry record. Now
4167 it does not use this if it is later than than the arrival time of the
4168 message. Instead it uses the arrival time. This makes for better
4169 behaviour in cases where some deliveries succeed, thus re-setting the
4170 "first failed" field. An example is a quota failure for a huge message
4171 when small messages continue to be delivered. Without this change, the
4172 "time since failure" will always be short, possible causing more frequent
4173 delivery attempts for the huge message than are intended.
4174 [Note: This change was subsequently modified - see PH/04 for 4.62.]
4176 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
4177 $1, $2, $3) because the numerical variables can be reset during some
4178 expansion items (e.g. "match"), thereby losing the authentication data.
4180 PH/21 Make -bV show the size of off_t variables so that the test suite can
4181 decide whether to run tests for quotas > 2G.
4183 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
4184 mailbox_size, and mailbox_filecount in the appendfile transport. If a
4185 filecount value is greater than 2G or if a quota value is greater than 2G
4186 on a system where the size of off_t is not greater than 4, a panic error
4189 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
4190 never match. The debug and -bh output now contains an explicit error
4191 message indicating a malformed IPv4 address or mask.
4193 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
4194 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
4195 PH/23 above applies.
4197 PH/25 Do not write to syslog when running in the test harness. The only
4198 occasion when this arises is a failure to open the main or panic logs
4199 (for which there is an explicit test).
4201 PH/26 Added the /no_tell option to "control=freeze".
4203 PH/27 If a host name lookup failed very early in a connection, for example, if
4204 the IP address matched host_lookup and the reverse lookup yielded a name
4205 that did not have a forward lookup, an error message of the form "no IP
4206 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
4207 could be logged. Now it outputs the IP address instead of "NULL".
4209 PH/28 An enabling patch from MH: add new function child_open_exim2() which
4210 allows the sender and the authenticated sender to be set when
4211 submitting a message from within Exim. Since child_open_exim() is
4212 documented for local_scan(), the new function should be too.
4214 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
4215 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
4216 results in an empty string is now treated as unset.
4218 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
4220 PH/31 Added sender_verify_fail logging option.
4222 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
4223 needed by RFC 822 but not by RFC 2822 was commented out. I have now
4224 tidied the source and removed it altogether.
4226 PH/33 When a queue run was abandoned because the load average was too high, a
4227 log line was always written; now it is written only if the queue_run log
4228 selector is set. In addition, the log line for abandonment now contains
4229 information about the queue run such as the pid. This is always present
4230 in "start" and "stop" lines but was omitted from the "abandon" line.
4232 PH/34 Omit spaces between a header name and the colon in the error message that
4233 is given when verify = headers_syntax fails (if there are lots of them,
4234 the message gets confusing).
4236 PH/35 Change the default for dns_check_names_pattern to allow slashes within
4237 names, as there are now some PTR records that contain slashes. This check
4238 is only to protect against broken name servers that fall over on strange
4239 characters, so the fact that it applies to all lookups doesn't matter.
4241 PH/36 Now that the new test suite is complete, we can remove some of the
4242 special code in Exim that was needed for the old test suite. For example,
4243 sorting DNS records because real resolvers return them in an arbitrary
4244 order. The new test suite's fake resolver always returns records in the
4247 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
4248 bounces) except when queue_only is set, to avoid logging races between
4249 the different processes.
4251 PH/38 Panic-die if .include specifies a non-absolute path.
4253 PH/39 A tweak to the "H" retry rule from its user.
4255 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
4256 a label. They prevented compilation on older perls.
4258 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
4259 a warning to be raised on newish perls.
4261 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
4262 on queue. Changes to match documented behaviour of showing count of
4263 messages matching specified criteria.
4265 PH/40 Changed the default ident timeout from 30s to 5s.
4267 PH/41 Added support for the use of login_cap features, on those BSD systems
4268 that have them, for controlling the resources used by pipe deliveries.
4270 PH/42 The content-scanning code uses fopen() to create files in which to put
4271 message data. Previously it was not paying any attention to the mode of
4272 the files. Exim runs with umask(0) because the rest of the code creates
4273 files with open(), and sets the required mode explicitly. Thus, these
4274 files were ending up world-writeable. This was not a big issue, because,
4275 being within the spool directory, they were not world-accessible. I have
4276 created a function called modefopen, which takes an additional mode
4277 argument. It sets umask(777), creates the file, chmods it to the required
4278 mode, then resets the umask. All the relevant calls to fopen() in the
4279 content scanning code have been changed to use this function.
4281 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
4282 to 24 hours. This avoids potential overflow problems when processing G
4283 and H retry rules. I suspect nobody ever tinkers with this value.
4285 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
4287 PH/45 When the plaintext authenticator is running as a client, the server's
4288 challenges are checked to ensure they are valid base64 strings. By
4289 default, the authentication attempt is cancelled if an invalid string is
4290 received. Setting client_ignore_invalid_base64 true ignores these errors.
4291 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
4292 they are received. Thus, the responses can be made to depend on the
4293 challenges. If an invalid string is ignored, an empty string is placed in
4296 PH/46 Messages that are created by the autoreply transport now contains a
4297 References: header, in accordance with RFCs 2822 and 3834.
4299 PH/47 Added authenticated_sender_force to the smtp transport.
4301 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
4303 PH/49 Installed latest patch from the Sieve maintainer.
4305 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
4306 was also set, the appendfile transport was unnecessarily scanning a
4307 directory of message files (e.g. for maildir delivery) to find the count
4308 of files (along with the size), even though it did not need this
4309 information. It now does the scan only if it needs to find either the
4310 size of the count of files.
4312 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
4314 PH/52 Two bugs concerned with error handling when the smtp transport is
4317 (i) Exim was not creating retry information for temporary errors given
4318 for individual recipients after the DATA command when the smtp transport
4319 was used in LMTP mode. This meant that they could be retried too
4320 frequently, and not timed out correctly.
4322 (ii) Exim was setting the flag that allows error details to be returned
4323 for LMTP errors on RCPT commands, but not for LMTP errors for individual
4324 recipients that were returned after the DATA command.
4326 PH/53 This is related to PH/52, but is more general: for any failing address,
4327 when detailed error information was permitted to be returned to the
4328 sender, but the error was temporary, then after the final timeout, only
4329 "retry timeout exceeded" was returned. Now it returns the full error as
4330 well as "retry timeout exceeded".
4332 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
4333 do this, and (what is worse) MTAs that accept it.
4335 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
4336 will now be deprecated.
4338 PH/56 New os.c-cygwin from the Cygwin maintainer.
4340 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
4341 formats (previously only available in exim formats via -bpr, -bpru,
4342 and -bpra. Now also available in native and exiqgrep formats)
4344 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
4345 with very large, slow to parse queues
4347 JJ/08 exipick: added ! as generic prefix to negate any criteria format
4349 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
4351 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
4352 responses to authentication challenges, though it was showing the
4353 challenges; (ii) I've removed the CR characters from the debug output for
4356 PH/58 Allow for the insertion of a newline as well as a space when a string
4357 is turned into more than one encoded-word during RFC 2047 encoding. The
4358 Sieve code now uses this.
4360 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
4361 data_4xx, lost_connection, tls_required.
4363 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
4364 message was being sent as an SMTP response.
4366 PH/61 Add -l and -k options to exicyclog.
4368 PH/62 When verifying, if an address was redirected to one new address, so that
4369 verification continued, and the new address failed or deferred after
4370 having set something in $address_data, the value of $address_data was not
4371 passed back to the ACL. This was different to the case when no
4372 redirection occurred. The value is now passed back in both cases.
4374 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
4375 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
4376 use login_cap.h, so on its own it isn't the distinguishing feature. The
4377 new name refers directly to the setclassresources() function.
4379 PH/65 Added configuration files for NetBSD3.
4381 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
4383 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
4384 is preferred over IPv4.
4386 PH/68 The bounce_return_message and bounce_return_body options were not being
4387 honoured for bounces generated during the reception of non-SMTP messages.
4388 In particular, this applied to messages rejected by the ACL. This bug has
4389 been fixed. However, if bounce_return_message is true and bounce_return_
4390 body is false, the headers that are returned for a non-SMTP message
4391 include only those that have been read before the error was detected.
4392 (In the case of an ACL rejection, they have all been read.)
4394 PH/69 The HTML version of the specification is now built in a directory called
4395 spec_html instead of spec.html, because the latter looks like a path with
4396 a MIME-type, and this confuses some software.
4398 PH/70 Catch two compiler warnings in sieve.c.
4400 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
4401 function verify_get_ident() calls ip_connect() to connect a socket, but
4402 if the "connect()" function timed out, ip_connect() used to close the
4403 socket. However, verify_get_ident() also closes the socket later, and in
4404 between Exim writes to the log, which may get opened at this point. When
4405 the socket was closed in ip_connect(), the log could get the same file
4406 descriptor number as the socket. This naturally causes chaos. The fix is
4407 not to close the socket in ip_connect(); the socket should be closed by
4408 the function that creates it. There was only one place in the code where
4409 this was missing, in the iplookup router, which I don't think anybody now
4410 uses, but I've fixed it anyway.
4412 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
4413 well as to direct DNS lookups. Otherwise the handling of names in host
4414 lists is inconsistent and therefore confusing.
4420 PH/01 Two changes to the default runtime configuration:
4422 (1) Move the checks for relay_from_hosts and authenticated clients from
4423 after to before the (commented out) DNS black list checks.
4425 (2) Add control=submission to the relay_from_hosts and authenticated
4426 clients checks, on the grounds that messages accepted by these
4427 statements are most likely to be submissions.
4429 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
4431 (1) Generate an error if the third argument for the ${prvs expansion is
4434 (2) Treat a missing third argument of ${prvscheck as if it were an empty
4437 (3) Reset the variables that are obtained from the first argument of
4438 ${prvscheck and used in the second argument before leaving the code,
4439 because their memory is reclaimed, so using them afterwards may do
4442 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
4443 one (it's much easier than Tom thought :-).
4445 (5) Because of (4), we can now allow for the use of $prvscheck_result
4446 inside the third argument.
4448 PH/03 For some reason, the default setting of PATH when running a command from
4449 a pipe transport was just "/usr/bin". I have changed it to
4452 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
4453 anything to be listed in the output from -bV.
4455 PH/05 When a filter generated an autoreply, the entire To: header line was
4456 quoted in the delivery log line, like this:
4458 => >A.N.Other <ano@some.domain> <original@ddress> ...
4460 This has been changed so that it extracts the operative address. There
4461 may be more than one such address. If so, they are comma-separated, like
4464 => >ano@some.domain,ona@other.domain <original@ddress> ...
4466 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
4467 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
4468 not being looked up in the rDNS to get a host name, Exim was showing the
4469 IP address twice in Received: lines, even though the IP addresses were
4470 identical. For example:
4472 Received: from [1.2.3.4] (helo=[1.2.3.4])
4474 However, if the real host name was known, it was omitting the HELO data
4475 if it matched the actual IP address. This has been tidied up so that it
4476 doesn't show the same IP address twice.
4478 PH/07 When both +timestamp and +memory debugging was on, the value given by
4479 $tod_xxx expansions could be wrong, because the tod_stamp() function was
4480 called by the debug printing, thereby overwriting the timestamp buffer.
4481 Debugging no longer uses the tod_stamp() function when +timestamp is set.
4483 PH/08 When the original message was included in an autoreply transport, it
4484 always said "this is a copy of the message, including all the headers",
4485 even if body_only or headers_only was set. It now gives an appropriate
4488 PH/09 Applied a patch from the Sieve maintainer which:
4490 o fixes some comments
4491 o adds the (disabled) notify extension core
4492 o adds some debug output for the result of if/elsif tests
4493 o points to the current vacation draft in the documentation
4494 and documents the missing references header update
4498 o fixes a bug in processing the envelope test (when testing
4499 multiple envelope elements, the last element determined the
4502 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
4503 Electronic Mail") by including:
4505 Auto-submitted: auto-generated
4507 in the messages that it generates (bounce messages and others, such as
4508 warnings). In the case of bounce messages for non-SMTP messages, there was
4509 also a typo: it was using "Auto_submitted" (underscore instead of
4510 hyphen). Since every message generated by Exim is necessarily in response
4511 to another message, thes have all been changed to:
4513 Auto-Submitted: auto-replied
4515 in accordance with these statements in the RFC:
4517 The auto-replied keyword:
4519 - SHOULD be used on messages sent in direct response to another
4520 message by an automatic process,
4522 - MUST NOT be used on manually-generated messages,
4524 - MAY be used on Delivery Status Notifications (DSNs) and Message
4525 Disposition Notifications (MDNs),
4527 - MUST NOT be used on messages generated by automatic or periodic
4528 processes, except for messages which are automatic responses to
4531 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
4532 to the default Received: header definition.
4534 PH/12 Added log selector acl_warn_skipped (default on).
4536 PH/13 After a successful wildlsearch lookup, discard the values of numeric
4537 variables because (a) they are in the wrong storage pool and (b) even if
4538 they were copied, it wouldn't work properly because of the caching.
4540 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
4541 checking when decoding. Apparently there are clients that generate
4542 overlong encoded strings. Why am I not surprised?
4544 PH/15 If the first argument of "${if match_address" was not empty, but did not
4545 contain an "@" character, Exim crashed. Now it writes a panic log message
4546 and treats the condition as false.
4548 PH/16 In autoreply, treat an empty string for "once" the same as unset.
4550 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
4551 extension "envelope-auth". The code is finished and in agreement with
4552 other implementations, but there is no documentation so far and in fact,
4553 nobody wrote the draft yet. This extension is currently #undef'ed, thus
4554 not changing the active code.
4556 Print executed "if" and "elsif" statements when debugging is used. This
4557 helps a great deal to understand what a filter does.
4559 Document more things not specified clearly in RFC3028. I had all this
4560 sorted out, when out of a sudden new issues came to my mind. Oops."
4562 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
4565 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
4566 canonical form (as documented). However, after a host name lookup from
4567 the IP address, check_host() was doing a simple string comparison with
4568 addresses acquired from the DNS when checking that the found name did
4569 have the original IP as one of its addresses. Since any found IPv6
4570 addresses are likely to be in abbreviated form, the comparison could
4571 fail. Luckily, there already exists a function for doing the comparison
4572 by converting both addresses to binary, so now that is used instead of
4573 the text comparison.
4575 PH/20 There was another similar case to PH/19, when a complete host name was
4576 given in a host list; looking up its IP address could give an abbreviated
4577 form, whereas the current host's name might or might not be abbreviated.
4578 The same fix has been applied.
4584 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
4585 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
4588 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
4589 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
4591 PH/03 Typo: missing ".o" in src/pcre/Makefile.
4593 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
4594 header line, restrict the check to what is listed in RFCs 2369 and 2929.
4595 Also, for "Auto-Submitted", treat anything other than "no" as
4596 non-personal, in accordance with RFC 3834. (Previously it treated
4597 anything starting "auto-" as non-personal.)
4599 TF/01 The control=submission/name=... option had a problem with syntax
4600 errors if the name included a slash character. The /name= option
4601 now slurps the rest of the string, so it can include any characters
4602 but it must come last in the list of options (after /sender_retain
4605 PH/05 Some modifications to the interface to the fake nameserver for the new
4613 TK/01 Added the "success_on_redirect" address verification option. See
4614 NewStuff for rationale and an example.
4616 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
4618 PH/02 Patch to exigrep to allow it to work on syslog lines.
4620 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
4621 fread() to skip over the body file's header line, because in Cygwin the
4622 header line is locked and is inaccessible.
4624 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
4625 co-exist for some time) to make it clear that it is the Exim ID that is
4626 referenced, not the Message-ID: header line.
4628 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
4629 string_format() function, because snprintf() does not exist on all
4632 PH/06 The use of forbid_filter_existstest now also locks out the use of the
4633 ${stat: expansion item.
4635 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
4636 protocol synchronization error", to keep the pedants happy.
4638 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
4639 well as for IRIX systems, when gcc is being used. See the host.c source
4642 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
4644 PH/10 Named domain lists were not working if used in a queue_smtp_domains
4647 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
4648 transport and to the smtp transport in LMTP mode.
4650 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
4652 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
4653 run a filter in a subprocess. This could lead to confusion in subsequent
4654 lookups in the parent process. There should also be a search_tidyup() at
4655 the end of the subprocess.
4657 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
4658 only if the host matched helo_try_verify_hosts, which caused the
4659 verification to occur when the EHLO/HELO command was issued. The ACL just
4660 tested the remembered result. Now, if a previous verification attempt has
4661 not happened, "verify = helo" does it there and then.
4663 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
4665 TK/03 Fix log output including CR from clamd.
4667 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
4668 exist provoked a memory error which could cause a segfault.
4670 PH/15 Installed PCRE 6.2
4672 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
4674 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
4675 of the problem. Specifically, suggested +O2 rather than +O1 for the
4678 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
4680 PH/20 If a delivery was routed to a non-standard port by means of an SRV
4681 record, the port was not correctly logged when the outgoing_port log
4682 selector was set (it logged the transort's default port).
4684 PH/21 Added support for host-specific ports to manualroute, queryprogram,
4685 fallback_hosts, and "hosts" in the smtp transport.
4687 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
4688 host errors such as "Connection refused".
4690 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
4691 authentication with radiusclient 0.4.9:
4693 - Error returned from rc_read_config was caught wrongly
4694 - Username/password not passed on to radius server due to wrong length.
4696 The presumption is that some radiusclient API changes for 4.51/PH/17
4697 were not taken care of correctly. The code is still untested by me (my
4698 Linux distribution still has 0.3.2 of radiusclient), but it was
4699 contributed by a Radius user.
4701 PH/24 When doing a callout, the value of $domain wasn't set correctly when
4702 expanding the "port" option of the smtp transport.
4704 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
4705 while reading a MIME header. Thanks to Tom Hughes for a patch.
4707 PH/24 Include config.h inside local_scan.h so that configuration settings are
4710 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
4711 that in an ACL for RCPT (for example), you can examine exactly what was
4714 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
4715 commands, but it was not correctly comparing the address with the actual
4716 client host address. Thus, it would show the EHLO address in Received:
4717 header lines when this was not necessary.
4719 PH/27 Added the % operator to ${eval:}.
4721 PH/28 Exim tries to create and chdir to its spool directory when it starts;
4722 it should be ignoring failures (because with -C, for example, it has lost
4723 privilege). It wasn't ignoring creation failures other than "already
4726 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
4729 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
4730 because an input line was too long, either on its own, or by virtue of
4731 too many continuations, the temporary file was not being removed, and the
4732 return code was incorrect.
4734 PH/31 Missing "BOOL" in function definition in filtertest.c.
4736 PH/32 Applied Sieve patches from the maintainer.
4738 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
4740 PH/33 Added "verify = not_blind".
4742 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
4743 Local/Makefile (with some defaults set). These are used in built scripts
4744 such as exicyclog, but they have never been used in the exim_install
4745 script (though there are many overriding facilities there). I have
4746 arranged that the exim_install script now takes note of these two
4749 PH/35 Installed configuration files for Dragonfly.
4751 PH/36 When a locally submitted message by a trusted user did not contain a
4752 From: header, and the sender address was obtained from -f or from an SMTP
4753 MAIL command, and the trusted user did not use -F to supply a sender
4754 name, $originator_name was incorrectly used when constructing a From:
4755 header. Furthermore, $originator_name was used for submission mode
4756 messages from external hosts without From: headers in a similar way,
4757 which is clearly wrong.
4759 PH/37 Added control=suppress_local_fixups.
4761 PH/38 When log_selector = +received_sender was set, and the addition of the
4762 sender made the log line's construction buffer exactly full, or one byte
4763 less than full, an overflow happened when the terminating "\n" was
4766 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
4767 when the result of a list match is failure because a DNS lookup failed.
4769 PH/40 RM_COMMAND is now used in the building process.
4771 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
4772 the "build-* directories that it finds.
4774 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
4775 address in a domain literal was a prefix of an interface address.
4777 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
4778 when verifying a sender address, unless rewrite_headers is false.
4780 PH/44 Wrote a long comment about why errors_to addresses are verified as
4781 recipients, not senders.
4783 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
4784 the ratelimit ACL was added.
4786 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
4788 PH/46 Added extra information about PostgreSQL errors to the error string.
4790 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
4791 avoiding the need to install special zones in a real server. This is
4792 backwards compatible; if it can't find the fake resolver, it drops back.
4793 Thus, both old and new test suites can be run.
4795 TF/02 Added util/ratelimit.pl
4797 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
4798 clock is set back in time.
4800 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
4801 Candler <B.Candler@pobox.com>.
4803 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
4804 OK for addresses that are the result of redirections.
4806 PH/48 A number of further additions for the benefit of the new test suite,
4807 including a fake gethostbyname() that interfaces to the fake DNS resolver
4810 TF/06 The fix for widen_domains has also been applied to qualify_single and
4811 search_parents which are the other dnslookup options that can cause
4814 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
4817 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
4819 TF/07 Exim produced the error message "an SRV record indicated no SMTP
4820 service" if it encountered an MX record with an empty target hostname.
4821 The message is now "an MX or SRV record indicated no SMTP service".
4823 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
4824 if the DNS of the sending site is misconfigured. This is quite a
4825 common situation. This change restores the behaviour of treating a
4826 helo verification defer as a failure.
4828 PH/51 If self=fail was set on a router, the bounce message did not include the
4829 actual error message.
4835 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
4837 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
4838 command itself ended in error, the underlying message about the transport
4839 filter timeout was being overwritten with the pipe command error. Now the
4840 underlying error message should be appended to the second error message.
4842 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4844 PH/02 Reduce the amount of output that "make" produces by default. Full output
4845 can still be requested.
4847 PH/03 The warning log line about a condition test deferring for a "warn" verb
4848 was being output only once per connection, rather than after each
4849 occurrence (because it was using the same function as for successful
4850 "warn" verbs). This seems wrong, so I have changed it.
4852 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4853 it should not have, which might have caused a crash in the right
4854 circumstances, but probably never did.
4856 PH/04 Installed a modified version of Tony Finch's patch to make submission
4857 mode fix the return path as well as the Sender: header line, and to
4858 add a /name= option so that you can make the user's friendly name appear
4861 TF/03 Added the control = fakedefer ACL modifier.
4863 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4864 Mark Lowes for thorough testing.
4866 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4868 TK/03 Merged latest SRS patch from Miles Wilton.
4870 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4871 with the definition in sysexits.h (which is #included earlier).
4872 Fortunately, Exim does not actually use EX_OK. The code used to try to
4873 preserve the sysexits.h value, by assuming that macro definitions were
4874 scanned for macro replacements. I have been disabused of this notion,
4875 so now the code just undefines EX_OK before #including unistd.h.
4877 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4878 in the smtp transport. When a block could not be written in a single
4879 write() function, the timeout was being re-applied to each part-write.
4880 This seems wrong - if the receiver was accepting one byte at a time it
4881 would take for ever. The timeout is now adjusted when this happens. It
4882 doesn't have to be particularly precise.
4884 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4885 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4887 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4888 without a domain if the check to <postmaster@domain> fails.
4890 SC/01 Eximstats: added -xls and the ability to specify output files
4891 (patch written by Frank Heydlauf).
4893 SC/02 Eximstats: use FileHandles for outputting results.
4895 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4897 SC/04 Eximstats: fixed display of large numbers with -nvr option
4899 SC/05 Eximstats: fixed merging of reports with empty tables.
4901 SC/06 Eximstats: added the -include_original_destination flag
4903 SC/07 Eximstats: removed tabs and trailing whitespace.
4905 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4907 TK/06 MBOX spool code: Add real "From " MBOX separator line
4908 so the .eml file is really in mbox format (even though
4909 most programs do not really care). Patch from Alex Miller.
4911 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4912 The latter is generated from $received_to and is only set if the
4913 message has one envelope recipient. SA can use these headers,
4914 obviously out-of-the-box. Patch from Alex Miller.
4916 PH/08 The ${def test on a variable was returning false if the variable's
4917 value was "0", contrary to what the specification has always said!
4918 The result should be true unless the variable is empty.
4920 PH/09 The syntax error of a character other than { following "${if
4921 def:variable_name" (after optional whitespace) was not being diagnosed.
4922 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4923 accidental colon was present, for example, could give incorrect results.
4925 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4926 result is used (not including appendfile, where other changes are about
4929 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4930 This involved changing a lot of size variables from int to off_t. It
4931 should work with maildirs and everything.
4933 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4934 spamd dying while we are connected to it.
4936 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4939 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4940 function for BDB 4.3.
4942 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4944 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4945 the sources for PCRE have been re-arranged and more files are now
4948 PH/15 The code I had for printing potentially long long variables in PH/11
4949 above was not the best (it lost precision). The length of off_t variables
4950 is now inspected at build time, and an appropriate printing format (%ld
4951 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4952 to be "long long int" or "long int". This is needed for the internal
4953 formatting function string_vformat().
4955 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4956 the configuration file to be ":syslog", then the script "guesses" where
4957 the logs files are, rather than using the compiled in default. In our
4958 case the guess is not the same as the compiled default, so the script
4959 suddenly stopped working when I started to use syslog. The patch checks
4960 to see if log_file_path is "". If so, it attempts to read it from exim
4961 with no configuration file to get the compiled in version, before it
4962 falls back to the previous guessing code."
4964 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4965 implementing BATV in an Exim configuration. See NewStuff for the gory
4968 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4969 Makefile that are specific to HP-UX.
4971 PH/18 If the "use_postmaster" option was set for a recipient callout together
4972 with the "random" option, the postmaster address was used as the MAIL
4973 FROM address for the random test, but not for the subsequent recipient
4974 test. It is now used for both.
4976 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4977 patch removes a few documentation additions to RFC 3028, because the
4978 latest draft now contains them. It adds the new en;ascii-case comparator
4979 and a new error check for 8bit text in MIME parts. Comparator and
4980 require names are now matched exactly. I enabled the subaddress
4981 extension, but it is not well tested yet (read: it works for me)."
4983 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4984 rework some of the code of TK/09 above to avoid the hardwired use of
4985 "%lld" and "long long". Replaced the call to snprintf() with a call to
4988 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4989 records point to non-existent hosts", "retry timeout exceeded", and
4990 "retry time not reached for any host after a long failure period".
4992 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4993 experimental DomainKeys support:
4995 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4996 (2) On an error such as an illegally used "control", the wrong name for
4997 the control was given.
4999 These problems did NOT occur unless DomainKeys support was compiled.
5001 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
5003 PH/24 Added ${if match_ip condition.
5005 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
5006 databases so that it will be absolutely obvious if a crash occurs in the
5007 DB library. This is a regular occurrence (often caused by mis-matched
5010 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
5011 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
5012 picked up on a user's system that detects such things. There doesn't seem
5013 to be a gcc warning option for this - only an attribute that has to be
5014 put on the function's prototype. It seems that in Fedora Core 4 they have
5015 set this on a number of new functions. No doubt there will be more in due
5018 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
5019 specify a transport. However, if an address that was verified by such a
5020 router was the subject of a callout, Exim crashed because it tried to
5021 read the rcpt_include_affixes from the non-existent transport. Now it
5022 just assumes that the setting of that option is false. This bug was
5023 introduced by 4.51/PH/31.
5025 PH/28 Changed -d+all to exclude +memory, because that information is very
5026 rarely of interest, but it makes the output a lot bigger. People tend to
5027 do -d+all out of habit.
5029 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
5030 code in os-type was giving problems when libc.so lives in lib64, like on
5033 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
5034 aren't the modern standard, and it seems that some systems' include files
5035 don't always have them. Exim was already checking for some of the newer
5036 ones like T_AAAA, and defining it itself. I've added checks for all the
5037 record types that Exim uses.
5039 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
5040 not automatically generating a new one, as it is supposed to. This
5041 prevented TLS from working. If the file did exist, but contained invalid
5042 data, a new version was generated, as expected. It was only the case of a
5043 non-existent file that was broken.
5045 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
5046 with a change in libdomainkeys > 0.64.
5048 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
5049 from DNS. If the selector record carries the flag, it now has
5050 precedence over the domain-wide flag.
5052 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
5054 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
5055 the use of a port name that isn't defined in /etc/services) occurred, the
5056 message was deferred as in a normal delivery, and thus remained on the
5057 spool, instead of being failed because of the mua_wrapper setting. This
5058 is now fixed, and I tidied up some of the mua_wrapper messages at the
5061 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
5062 the messages in a hash of arrays rather than using individual hashes.
5063 This is a bit cleaner and results in dramatic memory savings, albeit
5064 at a slight CPU cost.
5066 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
5067 as requested by Marc Sherman.
5069 SC/10 Eximstats: added histograms for user specified patterns as requested
5072 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
5074 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
5075 fopen() in the content-scanning modules that did not already have it.
5081 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
5082 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
5084 TK/02 Fix ACL "control" statement not being available in MIME ACL.
5086 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
5088 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
5089 to test Sieve filters that use "vacation".
5091 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
5092 that changes the way the GnuTLS parameters are stored in the cache file.
5093 The new format can be generated externally. For backward compatibility,
5094 if the data in the cache doesn't make sense, Exim assumes it has read an
5095 old-format file, and it generates new data and writes a new file. This
5096 means that you can't go back to an older release without removing the
5099 PH/03 A redirect router that has both "unseen" and "one_time" set does not
5100 work if there are any delivery delays because "one_time" forces the
5101 parent to be marked "delivered", so its unseen clone is never tried
5102 again. For this reason, Exim now forbids the simultaneous setting of
5105 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
5106 redirected to themselves ("homonym" addresses). Read the long ChangeLog
5107 entry if you want to know the details. The fix, however, neglected to
5108 consider the case when local delivery batching is involved. The test for
5109 "previously delivered" was not happening when checking to see if an
5110 address could be batched with a previous (undelivered) one; under
5111 certain circumstances this could lead to multiple deliveries to the same
5114 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
5115 in its include files, and this causes problems building Exim.
5117 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
5118 header_syntax) but Exim was just ignoring anything given after a slash.
5119 In particular, this caused confusion with an attempt to use "verify =
5120 reverse_host_lookup/defer_ok". An error is now given when options are
5121 supplied for verify items that do not have them. (Maybe reverse_host_
5122 lookup should have a defer_ok option, but that's a different point.)
5124 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
5125 defined by RFC 821) to 2048, because there were problems with some AUTH
5126 commands, and RFC 1869 says the size should be increased for extended
5127 SMTP commands that take arguments.
5129 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
5132 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
5133 "unknown" error; now it says that the functionality isn't in the binary.
5135 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
5136 an address' error message when a string expansion fails (syntax or
5137 whatever). Otherwise the password may appear in the log. Following change
5138 PH/42 below, there is no longer a chance of it appearing in a bounce
5141 PH/11 Installed exipick version 20050225.0 from John Jetmore.
5143 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
5144 of its addresses was ever tried. (Bugzilla bug #2.)
5146 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
5147 the result incorrectly in the debug output. (It correctly added a newline
5148 to what was transported.)
5150 TF/01 Added $received_time.
5152 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
5153 commented out examples of how to interface to a virus scanner and to
5154 SpamAssassin. Also added commented examples of av_scanner and
5155 spamd_address settings.
5157 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
5158 and controls are allowed in which ACLs. There were a couple of minor
5159 errors. Some of the entries in the conditions table (which is a table of
5160 where they are NOT allowed) were getting very unwieldy; rewrote them as a
5161 negation of where the condition IS allowed.
5163 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
5165 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
5166 header file does not have a version number, so I've had to invent a new
5167 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
5168 API. The code is untested by me (my Linux distribution still has 0.3.2 of
5169 radiusclient), but it was contributed by a Radius user.
5171 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
5172 files or directories, for OpenSSL.
5174 PH/19 When an Exim process that is running as root has to create an Exim log
5175 file, it does so in a subprocess that runs as exim:exim so as to get the
5176 ownership right at creation (otherwise, other Exim processes might see
5177 the file with the wrong ownership). There was no test for failure of this
5178 fork() call, which would lead to the process getting stuck as it waited
5179 for a non-existent subprocess. Forks do occasionally fail when resources
5180 run out. I reviewed all the other calls to fork(); they all seem to check
5183 PH/20 When checking for unexpected SMTP input at connect time (before writing
5184 the banner), Exim was not dealing correctly with a non-positive return
5185 from the read() function. If the client had disconnected by this time,
5186 the result was a log entry for a synchronization error with an empty
5187 string after "input=" when read() returned zero. If read() returned -1
5188 (an event I could not check), uninitialized data bytes were printed.
5189 There were reports of junk text (parts of files, etc) appearing after
5192 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
5194 PH/22 Added support for macro redefinition, and (re)definition in between
5195 driver and ACL definitions.
5197 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
5198 forgetting to use the resulting value; it was using the unexpanded value.
5200 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
5201 hadn't been configured. The fix is from Juergen Kreileder, who
5202 understands it better than I do:
5204 "Here's what I see happening with three configured cyrus_sasl
5205 authenticators configured (plain, login, cram-md5):
5207 On startup auth_cyrus_sasl_init() gets called for each of these.
5208 This means three calls to sasl_listmech() without a specified mech_list.
5209 => SASL tests which mechs of all available mechs actually work
5210 => three warnings about OTP not working
5211 => the returned list contains: plain, login, cram-md5, digest-md5, ...
5213 With the patch, sasl_listmech() also gets called three times. But now
5214 SASL's mech_list option is set to the server_mech specified in the the
5215 authenticator. Or in other words, the answer from sasl_listmech()
5216 gets limited to just the mech you're testing for (which is different
5218 => the return list contains just 'plain' or 'login', 'cram-md5' or
5219 nothing depending on the value of ob->server_mech.
5221 I've just tested the patch: Authentication still works fine,
5222 unavailable mechs specified in the exim configuration are still
5223 caught, and the auth.log warnings about OTP are gone."
5225 PH/25 When debugging is enabled, the contents of the command line are added
5226 to the debugging output, even when log_selector=+arguments is not
5229 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
5230 answer is "GNU", and only if the return is "GNU/something" is the answer
5233 PH/27 $acl_verify_message is now set immediately after the failure of a
5234 verification in an ACL, and so is available in subsequent modifiers. In
5235 particular, the message can be preserved by coding like this:
5237 warn !verify = sender
5238 set acl_m0 = $acl_verify_message
5240 Previously, $acl_verify_message was set only while expanding "message"
5241 and "log_message" when a very denied access.
5243 PH/28 Modified OS/os.c-Linux with
5245 -#ifndef OS_LOAD_AVERAGE
5246 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
5248 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
5249 nomenclature these days.)
5251 PH/29 Installed patch from the Sieve maintainer that adds the options
5252 sieve_useraddress and sieve_subaddress to the redirect router.
5254 PH/30 In these circumstances:
5255 . Two addresses routed to the same list of hosts;
5256 . First host does not offer TLS;
5257 . First host accepts first address;
5258 . First host gives temporary error to second address;
5259 . Second host offers TLS and a TLS session is established;
5260 . Second host accepts second address.
5261 Exim incorrectly logged both deliveries with the TLS parameters (cipher
5262 and peerdn, if requested) that were in fact used only for the second
5265 PH/31 When doing a callout as part of verifying an address, Exim was not paying
5266 attention to any local part prefix or suffix that was matched by the
5267 router that accepted the address. It now behaves in the same way as it
5268 does for delivery: the affixes are removed from the local part unless
5269 rcpt_include_affixes is set on the transport.
5271 PH/32 Add the sender address, as F=<...>, to the log line when logging a
5272 timeout during the DATA phase of an incoming message.
5274 PH/33 Sieve envelope tests were broken for match types other than :is. I have
5275 applied a patch sanctioned by the Sieve maintainer.
5277 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
5278 the uid or gid is negative. A case of a negative gid caused this to be
5279 noticed. The fix allows for either to be negative.
5281 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
5282 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
5285 PH/36 The change PH/12 above was broken. Fixed it.
5287 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
5288 the grounds that routing the same address twice would always produce the
5289 same answer. This might have been true once, but it is certainly no
5290 longer true now. Routing a child address may depend on the previous
5291 routing that produced that child. Some complicated redirection strategies
5292 went wrong when messages had multiple recipients, and made Exim's
5293 behaviour dependent on the order in which the addresses were given.
5295 I have moved the duplicate checking until after the routing is complete.
5296 Exim scans the addresses that are assigned to local and remote
5297 transports, and removes any duplicates. This means that more work will be
5298 done, as duplicates will always all be routed, but duplicates are
5299 presumably rare, so I don't expect this is of any significance.
5301 For deliveries to pipes, files, and autoreplies, the duplicate checking
5302 still happens during the routing process, since they are not going to be
5305 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
5306 It corrects a timeout issue with spamd. This is Ian's comment: "The
5307 background is that sometimes spamd either never reads data from a
5308 connection it has accepted, or it never writes response data. The exiscan
5309 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
5310 blindly assumes that writes won't block so it may never time out."
5312 PH/39 Allow G after quota size as well as K and M.
5314 PH/40 The value set for $authenticated_id in an authenticator may not contain
5315 binary zeroes or newlines because the value is written to log lines and
5316 to spool files. There was no check on this. Now the value is run through
5317 the string_printing() function so that such characters are converted to
5318 printable escape sequences.
5320 PH/41 $message_linecount is a new variable that contains the total number of
5321 lines in the message. Compare $body_linecount, which is the count for the
5324 PH/42 Exim no longer gives details of delivery errors for specific addresses in
5325 bounce and delay warning messages, except in certain special cases, which
5328 (a) An SMTP error message from a remote host;
5329 (b) A message specified in a :fail: redirection;
5330 (c) A message specified in a "fail" command in a system filter;
5331 (d) A message specified in a FAIL return from the queryprogram router;
5332 (e) A message specified by the cannot_route_message router option.
5334 In these cases only, Exim does include the error details in bounce and
5335 warning messages. There are also a few cases where bland messages such
5336 as "unrouteable address" or "local delivery error" are given.
5338 PH/43 $value is now also set for the "else" part of a ${run expansion.
5340 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
5341 being worked on, but at least Exim now implements the latest version to
5344 PH/45 In a pipe transport, although a timeout while waiting for the pipe
5345 process to complete was treated as a delivery failure, a timeout while
5346 writing the message to the pipe was logged, but erroneously treated as a
5347 successful delivery. Such timeouts include transport filter timeouts. For
5348 consistency with the overall process timeout, these timeouts are now
5349 treated as errors, giving rise to delivery failures by default. However,
5350 there is now a new Boolean option for the pipe transport called
5351 timeout_defer, which, if set TRUE, converts the failures into defers for
5352 both kinds of timeout. A transport filter timeout is now identified in
5355 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
5356 systems where "make" and "gmake" are different, calling "gmake" at top
5357 level broke things. I've arranged for the value of $(MAKE) to be passed
5358 from the Makefile to this script so that it can call the same version of
5362 A note about Exim versions 4.44 and 4.50
5363 ----------------------------------------
5365 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
5366 changes of various kinds. As a consequence, a big documentation update was
5367 needed. This delayed the release for rather longer than seemed good, especially
5368 in the light of a couple of (minor) security issues. Therefore, the changes
5369 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
5370 release. So 4.44 and 4.50 are in effect two different branches that both start
5373 I have left the 4.50 change log unchanged; it contains all the changes since
5374 4.43. The change log for 4.44 is below; many of its items are identical to
5375 those for 4.50. This seems to be the most sensible way to preserve the
5376 historical information.
5382 1. Minor wording change to the doc/README.SIEVE file.
5384 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
5385 computation of the current number of files was incorrect.
5387 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5388 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5391 4. Give more explanation in the error message when the command for a transport
5392 filter fails to execute.
5394 5. There are several places where Exim runs a non-Exim command in a
5395 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5396 was being done only for the command run by the queryprogram router. It is
5397 now done for all such subprocesses. The other cases are: ${run, transport
5398 filters, and the commands run by the lmtp and pipe transports.
5400 6. Added CONFIGURE_GROUP build-time option.
5402 7. Some older OS have a limit of 256 on the maximum number of file
5403 descriptors. Exim was using setrlimit() to set 1000 as a large value
5404 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5405 systems. I've change it so that if it can't get 1000, it tries for 256.
5407 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5408 was an oversight, and furthermore, ever since the addition of extra
5409 controls (e.g. 4.43/32), the checks on when to allow different forms of
5410 "control" were broken. There should now be diagnostics for all cases when a
5411 control that does not make sense is encountered.
5413 9. Added the /retain_sender option to "control=submission".
5415 10. $recipients is now available in the predata ACL (oversight).
5417 11. Tidy the search cache before the fork to do a delivery from a message
5418 received from the command line. Otherwise the child will trigger a lookup
5419 failure and thereby defer the delivery if it tries to use (for example) a
5420 cached ldap connection that the parent has called unbind on.
5422 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5423 of $address_data from the recipient verification was clobbered by the
5424 sender verification.
5426 13. The value of address_data from a sender verification is now available in
5427 $sender_address_data in subsequent conditions in the ACL statement.
5429 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
5431 15. Added a new option "connect=<time>" to callout options, to set a different
5434 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5435 was its contents. (It was OK if the option was not defined at all.)
5437 17. A "Completed" log line is now written for messages that are removed from
5438 the spool by the -Mrm option.
5440 18. New variables $sender_verify_failure and $recipient_verify_failure contain
5441 information about exactly what failed.
5443 19. Added -dd to debug only the daemon process.
5445 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
5446 handles timeouts, both on the server side and network timeouts. Renamed the
5447 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
5449 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
5450 It is now set to "smtps".
5452 22. $host_address is now set to the target address during the checking of
5453 ignore_target_hosts.
5455 23. When checking ignore_target_hosts for an ipliteral router, no host name was
5456 being passed; this would have caused $sender_host_name to have been used if
5457 matching the list had actually called for a host name (not very likely,
5458 since this list is usually IP addresses). A host name is now passed as
5461 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5462 code that specifies a non-restarting handler (typically sigaction() in
5463 modern systems) in an attempt to fix a rare and obscure crash bug.
5465 25. Narrowed the window for a race in the daemon that could cause it to ignore
5466 SIGCHLD signals. This is not a major problem, because they are used only to
5467 wake it up if nothing else does.
5469 26. A malformed maildirsize file could cause Exim to calculate negative values
5470 for the mailbox size or file count. Odd effects could occur as a result.
5471 The maildirsize information is now recalculated if the size or filecount
5474 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5475 support for a long time. Removed HAVE_SYS_VFS_H.
5477 28. Installed the latest version of exipick from John Jetmore.
5479 29. In an address list, if the pattern was not a regular expression, an empty
5480 subject address (from a bounce message) matched only if the pattern was an
5481 empty string. Non-empty patterns were not even tested. This was the wrong
5482 because it is perfectly reasonable to use an empty address as part of a
5483 database query. An empty address is now tested by patterns that are
5484 lookups. However, all the other forms of pattern expect the subject to
5485 contain a local part and a domain, and therefore, for them, an empty
5486 address still always fails if the pattern is not itself empty.
5488 30. Exim went into a mad DNS loop when attempting to do a callout where the
5489 host was specified on an smtp transport, and looking it up yielded more
5490 than one IP address.
5492 31. Re-factored the code for checking spool and log partition space into a
5493 function that finds that data and another that does the check. The former
5494 is then used to implement four new variables: $spool_space, $log_space,
5495 $spool_inodes, and $log_inodes.
5497 32. The RFC2047 encoding function was originally intended for short strings
5498 such as real names; it was not keeping to the 75-character limit for
5499 encoded words that the RFC imposes. It now respects the limit, and
5500 generates multiple encoded words if necessary. To be on the safe side, I
5501 have increased the buffer size for the ${rfc2047: expansion operator from
5504 33. It is now permitted to omit both strings after an "if" condition; if the
5505 condition is true, the result is "true". As before, when the second string
5506 is omitted, a false condition yields an empty string. This makes it less
5507 cumbersome to write custom ACL and router conditions.
5509 34. Failure to deliver a bounce message always caused it to be frozen, even if
5510 there was an errors_to setting on the router. The errors_to setting is now
5513 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5514 canonical form (fully expanded) before being placed in
5515 $sender_host_address.
5517 36. The table in the code that translates DNS record types into text (T_A to
5518 "A" for instance) was missing entries for NS and CNAME. It is just possible
5519 that this could have caused confusion if both these types were looked up
5520 for the same domain, because the text type is used as part of Exim's
5521 per-process caching. But the chance of anyone hitting this buglet seems
5524 37. The dnsdb lookup has been extended in a number of ways.
5526 (1) There is a new type, "zns", which walks up the domain tree until it
5527 finds some nameserver records. It should be used with care.
5529 (2) There is a new type, "mxh", which is like "mx" except that it returns
5530 just the host names, not the priorities.
5532 (3) It is now possible to give a list of domains (or IP addresses) to be
5533 looked up. The behaviour when one of the lookups defers can be
5534 controlled by a keyword.
5536 (4) It is now possible to specify the separator character for use when
5537 multiple records are returned.
5539 38. The dnslists ACL condition has been extended: it is now possible to supply
5540 a list of IP addresses and/or domains to be looked up in a particular DNS
5543 39. Added log_selector=+queue_time_overall.
5545 40. When running the queue in the test harness, wait just a tad after forking a
5546 delivery process, to get repeatability of debugging output.
5548 41. Include certificate and key file names in error message when GnuTLS fails
5549 to set them up, because the GnuTLS error message doesn't include the name
5550 of the failing file when there is a problem reading it.
5552 42. Allow both -bf and -bF in the same test run.
5554 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
5556 44. The "Exiscan patch" is now merged into the mainline Exim source.
5558 45. Sometimes the final signoff response after QUIT could fail to get
5559 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5560 < 0 before doing a fflush(). This bug looks as though it goes back to the
5561 introduction of TLS in release 3.20, but "sometimes" must have been rare
5562 because the tests only now provoked it.
5564 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
5565 (this can affect the format of dates).
5567 47. exim_tidydb, when checking for the continued existence of a message for
5568 which it has found a message-specific retry record, was not finding
5569 messages that were in split spool directories. Consequently, it was
5570 deleting retry records that should have stayed in existence.
5572 48. Steve fixed some bugs in eximstats.
5574 49. The SPA authentication driver was not abandoning authentication and moving
5575 on to the next authenticator when an expansion was forced to fail,
5576 contradicting the general specification for all authenticators. Instead it
5577 was generating a temporary error. It now behaves as specified.
5579 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
5580 (the order specifies the preference for clients). The order is now AES256,
5581 AES128, 3DES, ARCFOUR128.
5583 51. Small patch to Sieve code - explicitly set From: when generating an
5586 52. Exim crashed if a remote delivery caused a very long error message to be
5587 recorded - for instance if somebody sent an entire SpamAssassin report back
5588 as a large number of 550 error lines. This bug was coincidentally fixed by
5589 increasing the size of one of Exim's internal buffers (big_buffer) that
5590 happened as part of the Exiscan merge. However, to be on the safe side, I
5591 have made the code more robust (and fixed the comments that describe what
5594 53. Now that there can be additional text after "Completed" in log lines (if
5595 the queue_time_overall log selector is set), a one-byte patch to exigrep
5596 was needed to allow it to recognize "Completed" as not the last thing in
5599 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
5600 patch that reportedly fixes this has been added. I am not expert enough to
5601 create a test for it. This is what the patch creator wrote:
5603 "I found a little strange behaviour of ldap code when working with
5604 Windows 2003 AD Domain, where users was placed in more than one
5605 Organization Units. When I tried to give exim partial DN, the exit code
5606 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
5607 But simultaneously result of request was absolutely normal ldap result,
5608 so I produce this patch..."
5610 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
5611 REFERENCE, so I have modified the code to exclude the patch when that macro
5614 55. Some experimental protocols are using DNS PTR records for new purposes. The
5615 keys for these records are domain names, not reversed IP addresses. The
5616 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5617 leaves it alone. Component reversal etc. now happens only for IP addresses.
5620 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
5622 57. Double the size of the debug message buffer (to 2048) so that more of very
5623 long debug lines gets shown.
5625 58. The exicyclog utility now does better if the number of log files to keep
5626 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
5628 59. Two changes related to the smtp_active_hostname option:
5630 (1) $smtp_active_hostname is now available as a variable.
5631 (2) The default for smtp_banner uses $smtp_active_hostname instead
5632 of $primary_hostname.
5634 60. The host_aton() function is supposed to be passed a string that is known
5635 to be a valid IP address. However, in the case of IPv6 addresses, it was
5636 not checking this. This is a hostage to fortune. Exim now panics and dies
5637 if the condition is not met. A case was found where this could be provoked
5638 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5639 components; fortuitously, this particular loophole had already been fixed
5640 by change 4.50/55 above.
5642 If there are any other similar loopholes, the new check in host_aton()
5643 itself should stop them being exploited. The report I received stated that
5644 data on the command line could provoke the exploit when Exim was running as
5645 exim, but did not say which command line option was involved. All I could
5646 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5647 running as the user.
5650 61. There was a buffer overflow vulnerability in the SPA authentication code
5651 (which came originally from the Samba project). I have added a test to the
5652 spa_base64_to_bits() function which I hope fixes it.
5655 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
5656 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
5658 63. The daemon start-up calls getloadavg() while still root for those OS that
5659 need the first call to be done as root, but it missed one case: when
5660 deliver_queue_load_max is set with deliver_drop_privilege. This is
5661 necessary for the benefit of the queue runner, because there is no re-exec
5662 when deliver_drop_privilege is set.
5664 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
5665 This has been fixed.
5667 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5668 was in use, was not putting the data itself into the right store pool;
5669 consequently, it could be overwritten for a subsequent message in the same
5670 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5673 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
5675 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
5676 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
5677 the function were treating the return as a boolean value, which happened to
5678 work because 0=false and not-0=true, but is not correct code.
5680 68. The host_aton() function was not handling scoped IPv6 addresses (those
5681 with, for example, "%eth0" on the end) correctly.
5683 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
5684 negated items (that is, ~something) in unsigned ints. Some compilers
5685 apparently mutter when there is no cast.
5687 70. If an address verification called from an ACL failed, and did not produce a
5688 user-specific message (i.e. there was only a "system" message), nothing was
5689 put in $acl_verify_message. In this situation, it now puts the system
5692 71. Change 4.23/11 added synchronization checking at the start of an SMTP
5693 session; change 4.31/43 added the unwanted input to the log line - except
5694 that it did not do this in the start of session case. It now does.
5696 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
5697 This is wrong and can cause the other end to generate a synchronization
5698 error if it is another Exim or anything else that does the synchronization
5699 check. A QUIT command is no longer sent after a timeout.
5701 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
5702 during host lookups.
5704 74. The defer_ok option of callout verification was not working if it was used
5705 when verifying addresses in header lines, that is, for this case:
5707 verify = header_sender/callout=defer_ok
5709 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
5710 those file descriptors could be used for SMTP connections. If anything
5711 wrote to stderr (the example that came up was "warn" in embedded Perl), it
5712 could be sent to the SMTP client, causing chaos. The daemon now opens
5713 stdin, stdout, and stderr to /dev/null when it puts itself into the
5716 76. Arrange for output from Perl's "warn" command to be written to Exim's main
5717 log by default. The user can override this with suitable Perl magic.
5719 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
5720 the log message when discard triggers, was not working for the DATA ACL or
5721 for the non-SMTP ACL.
5723 78. Error message wording change in sieve.c.
5725 79. If smtp_accept_max_per_host was set, the number of connections could be
5726 restricted to fewer than expected, because the daemon was trying to set up
5727 a new connection before checking whether the processes handling previous
5728 connections had finished. The check for completed processes is now done
5729 earlier. On busy systems, this bug wouldn't be noticed because something
5730 else would have woken the daemon, and it would have reaped the completed
5733 80. If a message was submitted locally by a user whose login name contained one
5734 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
5735 It caused a spool format error. I have fixed the spool reading code. A
5736 related problem was that the "from" clause in the Received: line became
5737 illegal because of the space(s). It is now covered by ${quote_local_part.
5739 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
5742 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
5744 83. Patch from Sieve maintainer to fix unterminated string problem in
5745 "vacation" handling.
5747 84. Some minor changes to the Linux configuration files to help with other
5748 OS variants using glibc.
5750 85. One more patch for Sieve to update vacation handling to latest spec.
5753 ----------------------------------------------------
5754 See the note above about the 4.44 and 4.50 releases.
5755 ----------------------------------------------------
5761 1. Change 4.43/35 introduced a bug that caused file counts to be
5762 incorrectly computed when quota_filecount was set in an appendfile
5765 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5766 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5769 3. Give more explanation in the error message when the command for a transport
5770 filter fails to execute.
5772 4. There are several places where Exim runs a non-Exim command in a
5773 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5774 was being done only for the command run by the queryprogram router. It is
5775 now done for all such subprocesses. The other cases are: ${run, transport
5776 filters, and the commands run by the lmtp and pipe transports.
5778 5. Some older OS have a limit of 256 on the maximum number of file
5779 descriptors. Exim was using setrlimit() to set 1000 as a large value
5780 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5781 systems. I've change it so that if it can't get 1000, it tries for 256.
5783 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5784 was an oversight, and furthermore, ever since the addition of extra
5785 controls (e.g. 4.43/32), the checks on when to allow different forms of
5786 "control" were broken. There should now be diagnostics for all cases when a
5787 control that does not make sense is encountered.
5789 7. $recipients is now available in the predata ACL (oversight).
5791 8. Tidy the search cache before the fork to do a delivery from a message
5792 received from the command line. Otherwise the child will trigger a lookup
5793 failure and thereby defer the delivery if it tries to use (for example) a
5794 cached ldap connection that the parent has called unbind on.
5796 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5797 of $address_data from the recipient verification was clobbered by the
5798 sender verification.
5800 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5801 was its contents. (It was OK if the option was not defined at all.)
5803 11. A "Completed" log line is now written for messages that are removed from
5804 the spool by the -Mrm option.
5806 12. $host_address is now set to the target address during the checking of
5807 ignore_target_hosts.
5809 13. When checking ignore_target_hosts for an ipliteral router, no host name was
5810 being passed; this would have caused $sender_host_name to have been used if
5811 matching the list had actually called for a host name (not very likely,
5812 since this list is usually IP addresses). A host name is now passed as
5815 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5816 code that specifies a non-restarting handler (typically sigaction() in
5817 modern systems) in an attempt to fix a rare and obscure crash bug.
5819 15. Narrowed the window for a race in the daemon that could cause it to ignore
5820 SIGCHLD signals. This is not a major problem, because they are used only to
5821 wake it up if nothing else does.
5823 16. A malformed maildirsize file could cause Exim to calculate negative values
5824 for the mailbox size or file count. Odd effects could occur as a result.
5825 The maildirsize information is now recalculated if the size or filecount
5828 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5829 support for a long time. Removed HAVE_SYS_VFS_H.
5831 18. Updated exipick to current release from John Jetmore.
5833 19. Allow an empty sender to be matched against a lookup in an address list.
5834 Previously the only cases considered were a regular expression, or an
5837 20. Exim went into a mad DNS lookup loop when doing a callout where the
5838 host was specified on the transport, if the DNS lookup yielded more than
5841 21. The RFC2047 encoding function was originally intended for short strings
5842 such as real names; it was not keeping to the 75-character limit for
5843 encoded words that the RFC imposes. It now respects the limit, and
5844 generates multiple encoded words if necessary. To be on the safe side, I
5845 have increased the buffer size for the ${rfc2047: expansion operator from
5848 22. Failure to deliver a bounce message always caused it to be frozen, even if
5849 there was an errors_to setting on the router. The errors_to setting is now
5852 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5853 canonical form (fully expanded) before being placed in
5854 $sender_host_address.
5856 24. Updated eximstats to version 1.33
5858 25. Include certificate and key file names in error message when GnuTLS fails
5859 to set them up, because the GnuTLS error message doesn't include the name
5860 of the failing file when there is a problem reading it.
5862 26. Expand error message when OpenSSL has problems setting up cert/key files.
5865 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5866 (this can affect the format of dates).
5868 28. exim_tidydb, when checking for the continued existence of a message for
5869 which it has found a message-specific retry record, was not finding
5870 messages that were in split spool directories. Consequently, it was
5871 deleting retry records that should have stayed in existence.
5873 29. eximstats updated to version 1.35
5874 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5875 1.35 - bugfix such that pie charts by volume are generated correctly
5877 30. The SPA authentication driver was not abandoning authentication and moving
5878 on to the next authenticator when an expansion was forced to fail,
5879 contradicting the general specification for all authenticators. Instead it
5880 was generating a temporary error. It now behaves as specified.
5882 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5883 (the order specifies the preference for clients). The order is now AES256,
5884 AES128, 3DES, ARCFOUR128.
5886 31. Small patch to Sieve code - explicitly set From: when generating an
5889 32. Exim crashed if a remote delivery caused a very long error message to be
5890 recorded - for instance if somebody sent an entire SpamAssassin report back
5891 as a large number of 550 error lines. This bug was coincidentally fixed by
5892 increasing the size of one of Exim's internal buffers (big_buffer) that
5893 happened as part of the Exiscan merge. However, to be on the safe side, I
5894 have made the code more robust (and fixed the comments that describe what
5897 33. Some experimental protocols are using DNS PTR records for new purposes. The
5898 keys for these records are domain names, not reversed IP addresses. The
5899 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5900 leaves it alone. Component reversal etc. now happens only for IP addresses.
5903 34. The host_aton() function is supposed to be passed a string that is known
5904 to be a valid IP address. However, in the case of IPv6 addresses, it was
5905 not checking this. This is a hostage to fortune. Exim now panics and dies
5906 if the condition is not met. A case was found where this could be provoked
5907 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5908 components; fortuitously, this particular loophole had already been fixed
5909 by change 4.50/55 or 4.44/33 above.
5911 If there are any other similar loopholes, the new check in host_aton()
5912 itself should stop them being exploited. The report I received stated that
5913 data on the command line could provoke the exploit when Exim was running as
5914 exim, but did not say which command line option was involved. All I could
5915 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5916 running as the user.
5919 35. There was a buffer overflow vulnerability in the SPA authentication code
5920 (which came originally from the Samba project). I have added a test to the
5921 spa_base64_to_bits() function which I hope fixes it.
5924 36. The daemon start-up calls getloadavg() while still root for those OS that
5925 need the first call to be done as root, but it missed one case: when
5926 deliver_queue_load_max is set with deliver_drop_privilege. This is
5927 necessary for the benefit of the queue runner, because there is no re-exec
5928 when deliver_drop_privilege is set.
5930 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5931 was in use, was not putting the data itself into the right store pool;
5932 consequently, it could be overwritten for a subsequent message in the same
5933 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5936 38. Sometimes the final signoff response after QUIT could fail to get
5937 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5938 < 0 before doing a fflush(). This bug looks as though it goes back to the
5939 introduction of TLS in release 3.20, but "sometimes" must have been rare
5940 because the tests only now provoked it.
5946 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5947 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5948 Now it returns an integer. A number of calls were still expecting a T/F
5949 return. Fortuitously, in all cases, the tests worked in OK situations,
5950 which is the norm. However, things would have gone wrong on any write
5951 failures on the smtp file descriptor. This function is used when sending
5952 messages over SMTP and also when doing verify callouts.
5954 2. When Exim is called to do synchronous delivery of a locally submitted
5955 message (the -odf or -odi options), it no longer closes stderr before doing
5958 3. Implemented the mua_wrapper option.
5960 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5962 5. Implemented the functions header_remove(), header_testname(),
5963 header_add_at_position(), and receive_remove_recipient(), and exported them
5966 6. If an ACL "warn" statement specified the addition of headers, Exim already
5967 inserted X-ACL-Warn: at the start if there was no header name. However, it
5968 was not making this test for the second and subsequent header lines if
5969 there were newlines in the string. This meant that an invalid header could
5970 be inserted if Exim was badly configured.
5972 7. Allow an ACL "warn" statement to add header lines at the start or after all
5973 the Received: headers, as well as at the end.
5975 8. Added the rcpt_4xx retry error code.
5977 9. Added postmaster_mailfrom=xxx to callout verification option.
5979 10. Added mailfrom=xxxx to the callout verification option, for verify=
5982 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5983 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5985 12. Inserted some casts to stop certain compilers warning when using pointer
5986 differences as field lengths or precisions in printf-type calls (mostly
5987 affecting debugging statements).
5989 13. Added optional readline() support for -be (dynamically loaded).
5991 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5992 same clock tick as a message's arrival, so that its received time was the
5993 same as the "first fail" time on the retry record, and that message
5994 remained on the queue past the ultimate address timeout, every queue runner
5995 would try a delivery (because it was past the ultimate address timeout) but
5996 after another failure, the ultimate address timeout, which should have then
5997 bounced the address, did not kick in. This was a "< instead of <=" error;
5998 in most cases the first failure would have been in the next clock tick
5999 after the received time, and all would be well.
6001 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
6002 being recognized when the domain list was tested by the match_domain
6003 condition in an expansion string.
6005 16. Added the ${str2b64: operator.
6007 17. Exim was always calling setrlimit() to set a large limit for the number of
6008 processes, without checking whether the existing limit was already
6009 adequate. (It did check for the limit on file descriptors.) Furthermore,
6010 errors from getrlimit() and setrlimit() were being ignored. Now they are
6011 logged to the main and panic logs, but Exim does carry on, to try to do its
6012 job under whatever limits there are.
6014 18. Imported PCRE 5.0.
6016 19. Trivial typo in log message " temporarily refused connection" (the leading
6019 20. If the log selector return_path_on_delivery was set and an address was
6020 redirected to /dev/null, the delivery process crashed because it assumed
6021 that a return path would always be set for a "successful" delivery. In this
6022 case, the whole delivery is bypassed as an optimization, and therefore no
6025 21. Internal re-arrangement: the function for sending a challenge and reading
6026 a response while authentication was assuming a zero-terminated challenge
6027 string. It's now changed to take a pointer and a length, to allow for
6028 binary data in such strings.
6030 22. Added the cyrus_sasl authenticator (code supplied by MBM).
6032 23. Exim was not respecting finduser_retries when seeking the login of the
6033 uid under which it was called; it was always trying 10 times. (The default
6034 setting of finduser_retries is zero.) Also, it was sleeping after the final
6035 failure, which is pointless.
6037 24. Implemented tls_on_connect_ports.
6039 25. Implemented acl_smtp_predata.
6041 26. If the domain in control=submission is set empty, Exim assumes that the
6042 authenticated id is a complete email address when it generates From: or
6043 Sender: header lines.
6045 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
6046 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
6047 chown and chgrp in /bin and hostname in /usr/bin.
6049 28. Exim was keeping the "process log" file open after each use, just as it
6050 does for the main log. This opens the possibility of it remaining open for
6051 long periods when the USR1 signal hits a daemon. Occasional processlog
6052 errors were reported, that could have been caused by this. Anyway, it seems
6053 much more sensible not to leave this file open at all, so that is what now
6056 29. The long-running daemon process does not normally write to the log once it
6057 has entered its main loop, and it closes the log before doing so. This is
6058 so that log files can straightforwardly be renamed and moved. However,
6059 there are a couple of unusual error situations where the daemon does write
6060 log entries, and I had neglected to close the log afterwards.
6062 30. The text of an SMTP error response that was received during a remote
6063 delivery was being truncated at 512 bytes. This is too short for some of
6064 the long messages that one sometimes sees. I've increased the limit to
6067 31. It is now possible to make retry rules that apply only when a message has a
6068 specific sender, in particular, an empty sender.
6070 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
6071 it possible to be selective about when SMTP synchronization is enforced.
6073 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
6075 32. Implemented hosts_connection_nolog.
6077 33. Added an ACL for QUIT.
6079 34. Setting "delay_warning=" to disable warnings was not working; it gave a
6082 35. Added mailbox_size and mailbox_filecount to appendfile.
6084 36. Added control = no_multiline_responses to ACLs.
6086 37. There was a bug in the logic of the code that waits for the clock to tick
6087 in the case where the clock went backwards by a substantial amount such
6088 that the microsecond fraction of "now" was more than the microsecond
6089 fraction of "then" (but the whole seconds number was less).
6091 38. Added support for the libradius Radius client library this is found on
6092 FreeBSD (previously only the radiusclient library was supported).
6098 1. When certain lookups returned multiple values in the form name=value, the
6099 quoting of the values was not always being done properly. Specifically:
6100 (a) If the value started with a double quote, but contained no whitespace,
6102 (b) If the value contained whitespace other than a space character (i.e.
6103 tabs or newlines or carriage returns) it was not quoted.
6104 This fix has been applied to the mysql and pgsql lookups by writing a
6105 separate quoting function and calling it from the lookup code. The fix
6106 should probably also be applied to nisplus, ibase and oracle lookups, but
6107 since I cannot test any of those, I have not disturbed their existing code.
6109 2. A hit in the callout cache for a specific address caused a log line with no
6110 reason for rejecting RCPT. Now it says "Previous (cached) callout
6111 verification failure".
6113 3. There was an off-by-one bug in the queryprogram router. An over-long
6114 return line was truncated at 256 instead of 255 characters, thereby
6115 overflowing its buffer with the terminating zero. As well as fixing this, I
6116 have increased the buffer size to 1024 (and made a note to document this).
6118 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
6119 when Exim is waiting for an SMTP response from a remote server, Exim
6120 restarts its select() call on the socket, thereby resetting its timeout.
6121 This is not a problem when such interrupts are rare. Somebody set up a cron
6122 job to run exiwhat every 2 minutes, which is less than the normal select()
6123 timeout (5 or 10 minutes). This meant that the select() timeout never
6124 kicked in because it was always reset. I have fixed this by comparing the
6125 time when an interrupt arrives with the time at the start of the first call
6126 to select(). If more time than the timeout has elapsed, the interrupt is
6127 treated as a timeout.
6129 5. Some internal re-factoring in preparation for the addition of Sieve
6130 extensions (by MH). In particular, the "personal" test is moved to a
6131 separate function, and given an option for scanning Cc: and Bcc: (which is
6132 not set for Exim filters).
6134 6. When Exim created an email address using the login of the caller as the
6135 local part (e.g. when creating a From: or Sender: header line), it was not
6136 quoting the local part when it contained special characters such as @.
6138 7. Installed new OpenBSD configuration files.
6140 8. Reworded some messages for syntax errors in "and" and "or" conditions to
6141 try to make them clearer.
6143 9. Callout options, other than the timeout value, were being ignored when
6144 verifying sender addresses in header lines. For example, when using
6146 verify = header_sender/callout=no_cache
6148 the cache was (incorrectly) being used.
6150 10. Added a missing instance of ${EXE} to the exim_install script; this affects
6151 only the Cygwin environment.
6153 11. When return_path_on_delivery was set as a log selector, if different remote
6154 addresses in the same message used different return paths and parallel
6155 remote delivery occurred, the wrong values would sometimes be logged.
6156 (Whenever a remote delivery process finished, the return path value from
6157 the most recently started remote delivery process was logged.)
6159 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
6160 lines when AUTH and/or TLS are in use. This is the "received protocol"
6161 field. Exim used to use "asmtp" for authenticated SMTP, without any
6162 indication (in the protocol name) for TLS use. Now it follows the RFC and
6163 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
6164 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
6165 names appear in log lines as well as in Received: header lines.
6167 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
6168 extensions, and comparison tests, and to fix some bugs.
6170 14. Changes to the "personal" filter test:
6172 (1) The test was buggy in that it was just doing the equivalent of
6173 "contains" tests on header lines. For example, if a user's address was
6174 anne@some.where, the "personal" test would incorrectly be true for
6176 To: susanne@some.where
6178 This test is now done by extracting each address from the header in turn,
6179 and checking the entire address. Other tests that are part of "personal"
6180 are now done using regular expressions (for example, to check local parts
6181 of addresses in From: header lines).
6183 (2) The list of non-personal local parts in From: addresses has been
6184 extended to include "listserv", "majordomo", "*-request", and "owner-*",
6185 taken from the Sieve specification recommendations.
6187 (3) If the message contains any header line starting with "List-" it is
6188 treated as non-personal.
6190 (4) The test for "circular" in the Subject: header line has been removed
6191 because it now seems ill-conceived.
6193 15. Minor typos in src/EDITME comments corrected.
6195 16. Installed latest exipick from John Jetmore.
6197 17. If headers_add on a router specified a text string that was too long for
6198 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
6199 of string_sprintf() is now avoided.
6201 18. $message_body_size was not set (it was always zero) when running the DATA
6202 ACL and the local_scan() function.
6204 19. For the "mail" command in an Exim filter, no default was being set for
6205 the once_repeat time, causing a random time value to be used if "once" was
6206 specified. (If the value happened to be <= 0, no repeat happened.) The
6207 default is now 0s, meaning "never repeat". The "vacation" command was OK
6208 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
6209 (I found it when inspecting the code).
6211 20. There is now an overall timeout for performing a callout verification. It
6212 defaults to 4 times the callout timeout, which applies to individual SMTP
6213 commands during the callout. The overall timeout applies when there is more
6214 than one host that can be tried. The timeout is checked before trying the
6215 next host. This prevents very long delays if there are a large number of
6216 hosts and all are timing out (e.g. when the network connections are timing
6217 out). The value of the overall timeout can be changed by specifying an
6218 additional sub-option for "callout", called "maxwait". For example:
6220 verify = sender/callout=5s,maxwait=20s
6222 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
6223 to the end before writing, but this should make it even safer).
6225 22. Exim was forgetting that it had advertised PIPELINING for the second and
6226 subsequent messages on an SMTP connection. It was also not resetting its
6227 memory on STARTTLS and an internal HELO.
6229 23. When Exim logs an SMTP synchronization error within a session, it now
6230 records whether PIPELINING has been advertised or not.
6232 24. Added 3 instances of "(long int)" casts to time_t variables that were being
6233 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
6234 rather than long int.
6236 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
6238 26. Added the never_mail option to autoreply.
6244 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
6245 crash if the getsockname() call failed; this can happen if a connection is
6246 closed very soon after it is established. The problem was simply in the
6247 order in which certain operations were done, causing Exim to try to write
6248 to the SMTP stream before it had set up the file descriptor. The bug has
6249 been fixed by making things happen in the correct order.
6255 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
6256 before the connection was closed, thus losing the rejection response.
6258 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
6259 some early Solaris releases, but causes trouble in current releases where
6260 socklen_t is defined.
6262 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
6265 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
6268 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
6269 that releasing the top of it at the end releases what was used for sub-
6270 expansions (unless the block got too big). However, discard this block if
6271 the first thing is a variable or header, so that we can use its block when
6272 it is dynamic (useful for very large $message_headers, for example).
6274 6. Lookups now cache *every* query, not just the most recent. A new, separate
6275 store pool is used for this. It can be recovered when all lookup caches are
6276 flushed. Lookups now release memory at the end of their result strings.
6277 This has involved some general refactoring of the lookup sources.
6279 7. Some code has been added to the store_xxx() functions to reduce the amount
6280 of flapping under certain conditions.
6282 8. log_incoming_interface used to affect only the <= reception log lines. Now
6283 it causes the local interface and port to be added to several more SMTP log
6284 lines, for example "SMTP connection from", and rejection lines.
6286 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
6288 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
6290 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
6291 could be overwritten at the end of the current message (or the start of a
6292 new message if it was set in a HELO ACL). The value is now preserved for
6293 the duration of the SMTP connection.
6295 12. If a transport had a headers_rewrite setting, and a matching header line
6296 contained an unqualified address, that address was qualified, even if it
6297 did not match any rewriting rules. The underlying bug was that the values
6298 of the flags that permit the existence of unqualified sender and recipient
6299 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
6300 non-local messages, and by -bnq for local messages) were not being
6301 preserved with the message after it was received.
6303 13. When Exim was logging an SMTP synchronization error, it could sometimes log
6304 "next input=" as part of the text comprising the host identity instead of
6305 the correct text. The code was using the same buffer for two different
6306 strings. However, depending on which order the printing function evaluated
6307 its arguments, the bug did not always show up. Under Linux, for example, my
6308 test suite worked just fine.
6310 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
6311 doesn't work with some older versions of Perl. It has been changed to "my",
6312 which in any case is probably the better facility to use.
6314 15. A really picky compiler found some instances of statements for creating
6315 error messages that either had too many or two few arguments for the format
6318 16. The size of the buffer for calls to the DNS resolver has been increased
6319 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
6320 for addresses that have a lot of PTR records. This alleviates a problem; it
6321 does not fully solve it.
6323 17. A dnsdb lookup for PTR records that receives more data than will fit in the
6324 buffer now truncates the list and logs the incident, which is the same
6325 action as happens when Exim is looking up a host name and its aliases.
6326 Previously in this situation something unpredictable would happen;
6327 sometimes it was "internal error: store_reset failed".
6329 18. If a server dropped the connection unexpectedly when an Exim client was
6330 using GnuTLS and trying to read a response, the client delivery process
6331 crashed while trying to generate an error log message.
6333 19. If a "warn" verb in an ACL added multiple headers to a message in a single
6334 string, for example:
6336 warn message = H1: something\nH2: something
6338 the text was added as a single header line from Exim's point of view
6339 though it ended up OK in the delivered message. However, searching for the
6340 second and subsequent header lines using $h_h2: did not work. This has been
6341 fixed. Similarly, if a system filter added multiple headers in this way,
6342 the routers could not see them.
6344 20. Expanded the error message when iplsearch is called with an invalid key to
6345 suggest using net-iplsearch in a host list.
6347 21. When running tests using -bh, any delays imposed by "delay" modifiers in
6348 ACLs are no longer actually imposed (and a message to that effect is
6351 22. If a "gecos" field in a passwd entry contained escaped characters, in
6352 particular, if it contained a \" sequence, Exim got it wrong when building
6353 a From: or a Sender: header from that name. A second bug also caused
6354 incorrect handling when an unquoted " was present following a character
6355 that needed quoting.
6357 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
6358 was not being matched caselessly.
6360 24. Arranged for all hyphens in the exim.8 source to be escaped with
6363 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
6364 properly. Recipient callout cache records were still being keyed to include
6365 the sender, even when use_sender was set false. This led to far more
6366 callouts that were necessary. The sender is no longer included in the key
6367 when use_sender is false.
6369 26. Added "control = submission" modifier to ACLs.
6371 27. Added the ${base62d: operator to decode base 62 numbers.
6373 28. dnsdb lookups can now access SRV records.
6375 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
6376 the configuration file.
6378 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
6379 (-v) mode. This makes the output for a verbose queue run more intelligible.
6381 31. Added a use_postmaster feature to recipient callouts.
6383 32. Added the $body_zerocount variable, containing the number of binary zero
6384 bytes in the message body.
6386 33. The time of last modification of the "new" subdirectory is now used as the
6387 "mailbox time last read" when there is a quota error for a maildir
6390 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
6392 35. Added +ignore_unknown as a special item in host lists.
6394 36. Code for decoding IPv6 addresses in host lists is now included, even if
6395 IPv6 support is not being compiled. This fixes a bug in which an IPv6
6396 address was recognized as an IP address, but was then not correctly decoded
6397 into binary, causing unexpected and incorrect effects when compared with
6404 1. Very minor rewording of debugging text in manualroute to say "list of
6405 hosts" instead of "hostlist".
6407 2. If verify=header_syntax was set, and a header line with an unqualified
6408 address (no domain) and a large number of spaces between the end of the
6409 name and the colon was received, the reception process suffered a buffer
6410 overflow, and (when I tested it) crashed. This was caused by some obsolete
6411 code that should have been removed. The fix is to remove it!
6413 3. When running in the test harness, delay a bit after writing a bounce
6414 message to get a bit more predictability in the log output.
6416 4. Added a call to search_tidyup() just before forking a reception process. In
6417 theory, someone could use a lookup in the expansion of smtp_accept_max_
6418 per_host which, without the tidyup, could leave open a database connection.
6420 5. Added the variables $recipient_data and $sender_data which get set from a
6421 lookup success in an ACL "recipients" or "senders" condition, or a router
6422 "senders" option, similar to $domain_data and $local_part_data.
6424 6. Moved the writing of debug_print from before to after the "senders" test
6427 7. Change 4.31/66 (moving the time when the Received: is generated) caused
6428 problems for message scanning, either using a data ACL, or using
6429 local_scan() because the Received: header was not generated till after they
6430 were called (in order to set the time as the time of reception completion).
6431 I have revised the way this works. The header is now generated after the
6432 body is received, but before the ACL or local_scan() are called. After they
6433 are run, the timestamp in the header is updated.
6439 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
6440 before starting a queue runner without re-exec. This happened only when
6441 deliver_drop_privilege was set or when the Exim user was set to root. The
6442 effect of the bug was that timeouts during subsequent deliveries caused
6443 crashes instead of being properly handled. The handler is now left at its
6444 default (and expected) setting.
6446 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
6447 message, again when deliver_drop_privilege is set or Exim is run as root.
6448 The bug described in (1) was not present in this case, but the tidying up
6449 of the other signals was missing. I have made the two cases consistent.
6451 3. The ignore_target_hosts setting on a manualroute router was being ignored
6452 for hosts that were looked up using the /MX notation.
6454 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
6457 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
6458 operated on the sender address. After changing the $sender_address to <>
6459 for the sender address verify, Exim was re-instated it as the original
6460 (before rewriting) address, but remembering that it had rewritten it, so it
6461 wasn't rewriting it again. This bug also had the effect of breaking the
6462 sender address verification caching when the sender address was rewritten.
6464 6. The ignore_target_hosts option was being ignored by the ipliteral router.
6465 This has been changed so that if the ip literal address matches
6466 ignore_target_hosts, the router declines.
6468 7. Added expansion conditions match_domain, match_address, and match_local_
6469 part (NOT match_host).
6471 8. The placeholder for the Received: header didn't have a length field set.
6473 9. Added code to Exim itself and to exim_lock to test for a specific race
6474 condition that could lead to file corruption when using MBX delivery. The
6475 issue is with the lockfile that is created in /tmp. If this file is removed
6476 after a process has opened it but before that process has acquired a lock,
6477 there is the potential for a second process to recreate the file and also
6478 acquire a lock. This could lead to two Exim processes writing to the file
6479 at the same time. The added code performs the same test as UW imapd; it
6480 checks after acquiring the lock that its file descriptor still refers to
6481 the same named file.
6483 10. The buffer for building added header lines was of fixed size, 8192 bytes.
6484 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
6487 11. Added the smtp_active_hostname option. If used, this will typically be made
6488 to depend on the incoming interface address. Because $interface_address is
6489 not set up until the daemon has forked a reception process, error responses
6490 that can happen earlier (such as "too many connections") no longer contain
6493 12. If an expansion in a condition on a "warn" statement fails because a lookup
6494 defers, the "warn" statement is abandoned, and the next ACL statement is
6495 processed. Previously this caused the whole ACL to be aborted.
6497 13. Added the iplsearch lookup type.
6499 14. Added ident_timeout as a log selector.
6501 15. Added tls_certificate_verified as a log selector.
6503 16. Added a global option tls_require_ciphers (compare the smtp transport
6504 option of the same name). This controls incoming TLS connections.
6506 17. I finally figured out how to make tls_require_ciphers do a similar thing
6507 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
6508 before starting the TLS session.
6510 18. Tabs are now shown as \t in -bP output.
6512 19. If the log selector return_path_on_delivery was set, Exim crashed when
6513 bouncing a message because it had too many Received: header lines.
6515 20. If two routers both had headers_remove settings, and the first one included
6516 a superfluous trailing colon, the final name in the first list and the
6517 first name in the second list were incorrectly joined into one item (with a
6518 colon in the middle).
6524 1. Added -C and -D options to the exinext utility, mainly to make it easier
6525 to include in the automated testing, but these could be helpful when
6526 multiple configurations are in use.
6528 2. The exinext utility was not formatting the output nicely when there was
6529 an alternate port involved in the retry record key, nor when there was a
6530 message id as well (for retries that were specific to a specific message
6531 and a specific host). It was also confused by IPv6 addresses, because of
6532 the additional colons they contain. I have fixed the IPv4 problem, and
6533 patched it up to do a reasonable job for IPv6.
6535 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
6536 delivery, the log line now contains "pipelined" if PIPELINING was used.
6538 4. An SMTP transport process used to panic and die if the bind() call to set
6539 an explicit outgoing interface failed. This has been changed; it is now
6540 treated in the same way as a connect() failure.
6542 5. A reference to $sender_host_name in the part of a conditional expansion
6543 that was being skipped was still causing a DNS lookup. This no longer
6546 6. The def: expansion condition was not recognizing references to header lines
6547 that used bh_ and bheader_.
6549 7. Added the _cache feature to named lists.
6551 8. The code for checking quota_filecount in the appendfile transport was
6552 allowing one more file than it should have been.
6554 9. For compatibility with Sendmail, the command line option
6562 and sets the incoming protocol and host name (for trusted callers). The
6563 host name and its colon can be omitted when only the protocol is to be set.
6564 Note the Exim already has two private options, -pd and -ps, that refer to
6565 embedded Perl. It is therefore impossible to set a protocol value of "d" or
6566 "s", but I don't think that's a major issue.
6568 10. A number of refactoring changes to the code, none of which should affect
6571 (a) The number of logging options was getting close to filling up the
6572 32-bit word that was used as a bit map. I have split them into two classes:
6573 those that are passed in the argument to log_write(), and those that are
6574 only ever tested independently outside of that function. These are now in
6575 separate 32-bit words, so there is plenty of room for expansion again.
6576 There is no change in the user interface or the logging behaviour.
6578 (b) When building, for example, log lines, the code previously used a
6579 macro that called string_cat() twice, in order to add two strings. This is
6580 not really sufficiently general. Furthermore, there was one instance where
6581 it was actually wrong because one of the argument was used twice, and in
6582 one call a function was used. (As it happened, calling the function twice
6583 did not affect the overall behaviour.) The macro has been replaced by a
6584 function that can join an arbitrary number of extra strings onto a growing
6587 (c) The code for expansion conditions now uses a table and a binary chop
6588 instead of a serial search (which was left over from when there were very
6589 few conditions). Also, it now recognizes conditions like "pam" even when
6590 the relevant support is not compiled in: a suitably worded error message is
6591 given if an attempt is made to use such a condition.
6593 11. Added ${time_interval:xxxxx}.
6595 12. A bug was causing one of the ddress fields not to be passed back correctly
6596 from remote delivery subprocesses. The field in question was not being
6597 subsequently used, so this caused to problems in practice.
6599 13. Added new log selectors queue_time and deliver_time.
6601 14. Might have fixed a bug in maildirsizefile handling that threw up
6602 "unexpected character" debug warnings, and recalculated the data
6603 unnecessarily. In any case, I expanded the warning message to give more
6606 15. Added the message "Restricted characters in address" to the statements in
6607 the default ACL that block characters like @ and % in local parts.
6609 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
6610 Three changes have been made:
6612 (a) There was a serious bug; a negative response to MAIL caused the whole
6613 recipient domain to be cached as invalid, thereby blocking all messages
6614 to all local parts at the same domain, from all senders. This bug has
6615 been fixed. The domain is no longer cached after a negative response to
6616 MAIL if the sender used is not empty.
6618 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
6621 (c) A new callout option, "use_sender" has been added for people who want
6622 the modified behaviour.
6628 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
6631 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
6632 indeed breaks things for older releases.
6634 3. Added additional logging to the case where there is a problem reading data
6635 from a filter that is running in a subprocess using a pipe, in order to
6636 try to track down a specific problem.
6638 4. Testing facility fudge: when running in the test harness and attempting
6639 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
6640 getting "No route to host". Convert this to a timeout.
6642 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
6645 6. Some OS don't have socklen_t but use size_t instead. This affects the
6646 fifth argument of getsockopt() amongst other things. This is now
6647 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
6648 can be set for individual OS. I have set it for SunOS5, OSF1, and
6649 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
6650 some earlier ones do not.
6652 7. Change 4.30/15 was not doing the test caselessly.
6654 8. The standard form for an IPv6 address literal was being rejected by address
6655 parsing in, for example, MAIL and RCPT commands. An example of this kind of
6656 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6657 this, as well as the form without the "IPv6" on the front (but only when
6658 address literals are enabled, of course).
6660 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
6662 10. Exim crashed if a message with an empty sender address specified by -f
6663 encountered a router with an errors_to setting. This could be provoked only
6664 by a command such as
6668 where an empty string was supplied; "<>" did not hit this bug.
6670 11. Installed PCRE release 4.5.
6672 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
6673 remained set. It is now erased.
6675 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
6676 times from message ids (which are base 36 rather than the normal 62).
6678 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
6679 were being counted as actual protocol errors, and logged if the log
6680 selector +smtp_protocol_error was set. One cannot be perfect in this test,
6681 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
6682 and DATA following a set of rejected RCPTs do not count as protocol errors.
6683 In other words, Exim assumes they were pipelined, though this may not
6684 actually be the case. Of course, in all cases the client gets an
6685 appropriate error code.
6687 15. If a lookup fails in an ACL condition, a message about the failure may
6688 be available; it is used if testing the ACL cannot continue, because most
6689 such messages specify what the cause of the deferral is. However, some
6690 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
6691 that caused an old message to be retained and used if a later statement
6692 caused a defer, replacing the real cause of the deferral.
6694 16. If an IP address had so many PTR records that the DNS lookup buffer
6695 was not large enough to hold them, Exim could crash while trying to process
6696 the truncated data. It now detects and logs this case.
6698 17. Further to 4.21/58, another change has been made: if (and only if) the
6699 first line of a message (the first header line) ends with CRLF, a bare LF
6700 in a subsequent header line has a space inserted after it, so as not to
6701 terminate the header.
6703 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
6704 unnecessarily, used atoi() instead of strtol(), and didn't check the
6705 termination when getting file sizes from file names by regex.
6707 19. Completely re-implemented the support for maildirsize files, in the light
6708 of a number of problems with the previous contributed implementation
6709 (4.30/29). In particular:
6711 . If the quota is zero, the maildirsize file is maintained, but no quota is
6714 . If the maildir directory does not exist, it is created before any attempt
6715 to write a maildirsize file.
6717 . The quota value in the file is just a cache; if the quota is changed in
6718 the transport, the new value overrides.
6720 . A regular expression is available for excluding directories from the
6723 20. The autoreply transport checks the characters in options that define the
6724 message's headers; it allows continued headers, but it was checking with
6725 isspace() after an embedded newline instead of explicitly looking for a
6728 21. If all the "regular" hosts to which an address was routed had passed their
6729 expiry times, and had not reached their retry times, the address was
6730 bounced, even if fallback hosts were defined. Now Exim should go on to try
6733 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
6734 equivalent code in the SMTP transport. Some hosts send humungous responses
6735 to HELO/EHLO, more than 1024 it seems.
6737 23. Refactoring: code in filter.c used (void *) for "any old type" but this
6738 gives compiler warnings in some environments. I've now done it "properly",
6741 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
6742 (because of problems with the built-in one) was declared to return uschar *
6743 instead of char *, causing compiler failure.
6745 25. Fixed a file descriptor leak when processing alias/forward files.
6747 26. Fixed a minor format string issue in dbfn.c.
6749 27. Typo in exim.c: ("dmbnz" for "dbmnz").
6751 28. If a filter file refered to $h_xxx or $message_headers, and the headers
6752 contained RFC 2047 "words", Exim's memory could, under certain conditions,
6755 29. When a sender address is verified, it is cached, to save repeating the test
6756 when there is more than one recipient in a message. However, when the
6757 verification involves a callout, it is possible for different callout
6758 options to be set for different recipients. It is too complicated to keep
6759 track of this in the cache, so now Exim always runs a verification when a
6760 callout is required, relying on the callout cache for the optimization.
6761 The overhead is duplication of the address routing, but this should not be
6764 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
6765 to be verified with callout=postmaster, and the main callout worked but the
6766 postmaster check failed, the verification correctly failed. However, if a
6767 subsequent RCPT command asked for sender verification *without* the
6768 postmaster check, incorrect caching caused this verification also to fail,
6771 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
6772 it was not caching the DNS options (qualify_single, search_parents) that
6773 were used when the lookup failed. A subsequent lookup with different
6774 options therefore always gave the same answer, though there were cases
6775 where it should not have. (Example: a "domains = !$mx_any" option on a
6776 dnslookup router: the "domains" option is always processed without any
6777 widening, but the router might have qualify_single set.) Now Exim uses the
6778 cached value only when the same options are set.
6780 32. Added John Jetmore's "exipick" utility to the distribution.
6782 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
6783 than a timeout (e.g. a certificate is required, and is not provided), an
6784 Exim server now closes the connection immediately. Previously it waited for
6785 the client to close - but if the client is SSL, it seems that they each
6786 wait for each other, leading to a delay before one of them times out.
6788 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
6789 maintained 0.8.x compatibility because I don't think many are using it, and
6790 it is clearly obsolete.
6792 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
6795 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
6796 client certificate was expired. A simple patch fixes this, though I don't
6797 understand the full logic of why the verify callback is called multiple
6800 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
6801 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
6802 which causes problems with some clients (such as the Certicom SSL Plus
6803 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
6804 disables the coutermeasure allowing Eudora to connect."
6806 38. Exim was not checking that a write() to a log file succeeded. This could
6807 lead to Bad Things if a log got too big, in particular if it hit a file
6808 size limit. Exim now panics and dies if it cannot write to a log file, just
6809 as it does if it cannot open a log file.
6811 39. Modified OS/Makefile-Linux so that it now contains
6813 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
6815 The two -D definitions ensure that Exim is compiled with large file
6816 support, which makes it possible to handle log files that are bigger than
6819 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
6820 instance) a domain was checked against a named list that involved a lookup,
6821 causing $domain_data to be set, then another domain was checked against the
6822 same list, then the first domain was re-checked, the value of $domain_data
6823 after the final check could be wrong. In particular, if the second check
6824 failed, it could be set empty. This bug probably also applied to
6827 41. The strip_trailing_dot option was not being applied to the address given
6828 with the -f command-line option.
6830 42. The code for reading a message's header from the spool was incrementing
6831 $received_count, but never initializing it. This meant that the value was
6832 incorrect (doubled) while delivering a message in the same process in which
6833 it was received. In the most common configuration of Exim, this never
6834 happens - a fresh exec is done - but it can happen when
6835 deliver_drop_privilege is set.
6837 43. When Exim logs an SMTP synchronization error - client data sent too soon -
6838 it now includes up to 150 characters of the unexpected data in the log
6841 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6842 and building data strings. The size of both of these buffers was 10 000
6843 bytes - far larger than anybody would *ever* want, thought I. Needless to
6844 say, somebody hit the limit. I have increased the maximum line length to
6845 20 000 and the maximum data length of concatenated lines to 100 000. I have
6846 also fixed two bugs, because there was no checking on these buffers. Tsk,
6847 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6848 buffer is too small.
6850 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6851 lsearch lookups. Now it does.
6853 46. When parsing a route_list item in a manualroute router, a fixed-length
6854 buffer was used for the list of hosts. I made this 1024 bytes long,
6855 thinking that nobody would ever have a list of hosts that long. Wrong.
6856 Somebody had a whole pile of complicated expansion conditions, and the
6857 string was silently truncated, leading to an expansion error. It turns out
6858 that it is easier to change to an unlimited length (owing to other changes
6859 that have happened since this code was originally written) than to build
6860 structure for giving a limitation error. The length of the item that
6861 expands into the list of hosts is now unlimited.
6863 47. The lsearch lookup could not handle data where the length of text line was
6864 more than 4095 characters. Such lines were truncated, leading to shortened
6865 data being returned. It should now handle lines of any length.
6867 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6868 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6871 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6872 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6873 no longer happens in comment lines. A list of replacements is now placed
6874 at the head of all of the source files, except those whose only change is
6875 to replace PERL_COMMAND in the very first #! line.
6877 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6878 messages on the queue, with a bottom-up merge sort, using code contributed
6879 by Michael Haardt. This should make operations like -bp somewhat faster on
6880 large queues. It won't affect queue runners, except when queue_run_in_order
6883 51. Installed eximstats 1.31 in the distribution.
6885 52. Added support for SRV lookups to the dnslookup router.
6887 53. If an ACL referred to $message_body or $message_body_end, the value was not
6888 reset for any messages that followed in the same SMTP session.
6890 54. The store-handling optimization for building very long strings was not
6891 differentiating between the different store pools. I don't think this
6892 actually made any difference in practice, but I've tidied it.
6894 55. While running the routers to verify a sender address, $sender_address
6895 was still set to the sender address. This is wrong, because when routing to
6896 send a bounce to the sender, it would be empty. Therefore, I have changed
6897 it so that, while verifying a sender address, $sender_address is set to <>.
6898 (There is no change to what happens when verifying a recipient address.)
6900 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6901 target A or AAAA records (if not already returned) without resetting the
6902 qualify_single or search_parents options of the DNS resolver. These are
6903 inappropriate in this case because the targets of MX and SRV records must
6904 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6905 target that, when qualified, matched something in the local domain. These
6906 two options are now turned off when doing these lookups.
6908 57. It seems that at least some releases of Reiserfs (which does not have the
6909 concept of a fixed number of inodes) returns zero and not -1 for the
6910 number of available inodes. This interacted badly with check_spool_inodes,
6911 which assumed that -1 was the "no such thing" setting. What I have done is
6912 to check that the total number of inodes is greater than zero before doing
6913 the test of how many are available.
6915 58. When a "warn" ACL statement has a log_message modifier, the message is
6916 remembered, and not repeated. This is to avoid a lot of repetition when a
6917 message has many recipients that cause the same warning to be written.
6918 However, Exim was preserving the list of already written lines for an
6919 entire SMTP session, which doesn't seem right. The memory is now reset if a
6920 new message is started.
6922 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6923 debugging output unless log_rewrite was also set.
6925 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6926 of (int)(handle) when we know that handle contains (void *)(-1).
6928 61. The Exim daemon panic-logs an error return when it closes the incoming
6929 connection. However "connection reset by peer" seems to be common, and
6930 isn't really an error worthy of noting specially, so that particular error
6933 62. When Exim is trying to find all the local interfaces, it used to panic and
6934 die if the ioctl to get the interface flags failed. However, it seems that
6935 on at least one OS (Solaris 9) it is possible to have an interface that is
6936 included in the list of interfaces, but for which you get a failure error
6937 for this call. This happens when the interface is not "plumbed" into a
6938 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6939 failure of the "get flags" call assumes that the interface is down.
6941 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6942 makes life easier for people who are doing arithmetic on fields extracted
6943 from dates, where you often get leading zeros that should not be
6944 interpreted as octal.
6946 64. Added qualify_domain to the redirect router, to override the global
6949 65. If a pathologically long header line contained very many addresses (the
6950 report of this problem mentioned 10 000) and each of them was rewritten,
6951 Exim could use up a very large amount of memory. (It kept on making new
6952 copies of the header line as it rewrote, and never released the old ones.)
6953 At the expense of a bit more processing, the header rewriting function has
6954 been changed so that it no longer eats memory in this way.
6956 66. The generation of the Received: header has been moved from the time that a
6957 message starts to be received, to the time that it finishes. The timestamp
6958 in the Received: header should now be very close to that of the <= log
6959 line. There are two side-effects of this change:
6961 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6962 logged header lines no longer include the local Received: line, because
6963 it has not yet been created. The same applies to a copy of the message
6964 that is returned to a non-SMTP sender when a message is rejected.
6966 (b) When a filter file is tested using -bf, no additional Received: header
6967 is added to the test message. After some thought, I decided that this
6970 This change does not affect the value of $received_for. It is still set
6971 after address rewriting, but before local_scan() is called.
6973 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6975 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6976 gave an unhelpful panic error message, and a defer error. I have managed to
6977 change this behaviour so that it now rejects any supplied certificate,
6978 which seems right, as the list of acceptable certificates is empty.
6980 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6981 gave an unhelpful defer error. I have not managed to make this reject any
6982 supplied certificates, but the error message it gives is "no certificate
6983 supplied", which is not helpful.
6985 70. exigrep's output now also includes lines that are not associated with any
6986 message, but which match the given pattern. Implemented by a patch from
6987 Martin Sluka, which also tidied up the Perl a bit.
6989 71. Recipient callout verification, like sender verification, was using <> in
6990 the MAIL FROM command. This isn't really the right thing, since the actual
6991 sender may affect whether the remote host accepts the recipient or not. I
6992 have changed it to use the actual sender in the callout; this means that
6993 the cache record is now keyed on a recipient/sender pair, not just the
6994 recipient address. There doesn't seem to be a real danger of callout loops,
6995 since a callout by the remote host to check the sender would use <>.
6996 [SEE ABOVE: changed after hitting problems.]
6998 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6999 temporary errors. However, in the case of such a code being given after
7000 the end of a data transmission (i.e. after ".") Exim was failing to write
7001 a retry record for the message. (Yes, there was some broken host that was
7002 actually sending 8xx at this point.)
7004 73. An unknown lookup type in a host list could cause Exim to panic-die when
7005 the list was checked. (An example that provoked this was putting <; in the
7006 middle of a list instead of at the start.) If this happened during a DATA
7007 ACL check, a -D file could be left lying around. This kind of configuration
7008 error no longer causes Exim to die; instead it causes a defer error. The
7009 incident is still logged to the main and panic logs.
7011 74. Buglet left over from Exim 3 conversion. The message "too many messages
7012 in one connection" was written to the rejectlog but not the mainlog, except
7013 when address rewriting (yes!) was being logged.
7015 75. Added write_rejectlog option.
7017 76. When a system filter was run not as root (that is, when system_filter_user
7018 was set), the values of the $n variables were not being returned to the
7019 main process; thus, they were not subsequently available in the $sn
7022 77. Added +return_path_on_delivery log selector.
7024 78. A connection timeout was being treated differently from recipients deferred
7025 when testing hosts_max_try with a message that was older than the host's
7026 retry timeout. (The host should not be counted, thus allowing all hosts to
7027 be tried at least once before bouncing.) This may have been the cause of an
7028 occasionally reported bug whereby a message would remain on the queue
7029 longer than the retry timeout, but would be bounced if a delivery was
7030 forced. I say "may" because I never totally pinned down the problem;
7031 setting up timeout/retry tests is difficult. See also the next item.
7033 79. The ultimate address timeout was not being applied to errors that involved
7034 a combination of host plus message (for example, a timeout on a MAIL
7035 command). When an address resolved to a number of possible hosts, and they
7036 were not all tried for each delivery (e.g. because of hosts_max_try), a
7037 message could remain on the queue longer than the retry timeout.
7039 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
7042 81. Fixed an obscure SMTP outgoing bug which required at least the following
7043 conditions: (a) there was another message waiting for the same server;
7044 (b) the server returned 5xx to all RCPT commands in the first message so
7045 that the message was not completed; (c) the server dropped the connection
7046 or gave a negative response to the RSET that Exim sends to abort the
7047 transaction. The observed case was a dropped connection after DATA that had
7048 been sent in pipelining mode. That is, the server had advertised PIPELINING
7049 but was not implementing it correctly. The effect of the bug was incorrect
7050 behaviour, such as trying another host, and this could lead to a crash.
7056 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
7057 and daemon.c were passed as pointers to ints; they should have been
7058 pointers to socklen_t variables (which are typically unsigned ints).
7060 2. Some signed/unsigned type warnings in the os.c file for Linux have been
7063 3. Fixed a really odd bug that affected only the testing scheme; patching a
7064 certain fixed string in the binary changed the value of another string that
7065 happened to be identical to the end of the original first string.
7067 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
7068 name", it returns that address as the IP address. On some operating
7069 systems (e.g. Solaris), it also passes back the IP address string as the
7070 "host name". However, on others (e.g. Linux), it passes back an empty
7071 string. Exim wasn't checking for this, and was changing the host name to an
7072 empty string, assuming it had been canonicalized.
7074 5. Although rare, it is permitted to have more than one PTR record for a given
7075 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
7076 all the names associated with an address, because they do in Solaris.
7077 However, it seems that they do not in Linux for data that comes from the
7078 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
7079 I found this out when I moved to a new Linux workstation and tried to run
7080 the Exim test suite.
7082 To get round this problem I have changed the code so that it now does its
7083 own call to the DNS to look up PTR records when searching for a host name.
7084 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
7085 addresses that are only in /etc/hosts are still found.
7087 This behaviour is, however, controlled by an option called host_lookup_
7088 order, which defaults to "bydns:byaddr". If people want to use the other
7089 order, or indeed, just use one or the other means of lookup, they can
7090 specify it in this variable.
7092 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
7093 some operating systems, this comes back from gethostbyaddr() as an empty
7094 string, and this is what Exim used to test for. However, it seems that in
7095 other systems, "." is yielded. Exim now tests for this case too.
7097 7. The values of check_spool_space and check_log_space are now held internally
7098 as a number of kilobytes instead of an absolute number of bytes. If a
7099 numbers is specified without 'K' or 'M', it is rounded up to the nearest
7100 kilobyte. This means that much larger values can be stored.
7102 8. Exim monitor: an attempt to get the action menu when not actually pointing
7103 at a message produces an empty menu entitled "No message selected". This
7104 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
7105 no entries in it ("Shell widget menu has zero width and/or height"). So I
7106 have added a single, blank menu entry in this case.
7108 9. Added ${quote_local_part.
7110 10. MIME decoding is now applied to the contents of Subject: header lines when
7113 11. Now that a reference to $sender_host_address automatically causes a reverse
7114 lookup to occur if necessary (4.13/18), there is no need to arrange for a
7115 host lookup before query-style lookups in lists that might use this
7116 variable. This has therefore been abolished, and the "net-" prefix is no
7117 longer necessary for query-style lookups.
7119 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
7120 have been a typo for LFLAGS, so it has been changed.
7122 13. The install script calls Exim with "-C /dev/null" in order to find the
7123 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
7124 to be output. However, since Exim outputs its version number before the
7125 error, it didn't break the script. It just looked ugly. I fixed this by
7126 always allowing "-C /dev/null" if the caller is root.
7128 14. Ignore overlarge ACL variable number when reading spool file - insurance
7129 against a later release with more variables having written the file.
7131 15. The standard form for an IPv6 address literal was being rejected by EHLO.
7132 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
7133 this, as well as the form without the "IPv6" on the front.
7135 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
7136 OS/Makefile-Darwin file.
7138 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
7139 only to LDAP libraries that do not have LDAP_OPT_DEREF.
7141 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
7142 variable. However, some OS use ints for inodes. I've added cast to long int
7143 to get rid of the compiler warning.
7145 19. I had forgotten to lock out "/../" in configuration file names when
7146 ALT_CONFIG_PREFIX was set.
7148 20. Routers used for verification do not need to specify transports. However,
7149 if such a router generated a host list, and callout was configured, Exim
7150 crashed, because it could not find a port number from the (non-existent)
7151 transport. It now assumes port 25 in this circumstance.
7153 21. Added the -t option to exigrep.
7155 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
7156 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
7157 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
7158 them is set, LOOKUP_LSEARCH is forced.
7160 23. "exim -bV" now outputs a list of lookups that are included in the binary.
7162 24. Added sender and host information to the "rejected by local_scan()" log
7163 line; previously there was no indication of these.
7165 25. Added .include_if_exists.
7167 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
7168 Linux. It turns out that not all file systems support this. Apparently some
7169 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
7170 do it.) To cope with this, the error EINVAL, which means that sync-ing is
7171 not supported on the file descriptor, is now ignored when Exim is trying to
7172 sync a directory. This applies only to Linux.
7174 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
7176 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
7177 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
7180 29. Added support for maildirsize files from supplied patch (modified a bit).
7182 30. The use of :fail: followed by an empty string could lead Exim to respond to
7183 sender verification failures with (e.g.):
7185 550 Verification failed for <xxx>
7186 550 Sender verify failed
7188 where the first response line was missing the '-' that indicates it is not
7189 the final line of the response.
7191 31. The loop for finding the name of the user that called Exim had a hardwired
7192 limit of 10; it now uses the value of finduser_retries, which is used for
7193 all other user lookups.
7195 32. Added $received_count variable, available in data and not_smtp ACLs, and at
7198 33. Exim was neglecting to zero errno before one call of strtol() when
7199 expanding a string and expecting an integer value. On some systems this
7200 resulted in spurious "integer overflow" errors. Also, it was casting the
7201 result into an int without checking.
7203 34. Testing for a connection timeout using "timeout_connect" in the retry rules
7204 did not work. The code looks as if it has *never* worked, though it appears
7205 to have been documented since at least release 1.62. I have made it work.
7207 35. The "timeout_DNS" error in retry rules, also documented since at least
7208 1.62, also never worked. As it isn't clear exactly what this means, and
7209 clearly it isn't a major issue, I have abolished the feature by treating it
7210 as "timeout", and writing a warning to the main and panic logs.
7212 36. The display of retry rules for -brt wasn't always showing the error code
7215 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
7216 timeout_connect_A, timeout_connect_MX.
7218 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
7219 to the empty sender.
7221 39. The daemon was not analysing the content of -oX till after it had closed
7222 stderr and disconnected from the controlling terminal. This meant that any
7223 syntax errors were only noted on the panic log, and the return code from
7224 the command was 0. By re-arranging the code a little, I've made the
7225 decoding happen first, so such errors now appear on stderr, and the return
7226 code is 1. However, the actual setting up of the sockets still happens in
7227 the disconnected process, so errors there are still only recorded on the
7230 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
7231 connections (as happens on some IP stacks) was logged at start up time as
7232 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
7233 it from "IPv6 and IPv4", which means that two separate sockets are being
7236 41. The debug output for gethostbyname2() or getipnodebyname() failures now
7237 says whether AF_INET or AF_INET6 was passed as an argument.
7239 42. Exiwhat output was messed up when time zones were included in log
7242 43. Exiwhat now gives more information about the daemon's listening ports,
7243 and whether -tls-on-connect was used.
7245 44. The "port" option of the smtp transport is now expanded.
7247 45. A "message" modifier in a "warn" statement in a non-message ACL was being
7248 silently ignored. Now an error message is written to the main and panic
7251 46. There's a new ACL modifier called "logwrite" which writes to a log file
7252 as soon as it is encountered.
7254 47. Added $local_user_uid and $local_user_gid at routing time.
7256 48. Exim crashed when trying to verify a sender address that was being
7259 49. Exim was recognizing only a space character after ".include". It now also
7260 recognizes a tab character.
7262 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
7263 extracting the relevant information from the specification. The man page no
7264 longer contains scrambled data for the -d option, and I've added a section
7265 at the front about calling Exim under different names.
7267 51. Added "extra_headers" argument to the "mail" command in filter files.
7269 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
7272 53. Installed eximstats 1.29.
7274 54. Added transport_filter_timeout as a generic transport option.
7276 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
7277 Cc: header lines. This was required by RFC 822, but it not required by RFC
7280 56. Exim used to add From:, Date:, and Message-Id: header lines to any
7281 incoming messages that did not have them. Now it does so only if the
7282 message originates locally, that is, if there is no associated remote host
7283 address. When Resent- header lines are present, this applies to the Resent-
7284 lines rather than the non-Resent- lines.
7286 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
7287 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
7289 58. Messages for configuration errors now include the name of the main
7290 configuration file - useful now that there may be more than one file in a
7291 list (.included file names were always shown).
7293 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
7294 for those rare installations that do not start the daemon as root or run it
7295 setuid root. I've cut out the call to initgroups() if the daemon is not
7298 60. The Exim user and group can now be bound into the binary as text strings
7299 that are looked up at the start of Exim's processing.
7301 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
7303 62. Added $mailstore_basename variable.
7305 63. Installed patch to sieve.c from Michael Haardt.
7307 64. When Exim failed to open the panic log after failing to open the main log,
7308 the original message it was trying to log was written to stderr and debug
7309 output, but if they were not available (the usual case in production), it
7310 was lost. Now it is written to syslog before the two lines that record the
7311 failures to open the logs.
7313 65. Users' Exim filters run in subprocesses under the user's uid. It is
7314 possible for a "deliver" command or an alias in a "personal" command to
7315 provoke an address rewrite. If logging of address rewriting is configured,
7316 this fails because the process is not running as root or exim. There may be
7317 a better way of dealing with this, but for the moment (because 4.30 needs
7318 to be released), I have disabled address rewrite logging when running a
7319 filter in a non-root, non-exim process.
7325 1. The buildconfig auxiliary program wasn't quoting the value set for
7326 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
7327 not defined. This bug was masked in 4.22 by the effect that was fixed in
7330 2. Some messages that were rejected after a message id was allocated were
7331 shown as "incomplete" by exigrep. It no longer does this for messages that
7332 are rejected by local_scan() or the DATA or non-SMTP ACLs.
7334 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
7335 have allow_domain_literals set, the ID did not get logged in the <= line.
7336 Domain literals are now always recognized in Message-ID: header lines.
7338 4. The first argument for a ${extract expansion item is the key name or field
7339 number. Leading and trailing spaces in this item were not being ignored,
7340 causing some misleading effects.
7342 5. When deliver_drop_privilege was set, single queue runner processes started
7343 manually (i.e. by the command "exim -q") or by the daemon (which uses the
7344 same command in the process it spins off) were not dropping privilege.
7346 6. When the daemon running as "exim" started a queue runner, it always
7347 re-executed Exim in the spun-off process. This is a waste of effort when
7348 deliver_drop_privilege is set. The new process now just calls the
7349 queue-runner function directly.
7355 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
7358 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
7359 ignored. Though the use of -oP was forcing the writing of a pid file, it
7360 was always written to the default place.
7362 3. If the message "no IP address found for host xxxx" is generated during
7363 incoming verification, it is now followed by identification of the incoming
7364 connection (so you can more easily find what provoked it).
7366 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
7368 5. Added some features to "harden" Exim a bit more against certain attacks:
7370 (a) There is now a build-time option called FIXED_NEVER_USERS that can
7371 be put in Local/Makefile. This is like the never_users runtime option,
7372 but it cannot be overridden. The default setting is "root".
7374 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
7375 prefix string with which any file named in a -C command line option
7378 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
7379 is retained for -C and -D only if the caller of Exim is root. Without
7380 it, the exim user may also use -C and -D and retain privilege.
7382 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
7383 command line option is disabled.
7385 6. Macro names set by the -D option must start with an upper case letter, just
7386 like macro names defined in the configuration file.
7388 7. Added "dereference=" facility to LDAP.
7390 8. Two instances of the typo "uknown" in the source files are fixed.
7392 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
7393 the Configure-Makefile script screwed up while processing it.
7395 10. Incorporated PCRE 4.4.
7397 11. The SMTP synchronization check was not operating right at the start of an
7398 SMTP session. For example, it could not catch a HELO sent before the client
7399 waited for the greeting. There is now a check for outstanding input at the
7400 point when the greeting is written. Because of the duplex, asynchronous
7401 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
7402 way, but not yet received, when the check is performed.
7404 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
7405 on TCP/IP sockets, because this apparently causes some broken clients to
7408 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
7409 unchanged) from the Cygwin maintainer.
7411 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
7412 was supported instead of testing for mbx. Effectively a typo.
7414 15. The spa authenticator server code was not checking that the input it
7415 received was valid base64.
7417 16. The debug output line for the "set" modifier in ACLs was not showing the
7418 name of the variable that was being set.
7420 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
7422 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
7423 lookup on its own. Something else was needed to trigger the lookup. For
7424 example, a match in host_lookup or the need for a host name in a host list.
7425 Now, if $sender_host_name is referenced and the host name has not yet been
7426 looked up, a lookup is performed. If the lookup fails, the variable remains
7427 empty, and $host_lookup_failed is set to "1".
7429 19. Added "eqi" as a case-independent comparison operator.
7431 20. The saslauthd authentication condition could segfault if neither service
7432 nor realm was specified.
7434 21. If an overflowing value such as "2048M" was set for message_size_limit, the
7435 error message that was logged was misleading, and incoming SMTP
7436 connections were dropped. The message is now more accurate, and temporary
7437 errors are given to SMTP connections.
7439 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
7440 (except RSET) with a 421 error, until QUIT is received. However, it was
7441 failing to send a response to QUIT.
7443 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
7444 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
7445 ACL is now run after the helo_try_verify_hosts code.
7447 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
7448 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
7449 case-independent, so other case variants are also recognized). Apparently
7450 some systems use these upper case variants.
7452 25. If more than two messages were waiting for the same host, and a transport
7453 filter was specified for the transport, Exim sent two messages over the
7454 same TCP/IP connection, and then failed with "socket operation on non-
7455 socket" when it tried to send the third.
7457 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
7459 27. The extern definition of crypt16() in expand.c was not being excluded when
7460 the OS had its own crypt16() function.
7462 28. Added bounce_return_body as a new option, and bounce_return_size_limit
7463 as a preferred synonym for return_size_limit, both as an option and as an
7466 29. Added LIBS=-liconv to OS/Makefile-OSF1.
7468 30. Changed the default configuration ACL to relax the local part checking rule
7469 for addresses that are not in any local domains. For these addresses,
7470 slashes and pipe symbols are allowed within local parts, but the sequence
7471 /../ is explicitly forbidden.
7473 31. SPA server authentication was not clearing the challenge buffer before
7476 32. log_message in a "warn" ACL statement was writing to the reject log as
7477 well as to the main log, which contradicts the documentation and doesn't
7478 seem right (because no rejection is happening). So I have stopped it.
7480 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
7481 However, I am unable to do any testing of this.
7483 34. Fixed an infelicity in the appendfile transport. When checking directories
7484 for a mailbox, to see if any needed to be created, it was accidentally
7485 using path names with one or more superfluous leading slashes; tracing
7486 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
7488 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
7489 contents are added to the log line that is written for every discarded
7490 recipient. (Previously a log_message setting was ignored.)
7492 36. The ${quote: operator now quotes the string if it is empty.
7494 37. The install script runs exim in order to find its version number. If for
7495 some reason other than non-existence or emptiness, which it checks, it
7496 could not run './exim', it was installing it with an empty version number,
7497 i.e. as "exim-". This error state is now caught, and the installation is
7500 38. An argument was missing from the function that creates an error message
7501 when Exim fails to connect to the socket for saslauthd authentication.
7502 This could cause Exim to crash, or give a corrupted message.
7504 39. Added isip, isip4, and isip6 to ${if conditions.
7506 40. The ACL variables $acl_xx are now saved with the message, and can be
7507 accessed later in routers, transports, and filters.
7509 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
7510 strings in the file are not string-expanded.
7512 42. If a MAIL command specified a SIZE value that was too large to fit into an
7513 int variable, the check against message_size_limit failed. Such values are
7514 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
7515 day this will have to be increased, but I don't think I want to be around
7516 when emails are that large.
7523 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
7524 iconv() is not standard in FreeBSD.
7526 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
7527 IPv6 enabled. The observed symptom was a segmentation fault on return from
7528 the function os_common_find_running_interfaces() in src/os.c.
7530 3. In the check_special_case() function in daemon.c I had used "errno" as an
7531 argument name, which causes warnings on some systems. This was basically a
7532 typo, since it was named "eno" in the comments!
7534 4. The code that waits for the clock to tick (at a resolution of some fraction
7535 of a second) so as to ensure message-id uniqueness was always waiting for
7536 at least one whole tick, when it could have waited for less. [This is
7537 almost certainly not relevant at current processor speeds, where it is
7538 unlikely to ever wait at all. But we try to future-proof.]
7540 5. The function that sleeps for a time interval that includes fractions of a
7541 second contained a race. It did not block SIGALRM between setting the
7542 timer, and suspending (a couple of lines later). If the interval was short
7543 and the sigsuspend() was delayed until after it had expired, the suspension
7544 never ended. On busy systems this could lead to processes getting stuck for
7547 6. Some uncommon configurations may cause a lookup to happen in a queue runner
7548 process, before it forks any delivery processes. The open lookup caching
7549 mechanism meant that the open file or database connection was passed into
7550 the delivery process. The problem was that delivery processes always tidy
7551 up cached lookup data. This could cause a problem for the next delivery
7552 process started by the queue runner, because the external queue runner
7553 process does not know about the closure. So the next delivery process
7554 still has data in the lookup cache. In the case of a file lookup, there was
7555 no problem because closing a file descriptor in a subprocess doesn't affect
7556 the parent. However, if the lookup was caching a connection to a database,
7557 the connection was closed, and the second delivery process was likely to
7558 see errors such as "PGSQL: query failed: server closed the connection
7559 unexpectedly". The problem has been fixed by closing all cached lookups
7560 in a queue runner before running a delivery process.
7562 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
7563 seem to have the "const" qualifier which it has on other OS. I've
7566 8. Change 4.21/2 was too strict. It is only if there are two authenticators
7567 *of the same type* (client or server) with the same public name that an
7568 error should be diagnosed.
7570 9. When Exim looked up a host name for an IP address, but failed to find the
7571 original IP address when looking up the host name (a safety check), it
7572 output the message "<ip address> does not match any IP for NULL", which was
7573 confusing, to say the least. The bug was that the host name should have
7574 appeared instead of "NULL".
7576 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
7577 user that is built into the binary, and the -C or -D options is used, root
7578 privilege is dropped before the configuration file is read. In addition,
7579 logging is switched to stderr instead of the normal log files. If the
7580 configuration then re-defines the Exim user, the unprivileged environment
7581 is probably not what is expected, so Exim logs a panic warning message (but
7584 However, if deliver_drop_privilege is set, the unprivileged state may well
7585 be exactly what is intended, so the warning has been cut out in that case,
7586 and Exim is allowed to try to write to its normal log files.
7592 1. smtp_return_error_details was not giving details for temporary sender
7593 or receiver verification errors.
7595 2. Diagnose a configuration error if two authenticators have the same public
7598 3. Exim used not to create the message log file for a message until the first
7599 delivery attempt. This could be confusing when incoming messages were held
7600 for policy or load reasons. The message log file is now created at the time
7601 the message is received, and an initial "Received" line is written to it.
7603 4. The automatically generated man page for command line options had a minor
7604 bug that caused no ill effects; however, a more serious problem was that
7605 the procedure for building the man page automatically didn't always
7606 operate. Consequently, release 4.20 contains an out-of-date version. This
7607 shouldn't happen again.
7609 5. When building Exim with embedded Perl support, the script that builds the
7610 Makefile was calling 'perl' to find its compile-time parameters, ignoring
7611 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
7613 6. The freeze_tell option was not being used for messages that were frozen on
7614 arrival, either by an ACL or by local_scan().
7616 7. Added the smtp_incomplete_transaction log selector.
7618 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
7619 was accepting AUTH without a new EHLO.
7621 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
7622 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
7625 10. Logging of TCP/IP connections (when configured) now happens in the main
7626 daemon process instead of the child process, so that the TCP/IP connection
7627 count is more accurate (but it can never be perfect).
7629 11. The use of "drop" in a nested ACL was not being handled correctly in the
7630 outer ACL. Now, if condition failure induced by the nested "drop" causes
7631 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
7632 or "require"), the connection is dropped.
7634 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
7635 that yield "discard" can now be used with an "accept" or a "discard" verb,
7636 but an error is generated for any others (because I can't see a useful way
7637 to define what should happen).
7639 13. When an ACL is read dynamically from a file (or anywhere else), the lines
7640 are now processed in the same way as lines in the Exim configuration file.
7641 In particular, continuation lines are supported.
7643 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
7645 15. Added -ti meaning -t -i.
7647 16. Check for letters, digits, hyphens, and dots in the names of dnslist
7648 domains, and warn by logging if others are found.
7650 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
7651 returned from GIFCONF when Exim is trying to find the list of interfaces on
7652 a host. The code in os.c has been modified to copy each ifreq to an aligned
7653 structure in all cases.
7655 Also, in some cases, the returned ifreq's were being copied to a 'struct
7656 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
7657 means the last couple of bytes of an IPv6 address could be chopped if the
7658 ifreq contained only a normal sockaddr (14 bytes storage).
7660 18. Named domain lists were not supported in the hosts_treat_as_local option.
7661 An entry such as +xxxx was not recognized, and was treated as a literal
7664 19. Ensure that header lines added by a DATA ACL are included in the reject log
7665 if the ACL subsequently rejects the message.
7667 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
7668 MD5 (which is deprecated).
7670 21. When testing a filter file using -bf, Exim was writing a message when it
7671 took the sender from a "From " line in the message, but it was not doing so
7672 when it took $return_path from a Return-Path: header line. It now does.
7674 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
7675 with a valid header line field name (a series of printing characters
7676 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
7678 23. Changed "disc" in the source to "disk" to conform to the documentation and
7679 the book and for uniformity.
7681 24. Ignore Sendmail's -Ooption=value command line item.
7683 25. When execve() failed while trying to run a command in a pipe transport,
7684 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
7685 could be confused with a return value of 69 from the command itself. This
7686 has been changed to 127, the value the shell returns if it is asked to run
7687 a non-existent command. The wording for the related log line suggests a
7688 non-existent command as the problem.
7690 26. If received_header_text expands to an empty string, do not add a Received:
7691 header line to the message. (Well, it adds a token one on the spool, but
7692 marks it "old" so that it doesn't get used or transmitted.)
7694 27. Installed eximstats 1.28 (addition of -nt option).
7696 28. There was no check for failure on the call to getsockname() in the daemon
7697 code. This can fail if there is a shortage of resources on the system, with
7698 ENOMEM, for example. A temporary error is now given on failure.
7700 29. Contrary to the C standard, it seems that in some environments, the
7701 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
7702 program. Exim now does this explicitly; it affects the formatting of
7703 timestamps using strftime().
7705 30. If exiqsumm was given junk data, it threw up some uninitialized variable
7706 complaints. I've now initialized all the variables, to avoid this.
7708 32. Header lines added by a system filter were not being "seen" during
7709 transport-time rewrites.
7711 33. The info_callback() function passed to OpenSSL is set up with type void
7712 (*)(SSL *, int, int), as described somewhere. However, when calling the
7713 function (actually a macro) that sets it up, the type void(*)() is
7714 expected. I've put in a cast to prevent warnings from picky compilers.
7716 34. If a DNS black list lookup found a CNAME record, but there were no A
7717 records associated with the domain it pointed at, Exim crashed.
7719 35. If a DNS black list lookup returned more than one A record, Exim ignored
7720 all but the first. It now scans all returned addresses if a particular IP
7721 value is being sought. In this situation, the contents of the
7722 $dnslist_value variable are a list of all the addresses, separated by a
7725 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
7726 to accept a host name and all its aliases if the forward lookup for any of
7727 them yielded the IP address of the incoming connection. Now it accepts only
7728 those names whose forward lookup yields the correct IP address. Any other
7729 names are discarded. This closes a loophole whereby a rogue DNS
7730 administrator could create reverse DNS records to break through a
7731 wildcarded host restriction in an ACL.
7733 37. If a user filter or a system filter that ran in a subprocess used any of
7734 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
7735 the wrong values were passed to the pipe command ($thisaddress had the
7736 value of $0, $0 had the value of $1, etc). This bug was introduced by
7737 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
7739 38. Improved the line breaking for long SMTP error messages from ACLs.
7740 Previously, if there was no break point between 40 and 75 characters, Exim
7741 left the rest of the message alone. Two changes have been made: (a) I've
7742 reduced the minimum length to 35 characters; (b) if it can't find a break
7743 point between 35 and 75 characters, it looks ahead and uses the first one
7744 that it finds. This may give the occasional overlong line, but at least the
7745 remaining text gets split now.
7747 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
7748 file descriptors might be low, and that setting 1000 would always raise it.
7749 It turns out that in some environments, the limit is already over 1000 and
7750 that lowering it causes trouble. So now Exim takes care not to decrease it.
7752 40. When delivering a message, the value of $return_path is set to $sender_
7753 address at the start of routing (routers may change the value). By an
7754 oversight, this default was not being set up when an address was tested by
7755 -bt or -bv, which affected the outcome if any router or filter referred to
7758 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
7759 log only when "message" or "log_message" are set. However, if one of the
7760 conditions was an address verification, or a call to a nested ACL, the
7761 messages generated by the underlying test were being passed through. This
7762 no longer happens. The underlying message is available in $acl_verify_
7763 message for both "message" and "log_message" expansions, so it can be
7764 passed through if needed.
7766 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
7767 new expansion $bh_ to give the encoded byte string without charset
7768 translation. Translation happens only if iconv() is available; HAVE_ICONV
7769 indicates this at build time. HEADERS_CHARSET gives the charset to
7770 translate to; headers_charset can change it in the configuration, and
7771 "headers charset" can change it in an individual filter file.
7773 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
7774 that creates RFC 2047 encoded "words" labels them as that charset instead
7775 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
7776 expansion operator; (ii) when Exim creates a From: line for a local
7777 message; (iii) when a header line is rewritten to include a "phrase" part.
7779 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
7780 buggy, causing it to skip the first lines of messages whose message ID
7781 ended in 'D'. This would not have bitten before Exim release 4.14, because
7782 message IDs were unlikely to end in 'D' before then. The effect was to have
7783 incorrect size information for certain domains.
7785 45. #include "config.h" was missing at the start of the crypt16.c module. This
7786 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
7789 46. If there was a timeout during a "random" callout check, Exim treated it as
7790 a failure of the random address, and carried on sending RSET and the real
7791 address. If the delay was just some slowness somewhere, the response to the
7792 original RCPT would be taken as a response to RSET and so on, causing
7793 mayhem of various kinds.
7795 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
7796 when I implemented it. It didn't allow for the fact that some option values
7797 may legitimately be negative (e.g. size_addition), and it didn't even do
7798 the right test for positive values.
7800 48. Domain names in DNS records are case-independent. Exim always looks them up
7801 in lower case. Some resolvers return domain names in exactly the case they
7802 appear in the zone file, that is, they may contain uppercase letters. Not
7803 all resolvers do this - some return always lower case. Exim was treating a
7804 change of case by a resolver as a change of domain, similar to a widening
7805 of a domain abbreviation. This triggered its re-routing code and so it was
7806 trying to route what was effectively the same domain again. This normally
7807 caused routing to fail (because the router wouldn't handle the domain
7808 twice). Now Exim checks for this case specially, and just changes the
7809 casing of the domain that it ultimately uses when it transmits the message
7812 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
7815 50. If a filter generated a file delivery with a non-absolute name (possible if
7816 no home directory exists for the router), the forbid_file option was not
7819 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
7820 the existing equality matching.
7822 52. Exim was using ints instead of ino_t variables in some places where it was
7823 dealing with inode numbers.
7825 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
7826 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
7827 called TMPDIR, and if it finds it is different, it changes its value.
7829 54. The smtp_printf() function is now made available to local_scan() so
7830 additional output lines can be written before returning. There is also an
7831 smtp_fflush() function to enable the detection of a dropped connection.
7832 The variables smtp_input and smtp_batched_input are exported to
7835 55. Changed the default runtime configuration: the message "Unknown user"
7836 has been removed from the ACL, and instead placed on the localuser router,
7837 using the cannot_route_message feature. This means that any verification
7838 failures that generate their own messages won't get overridden. Similarly,
7839 the "Unrouteable address" message that was in the ACL for unverifiable
7840 relay addresses has also been removed.
7842 56. Added hosts_avoid_esmtp to the smtp transport.
7844 57. The exicyclog script was not checking for the esoteric option
7845 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7846 will work only if exicyclog is run under the appropriate euid.
7848 58. Following a discussion on the list, the rules by which Exim recognises line
7849 endings on incoming messages have been changed. The -dropcr and drop_cr
7850 options are now no-ops, retained only for backwards compatibility. The
7851 following line terminators are recognized: LF CRLF CR. However, special
7852 processing applies to CR:
7854 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7855 nor a local message in the state where . is a terminator.
7857 (ii) If a bare CR is encountered in a header line, an extra space is added
7858 after the line terminator so as not to end the header. The reasoning
7859 behind this is that bare CRs in header lines are most likely either
7860 to be mistakes, or people trying to play silly games.
7862 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7863 was being incorrectly given as 18 bytes larger than it should have been.
7864 This is a VOB (very old bug).
7866 60. This may never have affected anything current, but just in case it has:
7867 When the local host is found other than at the start of a list of hosts,
7868 the local host, those with the same MX, and any that follow, are discarded.
7869 When the list in question was part of a longer list of hosts, the following
7870 hosts (not currently being processed) were also being discarded. This no
7871 longer happens. I'm not sure if this situation could ever has previously
7874 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7877 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7878 header. This is something that is recommended in a new Internet Draft, and
7879 is something that is documented as being done by Sendmail. There are two
7880 possible values. For messages generated by the autoreply transport, Exim
7883 Auto-Submitted: auto-replied
7885 whereas for all other generated messages (e.g. bounces) it adds
7887 Auto-Submitted: auto-generated
7889 63. The "personal" condition in filters now includes a test for the
7890 Auto-Submitted: header. If it contains the string "auto-" the message it
7891 not considered personal.
7893 64. Added rcpt_include_affixes as a generic transport option.
7895 65. Added queue_only_override (default true).
7897 66. Added the syslog_duplication option.
7899 67. If what should have been the first header line of a message consisted of
7900 a space followed by a colon, Exim was mis-interpreting it as a header line.
7901 It isn't of course - it is syntactically invalid and should therefore be
7902 treated as the start of the message body. The misbehaviour could have
7903 caused a number of strange effects, including loss of data in subsequent
7904 header lines, and spool format errors.
7906 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7907 client host had authenticated. This control can now be exercised by an ACL
7908 for more flexibility.
7910 69. By default, callouts do not happen when testing with -bh. There is now a
7911 variant, -bhc, which does actually run the callout code, including
7912 consulting and updating the callout cache.
7914 70. Added support for saslauthd authentication, courtesy of Alexander
7917 71. If statvfs() failed on the spool or log directories while checking their
7918 size for availability, Exim confusingly gave the error "space shortage".
7919 Furthermore, in debugging mode it crashed with a floating point exception.
7920 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7921 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7922 serious problem, Exim now writes to the main and panic logs when this
7923 happens, with details of the failure. It then refuses to accept the
7924 incoming message, giving the message "spool directory problem" or "log
7925 directory problem" with a 421 code for SMTP messages.
7927 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7928 0, 1, and 2 exist, because some OS complain for execs without them (see
7929 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7930 descriptors. However, the code omitted to check that the open succeeded,
7931 causing mysterious errors if for some reason the permissions on /dev/null
7932 got screwed. Now Exim writes a message to the main and panic logs, and
7933 bombs out if it can't open /dev/null.
7935 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7936 interact so that it is all more flexible. It is supposed to remain
7937 backwards compatible. Also added extra_local_interfaces.
7939 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7940 to bomb out with an assertion failure - to the client this appears as a
7941 connection drop. This problem occurs in the part of the code that was taken
7942 from the Samba project. Fortunately, the assertion is in a very simple
7943 function, so I have fixed this by reproducing the function inline in the
7944 one place where it is called, and arranging for authentication to fail
7945 instead of killing the process with assert().
7947 75. The SPA client code was not working when the server requested OEM rather
7948 than Unicode encoding.
7950 76. Added code to make require_files with a specific uid setting more usable in
7951 the case where statting the file as root fails - usually a non-root-mounted
7952 NFS file system. When this happens and the failure is EACCES, Exim now
7953 forks a subprocess and does the per-uid checking as the relevant uid.
7955 77. Added process_log_path.
7957 78. If log_file_path was not explicitly set, a setting of check_log_space or
7958 check_log_inodes was ignored.
7960 79. If a space check for the spool or log partitions fails, the incident is now
7961 logged. Of course, in the latter case the data may get lost...
7963 80. Added the %p formatting code to string_format() so that it can be used to
7964 print addresses in debug_print(). Adjusted all the address printing in the
7965 debugging in store.c to use %p rather than %d.
7967 81. There was a concern that a line of code in smtp_in.c could overflow a
7968 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7969 initially expressed, the concern was not well-founded, because trailing
7970 spaces are removed early. However, if the trailing spaces were followed by
7971 a NULL, they did not get removed, so the overflow was possible. Two fixes
7974 (a) I re-wrote the offending code in a cleaner fashion.
7975 (b) If an incoming SMTP command contains a NULL character, it is rejected
7978 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7979 runs initgroups(), so that if the Exim user is in any additional groups,
7980 they will be used during message reception.
7986 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob