1 This document describes *changes* to previous versions, that might
2 affect Exim's operation, with an unchanged configuration file. For new
3 options, and new features, see the NewStuff file next to this ChangeLog.
9 JH/01 Bug 1329: Fix format of Maildir-format filenames to match other mail-
10 related applications. Previously an "H" was used where available info
11 says that "M" should be, so change to match.
13 JH/02 Bug 2587: Fix pam expansion condition. Tainted values are commonly used
14 as arguments, so an implementation trying to copy these into a local
15 buffer was taking a taint-enforcement trap. Fix by using dynamically
16 created buffers. Similar fix for radius expansion condition.
18 JH/03 Bug 2586: Fix listcount expansion operator. Using tainted arguments is
19 reasonable, eg. to count headers. Fix by using dynamically created
20 buffers rather than a local. Do similar fixes for ACL actions "dcc",
21 "log_reject_target", "malware" and "spam"; the arguments are expanded
22 so could be handling tainted values.
24 JH/04 Bug 2590: Fix -bi (newaliases). A previous code rearrangement had
25 broken the (no-op) support for this sendmail command. Restore it
26 to doing nothing, silently, and returning good status.
28 JH/05 Bug 2593: Fix "vacation" in Exim filter. Previously, when a "once"
29 record path was given (or the default used) without a leading directory
30 path, an error occurred on trying to open it. Use the transport's working
33 JH/06 Bug 2594: Change the name used for certificate name checks in the smtp
34 transport. Previously it was the name on the DNS A-record; use instead
35 the head of the CNAME chain leading there (if there is one). This seems
36 to align better with RFC 6125.
38 JH/07 Bug 2597: Fix a resource leak. Using a lookup in obtaining a value for
39 smtp_accept_max_per_host allocated resources which were not released
40 when the limit was exceeded. This eventually crashed the daemon. Fix
41 by adding a relase action in that path.
43 JH/08 Bug 2598: Fix verify ACL condition. The options for the condition are
44 expanded; previously using tainted values was rejected. Fix by using
45 dynamically-created buffers.
47 JH/09 Relax restrictions on ACL verify condition needing access to message
48 headers. Previously they were only permitted in data and non-smtp ACLs;
49 permit also mime, dkim, prdr quit and notquit. Applies to header-syntax,
50 not_blind, header_sender and header_names_ascii verification.
52 JH/10 Bug 2603: Fix coding of string copying to only evaluate arguments once.
53 Previously a macro used one argument twice; when called with the
54 argument as an expression having side-effects, incorrect operation
55 resulted. Use an inlineable function.
57 JH/11 Bug 2604: Fix request to cutthrough-deliver when a connection is already
58 held open for a verify callout. Previously this wan not accounted for
59 and a corrupt onward SMTP conversation resulted.
61 JH/12 Bug 2607: Fix the ${srs_encode } expansion to handle quoted local_parts.
62 Previously they were embedded naively in the constructed address; when
63 needed, strip the quoting and quote the entire local_part.
64 Also make the inbound_srs expansion condition handle quoting.
66 JH/13 Fix dsearch "subdir" filter to ignore ".". Previously only ".." was
67 excluded, not matching the documentation.
69 JH/14 Bug 2606: Fix a segfault in sqlite lookups. When no, or a bad, filename
70 was given for the sqlite_dbfile a trap resulted.
72 JH/15 Bug 2620: Fix "spam" ACL condition. Previously, tainted values for the
73 "name" argument resulted in a trap. There is no reason to disallow such;
74 this was a coding error.
76 JH/16 Bug 2615: Fix pause during message reception, on systems that have been
77 suspended/resumed. The Linux CLOCK_MONOTONIC does not account for time
78 spent suspended, ignoring the Posix definition. Previously we assumed
79 it did and a constant offset from real time could be used as a correction.
80 Change to using the same clock source for the start-of-message and the
81 post-message next-tick-wait. Also change to using CLOCK_BOOTTIME if it
82 exists, just to get a clock slightly more aligned to reality.
84 JH/17 Bug 2295: Fix DKIM signing to always semicolon-terminate. Although the
85 RFC says it is optional some validators care. The missing char was not
86 intended but triggered by a line-wrap alignement. Discovery and fix by
87 Guillaume Outters, hacked on by JH.
89 JH/18 Bug 2617: Fix a taint trap in parse_fix_phrase(). Previously when the
90 name being quoted was tainted a trap would be taken. Fix by using
91 dynamicaly created buffers. The routine could have been called by a
92 rewrite with the "h" flag, by using the "-F" command-line option, or
93 by using a "name=" option on a control=submission ACL modifier.
95 JH/19 SPF: change the Authentication-Results expansion component to give
96 smtp.helo when the sender domain is empty. Previously it gave
99 JH/20 Bug 2631: ACL dnslist conditions now ignore and log any lookups returns
100 not in 127.0.0.0/8 to help in spotting list domains taken over by a
101 domain-parking registrar.
103 JH/21 Bug 2630: Fix eol-replacement string for the ${readsocket } expansion.
104 Previously when a whitespace character was specified it was not inserted
105 after removing the newline.
107 JH/22 Bug 2265: Force SNI usage for smtp transport DANE'd connections, to be
108 the domain part of the recipient address. This overrides any tls_sni
109 option set, which was previously used.
111 JH/23 Logging: with the +tls_sni log_selector, do not wrap the received SNI
114 JH/24 Bug 2634: Fix a taint trap seen on NetBSD: the testing coded for
115 is_tainted() had an off-by-one error in the overenthusiastic direction.
116 Find and fix by Gavan. Although NetBSD is not a supported platform for
117 4.94 this bug could affect other platforms.
119 PP/01 Fix default prime selection to be consistent.
120 One path used ike23 still, instead of exim.dev.20160529.3; now both
121 execution flows will use the same DH primes (currently
122 exim.dev.20160529.3).
124 JH/25 OpenSSL: Fix back-compatibility behaviour surrounding tls_certificates
125 option in smtp transport, to match the documentation. Previously
126 verification was not being done in some cases where it should have been.
128 JH/26 Bug 2646: fix a memory usage issue in ldap lookups. Previously, when more
129 than one server was defined and depending on the platform memory layout
130 details, an internal consistency trap could be hit while walking the list
133 JH/27 Bug 2648: fix the passing of an authenticator public-name through spool
134 files. The value is used by the authresults expansion item. Previously
135 if this was used in a router or transport, a crash could result.
137 JH/28 Fix spurious logging of select error. Some platforms, notably FreeBSD,
138 have a sufficient incidence of EINTR returns from select that an
139 interaction with other operations done by the main daemon loop exposed
140 a bug in the error-handling. This was benign apart from the log
143 JH/29 Bug 2675: add outgoing-interface I= element to deferred "==" log lines,
144 for consistency with delivered "=>" and failed "**" lines. While we're
145 there, handle PRX and TFO.
147 JH/30 Bug 2677: fix matching of long addresses. Since 4.93 a limit of 256 was
148 applied. This resulted, if any header-line rewrite rules were configured,
149 in a panic-log trigerrable by sending a message with a long address in
150 a header. Fix by increaing the arbitrary limit to larger than a single
151 (dewrapped) 5322 header line maximum size.
153 JH/31 The ESMTP option name advertised for the SUPPORT_EARLY_PIPE build option
154 is changed from X_PIPE_CONNECT to PIPE_CONNECT. This is in line with
155 RFC 6648 which deprecates X- options in protocols as a general practice.
156 Changeover between the implementations is handled by the mechanisms
159 JH/32 Bug 2599: fix delay of delivery to a local address where there is also
160 a remote which uses callout/hold. Previously the local was queued.
162 JH/33 Fix a taint trap in the ${listextract } expansion when the source data
165 JH/34 Fix the placement of a multiple-message delivery marker in the delivery
166 log line. The asterisk is now consistently appended to the remote IP
167 (and port, if given), and will also be provided on defer and fail log
168 lines. Previously it could be placed on the local IP if that was being
169 logged, and was only provided on delivery lines.
171 JH/35 Bug 2343: Harden exim_tidydb against corrupt wait- files.
173 JH/36 Bug 2687: Fix interpretation of multiple ^ chars in a plaintext
174 authenticator client_send option. Previously the next char, after a pair
175 was collapsed, was taken verbatim (so ^^^foo became ^^foo; ^^^^foo became
176 ^^\x00foo). Fixed to get ^\x00foo and ^^foo respectively to match the
177 documentation. There is still no way to get a leading ^ immediately
178 after a NUL (ie. for the password of a PLAIN method authenticator.
180 JH/37 Enforce the expected size, for fixed-size records read from hints-DB
181 files. For bad sizes read, delete the record and whine to paniclog.
183 JH/38 When logging an AUTH failure, as server, do not include sensitive
184 information. Previously, the credentials would be included if given
185 as part of the AUTH command line and an ACL denied authentidcation.
187 JH/39 Bug 2691: fix $local_part_data. When the matching list element
188 referred to a file, bad data was returned. This likely also affected
191 JH/40 The gsasl authenticator now supports caching of the salted password
192 generated by the client-side implementation. This required the addition
193 of a new variable: $auth4.
195 JH/41 Fix daemon SIGHUP on FreeBSD. Previously, a named socket for IPC was
196 left undeleted; the attempt to re-create it then failed - resulting in
197 the usual "SIGHUP tp have daemon reload configuration" to not work.
198 This affected any platform not supporting "abstract" Unix-domain
199 sockets (i.e. not Linux).
201 JH/42 Bug 2693: Harden against a peer which reneges on a 452 "too many
202 recipients" response to RCPT in a later response, with a 250. The
203 previous coding assumed this would not happen, and under PIPELINING
204 would result in both lost and duplicate recipients for a message.
206 JH/43 Bug 2694: Fix weighted distribution of work to multiple spamd servers.
207 Previously the weighting was incorrectly applied. Similar fix for socks
208 proxies. Found and fixed by Heiko Schlichting.
210 JH/44 Bug 2701: Fix list-expansion of dns_ipv4_lookup. Previously, it did
211 not handle sub-lists included using the +namedlist syntax. While
212 investigating, the same found for dns_trust_aa, dns_again_means_nonexist,
213 dnssec_require_domains, dnssec_request_domains, srv_fail_domains,
220 JH/01 Avoid costly startup code when not strictly needed. This reduces time
221 for some exim process initialisations. It does mean that the logging
222 of TLS configuration problems is only done for the daemon startup.
224 JH/02 Early-pipelining support code is now included unless disabled in Makefile.
226 JH/03 DKIM verification defaults no long accept sha1 hashes, to conform to
227 RFC 8301. They can still be enabled, using the dkim_verify_hashes main
230 JH/04 Support CHUNKING from an smtp transport using a transport_filter, when
231 DKIM signing is being done. Previously a transport_filter would always
232 disable CHUNKING, falling back to traditional DATA.
234 JH/05 Regard command-line recipients as tainted.
236 JH/06 Bug 340: Remove the daemon pid file on exit, when due to SIGTERM.
238 JH/07 Bug 2489: Fix crash in the "pam" expansion condition. It seems that the
239 PAM library frees one of the arguments given to it, despite the
240 documentation. Therefore a plain malloc must be used.
242 JH/08 Bug 2491: Use tainted buffers for the transport smtp context. Previously
243 on-stack buffers were used, resulting in a taint trap when DSN information
244 copied from a received message was written into the buffer.
246 JH/09 Bug 2493: Harden ARC verify against Outlook, whick has been seen to mix
247 the ordering of its ARC headers. This caused a crash.
249 JH/10 Bug 2492: Use tainted memory for retry record when needed. Previously when
250 a new record was being constructed with information from the peer, a trap
253 JH/11 Bug 2494: Unset the default for dmarc_tld_file. Previously a naiive
254 installation would get error messages from DMARC verify, when it hit the
255 nonexistent file indicated by the default. Distros wanting DMARC enabled
256 should both provide the file and set the option.
257 Also enforce no DMARC verification for command-line sourced messages.
259 JH/12 Fix an uninitialised flag in early-pipelining. Previously connections
260 could, depending on the platform, hang at the STARTTLS response.
262 JH/13 Bug 2498: Reset a counter used for ARC verify before handling another
263 message on a connection. Previously if one message had ARC headers and
264 the following one did not, a crash could result when adding an
265 Authentication-Results: header.
267 JH/14 Bug 2500: Rewind some of the common-coding in string handling between the
268 Exim main code and Exim-related utities. The introduction of taint
269 tracking also did many adjustments to string handling. Since then, eximon
270 frequently terminated with an assert failure.
272 JH/15 When PIPELINING, synch after every hundred or so RCPT commands sent and
273 check for 452 responses. This slightly helps the inefficieny of doing
274 a large alias-expansion into a recipient-limited target. The max_rcpt
275 transport option still applies (and at the current default, will override
276 the new feature). The check is done for either cause of synch, and forces
277 a fast-retry of all 452'd recipients using a new MAIL FROM on the same
278 connection. The new facility is not tunable at this time.
280 JH/16 Fix the variables set by the gsasl authenticator. Previously a pointer to
281 library live data was being used, so the results became garbage. Make
282 copies while it is still usable.
284 JH/17 Logging: when the deliver_time selector ise set, include the DT= field
285 on delivery deferred (==) and failed (**) lines (if a delivery was
286 attemtped). Previously it was only on completion (=>) lines.
288 JH/18 Authentication: the gsasl driver not provides the $authN variables in time
289 for the expansion of the server_scram_iter and server_scram_salt options.
291 WB/01 SPF: DNS lookups for the obsolete SPF RR type done by the libspf2 library
292 are now specifically given a NO_DATA response without hitting the system
293 resolver. The library goes on to do the now-standard TXT lookup.
294 Use of dnsdb lookups is not affected.
296 JH/19 Bug 2507: Modules: on handling a dynamic-module (lookups) open failure,
297 only retrieve the errormessage once. Previously two calls to dlerror()
298 were used, and the second one (for mainlog/paniclog) retrieved null
301 JH/20 Taint checking: disallow use of tainted data for
302 - the appendfile transport file and directory options
303 - the pipe transport command
304 - the autoreply transport file, log and once options
305 - file names used by the redirect router (including filter files)
307 - paths used by single-key lookups
308 Previously this was permitted.
310 JH/21 Bug 2501: Fix init call in the heimdal authenticator. Previously it
311 adjusted the size of a major service buffer; this failed because the
312 buffer was in use at the time. Change to a compile-time increase in the
313 buffer size, when this authenticator is compiled into exim.
315 JH/22 Taint-checking: move to safe-mode taint checking on all platforms. The
316 previous fast-mode was untenable in the face of glibs using mmap to
317 support larger malloc requests.
319 PP/01 Update the openssl_options possible values through OpenSSL 1.1.1c.
320 New values supported, if defined on system where compiled:
321 allow_no_dhe_kex, cryptopro_tlsext_bug, enable_middlebox_compat,
322 no_anti_replay, no_encrypt_then_mac, prioritize_chacha, tlsext_padding
324 JH/23 Performance improvement in the initial phase of a two-pass queue run. By
325 running a limited number of proceses in parallel, a benefit is gained. The
326 amount varies with the platform hardware and load. The use of the option
327 queue_run_in_order means we cannot do this, as ordering becomes
330 JH/24 Bug 2524: fix the cyrus_sasl auth driver gssapi usage. A previous fix
331 had introduced a string-copy (for ensuring NUL-termination) which was not
332 appropriate for that case, which can include embedded NUL bytes in the
333 block of data. Investigation showed the copy to actually be needless, the
334 data being length-specified.
336 JH/25 Fix use of concurrent TLS connections under GnuTLS. When a callout was
337 done during a receiving connection, and both used TLS, global info was
338 used rather than per-connection info for tracking the state of data
339 queued for transmission. This could result in a connection hang.
341 JH/26 Fix use of the SIZE parameter on MAIL commands, on continued connections.
342 Previously, when delivering serveral messages down a single connection
343 only the first would provide a SIZE. This was due to the size information
344 not being properly tracked.
346 JH/27 Bug 2530: When operating in a timezone with sub-minute offset, such as
347 TAI (at 37 seconds currently), pretend to be in UTC for time-related
348 expansion and logging. Previously, spurious values such as a future
349 minute could be seen.
351 JH/28 Bug 2533: Fix expansion of ${tr } item. When called in some situations
352 it could crash from a null-deref. This could also affect the
353 ${addresses: } operator and ${readsock } item.
355 JH/29 Bug 2537: Fix $mime_part_count. When a single connection had a non-mime
356 message following a mime one, the variable was not reset.
358 JH/30 When an pipelined-connect fails at the first response, assume incorrect
359 cached capability (perhaps the peer reneged?) and immediately retry in
362 JH/31 Fix spurious detection of timeout while writing to transport filter.
364 JH/32 Bug 2541: Fix segfault on bad cmdline -f (sender) argument. Previously
365 an attempt to copy the string was made before checking it.
367 JH/33 Fix the dsearch lookup to return an untainted result. Previously the
368 taint of the lookup key was maintained; we now regard the presence in the
369 filesystem as sufficient validation.
371 JH/34 Fix the readsocket expansion to not segfault when an empty "options"
372 argument is supplied.
374 JH/35 The dsearch lookup now requires that the directory is an absolute path.
375 Previously this was not checked, and nonempty relative paths made an
376 access under Exim's current working directory.
378 JH/36 Bug 2554: Fix msg:defer event for the hosts_max_try_hardlimit case.
379 Previously no event was raised.
381 JH/37 Bug 2552: Fix the check on spool space during reception to use the SIZE
382 parameter supplied by the sender MAIL FROM command. Previously it was
383 ignored, and only the check_spool_space option value for the required
386 JH/38 Fix $dkim_key_length. This should, after a DKIM verification, present
387 the size of the signing public-key. Previously it was instead giving
388 the size of the signature hash.
390 JH/39 DKIM verification: the RFC 8301 restriction on sizes of RSA keys is now
391 the default. See the (new) dkim_verify_min_keysizes option.
393 JH/40 Fix a memory-handling bug: when a connection carried multiple messages
394 and an ACL use a lookup for checking either the local_part or domain,
395 stale data could be accessed. Ensure that variable references are
396 dropped between messages.
398 JH/41 Bug 2571: Fix SPA authenticator. Running as a server, an offset supplied
399 by the client was not checked as pointing within response data before
400 being used. A malicious client could thus cause an out-of-bounds read and
401 possibly gain authentication. Fix by adding the check.
403 JH/42 Internationalisation: change the default for downconversion in the smtp
404 transport to be "if needed". Previously it was "as previously set" for
405 the message, which usually meant "if needed" for message-submission but
406 "no" for everything else. However, MTAs have been seen using SMTPUTF8
407 even when the envelope addresses did not need it, resulting in forwarding
408 failures to non-supporting MTAs. A downconvert in such cases will be
409 a no-op on the addresses, merely dropping the use of SMTPUTF8 by the
410 transport. The change does mean that addresses needing conversion will
411 be converted when previously a delivery failure would occur.
413 JH/43 Fix possible long line in DSN. Previously when a very long SMTP error
414 response was received it would be used unchecked in a fail-DSN, violating
415 standards on line-length limits. Truncate if needed.
417 HS/01 Remove parameters of the link to www.open-spf.org. The linked form
418 doesn't work. (Additionally add a new main config option to configure the
425 JH/01 OpenSSL: With debug enabled output keying information sufficient, server
426 side, to decode a TLS 1.3 packet capture.
428 JH/02 OpenSSL: Suppress the sending of (stateful) TLS1.3 session tickets.
429 Previously the default library behaviour applied, sending two, each in
432 JH/03 Debug output for ACL now gives the config file name and line number for
435 JH/04 The default received_header_text now uses the RFC 8314 tls cipher clause.
437 JH/05 DKIM: ensure that dkim_domain elements are lowercased before use.
439 JH/06 Fix buggy handling of autoreply bounce_return_size_limit, and a possible
440 buffer overrun for (non-chunking) other transports.
442 JH/07 GnuTLS: Our use of late (post-handshake) certificate verification, under
443 TLS1.3, means that a server rejecting a client certificate is not visible
444 to the client until the first read of encrypted data (typically the
445 response to EHLO). Add detection for that case and treat it as a failed
446 TLS connection attempt, so that the normal retry-in-clear can work (if
447 suitably configured).
449 JB/01 Bug 2375: fix expansions of 822 addresses having comments in local-part
450 and/or domain. Found and fixed by Jason Betts.
452 JH/08 Add hardening against SRV & TLSA lookups the hit CNAMEs (a nonvalid
453 configuration). If a CNAME target was not a wellformed name pattern, a
456 JH/09 Logging: Fix initial listening-on line for multiple ports for an IP when
457 the OS reports them interleaved with other addresses.
459 JH/10 OpenSSL: Fix aggregation of messages. Previously, when PIPELINING was
460 used both for input and for a verify callout, both encrypted, SMTP
461 responses being sent by the server could be lost. This resulted in
462 dropped connections and sometimes bounces generated by a peer sending
465 JH/11 Harden plaintext authenticator against a badly misconfigured client-send
466 string. Previously it was possible to cause undefined behaviour in a
467 library routine (usually a crash). Found by "zerons".
469 JH/12 Bug 2384: fix "-bP smtp_receive_timeout". Previously it returned no
472 JH/13 Bug 2386: Fix builds with Dane under LibreSSL 2.9.0 onward. Some old
473 API was removed, so update to use the newer ones.
475 JH/14 Bug 1891: Close the log file if receiving a non-smtp message, without
476 any timeout set, is taking a long time. Previously we would hang on to a
477 rotated logfile "forever" if the input was arriving with long gaps
478 (a previous attempt to fix addressed lack, for a long time, of initial
481 HS/01 Bug 2390: Use message_id for tempfile creation to avoid races in a
482 shared (NFS) environment. The length of the tempfile name is now
483 4 + 16 ("hdr.$message_exim_id") which might break on file
484 systems which restrict the file name length to lower values.
487 HS/02 Bug 2390: Use message_id for tempfile creation to avoid races in a
488 shared (NFS) environment.
490 HS/03 Bug 2392: exigrep does case sensitive *option* processing (as it
491 did for all versions <4.90). Notably -M, -m, --invert, -I may be
494 JH/15 Use unsigned when creating bitmasks in macros, to avoid build errors
495 on some platforms for bit 31.
497 JH/16 GnuTLS: rework ciphersuite strings under recent library versions. Thanks
498 to changes apparently associated with TLS1.3 handling some of the APIs
499 previously used were either nonfunctional or inappropriate. Strings
500 like TLS1.3:ECDHE_SECP256R1__RSA_PSS_RSAE_SHA256__AES_256_GCM__AEAD:256
501 and TLS1.2:ECDHE_SECP256R1__RSA_SHA256__AES_128_CBC__SHA256:128 replace
502 the previous TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256 .
503 This affects log line X= elements, the $tls_{in,out}_cipher variables,
504 and the use of specific cipher names in the encrypted= ACL condition.
506 JH/17 OpenSSL: the default openssl_options now disables ssl_v3.
508 JH/18 GnuTLS: fix $tls_out_ocsp under hosts_request_ocsp. Previously the
509 verification result was not updated unless hosts_require_ocsp applied.
511 JH/19 Bug 2398: fix listing of a named-queue. Previously, even with the option
512 queue_list_requires_admin set to false, non-admin users were denied the
515 JH/20 Bug 2389: fix server advertising of usable certificates, under GnuTLS in
516 directory-of-certs mode. Previously they were advertised despite the
519 JH/21 The smtp transport option "hosts_noproxy_tls" is now unset by default.
520 A single TCP connection by a client will now hold a TLS connection open
521 for multiple message deliveries, by default. Previously the default was to
524 JH/22 The smtp transport option "hosts_try_dane" now enables all hosts by
525 default. If built with the facility, DANE will be used. The facility
526 SUPPORT_DANE is now enabled in the prototype build Makefile "EDITME".
528 JH/23 The build default is now for TLS to be included; the SUPPORT_TLS define
529 is replaced with DISABLE_TLS. Either USE_GNUTLS or (the new) USE_OPENSSL
530 must be defined and you must still, unless you define DISABLE_TLS, manage
531 the the include-dir and library-file requirements that go with that
532 choice. Non-TLS builds are still supported.
534 JH/24 Fix duplicated logging of peer name/address, on a transport connection-
537 JH/25 The smtp transport option "hosts_try_fastopen" now enables all hosts by
538 default. If the platform supports and has the facility enabled, it will
539 be requested on all coneections.
541 JH/26 The PIPE_CONNECT facility is promoted from experimental status and is now
542 controlled by the build-time option SUPPORT_PIPE_CONNECT.
544 PP/01 Unbreak heimdal_gssapi, broken in 4.92.
546 JH/27 Bug 2404: Use the main-section configuration option "dsn_from" for
547 success-DSN messages. Previously the From: header was always the default
548 one for these; the option was ignored.
550 JH/28 Fix the timeout on smtp response to apply to the whole response.
551 Previously it was reset for every read, so a teergrubing peer sending
552 single bytes within the time limit could extend the connection for a
553 long time. Credit to Qualsys Security Advisory Team for the discovery.
555 JH/29 Fix DSN Final-Recipient: field. Previously it was the post-routing
556 delivery address, which leaked information of the results of local
557 forwarding. Change to the original envelope recipient address, per
560 JH/30 Bug 2411: Fix DSN generation when RFC 3461 failure notification is
561 requested. Previously not bounce was generated and a log entry of
562 error ignored was made.
564 JH/31 Avoid re-expansion in ${sort } expansion. (CVE-2019-13917)
566 JH/32 Introduce a general tainting mechanism for values read from the input
567 channel, and values derived from them. Refuse to expand any tainted
568 values, to catch one form of exploit.
570 JH/33 Bug 2413: Fix dkim_strict option. Previously the expansion result
571 was unused and the unexpanded text used for the test. Found and
572 fixed by Ruben Jenster.
574 JH/34 Fix crash after TLS shutdown. When the TCP/SMTP channel was left open,
575 an attempt to use a TLS library read routine dereffed a nul pointer,
578 JH/35 Bug 2409: filter out-of-spec chars from callout response before using
579 them in our smtp response.
581 JH/36 Have the general router option retry_use_local_part default to true when
582 any of the restrictive preconditions are set (to anything). Previously it
583 was only for check_local user. The change removes one item of manual
584 configuration which is required for proper retries when a remote router
585 handles a subset of addresses for a domain.
587 JH/37 Appendfile: when evaluating quota use (non-quota_size_regex) take the file
588 link count into consideration.
590 HS/04 Fix handling of very log lines in -H files. If a -<key> <value> line
591 caused the extension of big_buffer, the following lines were ignored.
593 JH/38 Bug 1395: Teach the DNS negative-cache about TTL value from the SOA in
594 accordance with RFC 2308. Previously there was no expiry, so a longlived
595 receive process (eg. due to ACL delays) versus a short SOA value could
598 HS/05 Handle trailing backslash gracefully. (CVE-2019-15846)
600 JH/39 Promote DMARC support to mainline.
602 JH/40 Bug 2452: Add a References: header to DSNs.
604 JH/41 With GnuTLS 3.6.0 (and later) do not attempt to manage Diffie-Hellman
605 parameters. The relevant library call is documented as "Deprecated: This
606 function is unnecessary and discouraged on GnuTLS 3.6.0 or later. Since
607 3.6.0, DH parameters are negotiated following RFC7919."
609 HS/06 Change the default of dnssec_request_domains to "*"
611 JH/42 Bug 2545: Fix CHUNKING for all RCPT commands rejected. Previously we
612 carried on and emitted a BDAT command, even when PIPELINING was not
615 JH/43 Bug 2465: Fix taint-handling in dsearch lookup. Previously a nontainted
616 buffer was used for the filename, resulting in a trap when tainted
617 arguments (eg. $domain) were used.
619 JH/44 With OpenSSL 1.1.1 (onwards) disable renegotiation for TLS1.2 and below;
620 recommended to avoid a possible server-load attack. The feature can be
621 re-enabled via the openssl_options main cofiguration option.
623 JH/45 local_scan API: documented the current smtp_printf() call. This changed
624 for version 4.90 - adding a "more data" boolean to the arguments.
625 Bumped the ABI version number also, this having been missed previously;
626 release versions 4.90 to 4.92.3 inclusive were effectively broken in
627 respect of usage of smtp_printf() by either local_scan code or libraries
628 accessed via the ${dlfunc } expansion item. Both will need coding
629 adjustment for any calls to smtp_printf() to match the new function
630 signature; a FALSE value for the new argument is always safe.
632 JH/46 FreeBSD: fix use of the sendfile() syscall. The shim was not updating
633 the file-offset (which the Linux syscall does, and exim expects); this
634 resulted in an indefinite loop.
636 JH/47 ARC: fix crash in signing, triggered when a configuration error failed
637 to do ARC verification. The Authentication-Results: header line added
638 by the configuration then had no ARC item.
644 JH/01 Remove code calling the customisable local_scan function, unless a new
645 definition "HAVE_LOCAL_SCAN=yes" is present in the Local/Makefile.
647 JH/02 Bug 1007: Avoid doing logging from signal-handlers, as that can result in
648 non-signal-safe functions being used.
650 JH/03 Bug 2269: When presented with a received message having a stupidly large
651 number of DKIM-Signature headers, disable DKIM verification to avoid
652 a resource-consumption attack. The limit is set at twenty.
654 JH/04 Add variables $arc_domains, $arc_oldest_pass for ARC verify. Fix the
655 report of oldest_pass in ${authres } in consequence, and separate out
656 some descriptions of reasons for verification fail.
658 JH/05 Bug 2273: Cutthrough delivery left a window where the received messsage
659 files in the spool were present and unlocked. A queue-runner could spot
660 them, resulting in a duplicate delivery. Fix that by doing the unlock
661 after the unlink. Investigation by Tim Stewart. Take the opportunity to
662 add more error-checking on spoolfile handling while that code is being
665 PP/01 Refuse to open a spool data file (*-D) if it's a symlink.
666 No known attacks, no CVE, this is defensive hardening.
668 JH/06 Bug 2275: The MIME ACL unlocked the received message files early, and
669 a queue-runner could start a delivery while other operations were ongoing.
670 Cutthrough delivery was a common victim, resulting in duplicate delivery.
671 Found and investigated by Tim Stewart. Fix by using the open message data
672 file handle rather than opening another, and not locally closing it (which
673 releases a lock) for that case, while creating the temporary .eml format
674 file for the MIME ACL. Also applies to "regex" and "spam" ACL conditions.
676 JH/07 Bug 177: Make a random-recipient callout success visible in ACL, by setting
677 $sender_verify_failure/$recipient_verify_failure to "random".
679 JH/08 When generating a selfsigned cert, use serial number 1 since zero is not
682 JH/09 Bug 2274: Fix logging of cmdline args when starting in an unlinked cwd.
683 Previously this would segfault.
685 JH/10 Fix ARC signing for case when DKIM signing failed. Previously this would
688 JH/11 Bug 2264: Exim now only follows CNAME chains one step by default. We'd
689 like zero, since the resolver should be doing this for us, But we need one
690 as a CNAME but no MX presence gets the CNAME returned; we need to check
691 that doesn't point to an MX to declare it "no MX returned" rather than
692 "error, loop". A new main option is added so the older capability of
693 following some limited number of chain links is maintained.
695 JH/12 Add client-ip info to non-pass iprev ${authres } lines.
697 JH/13 For receent Openssl versions (1.1 onward) use modern generic protocol
698 methods. These should support TLS 1.3; they arrived with TLS 1.3 and the
699 now-deprecated earlier definitions used only specified the range up to TLS
700 1.2 (in the older-version library docs).
702 JH/14 Bug 2284: Fix DKIM signing for body lines starting with a pair of dots.
704 JH/15 Rework TLS client-side context management. Stop using a global, and
705 explicitly pass a context around. This enables future use of TLS for
706 connections to service-daemons (eg. malware scanning) while a client smtp
707 connection is using TLS; with cutthrough connections this is quite likely.
709 JH/16 Fix ARC verification to do AS checks in reverse order.
711 JH/17 Support a "tls" option on the ${readsocket } expansion item.
713 JH/18 Bug 2287: Fix the protocol name (eg utf8esmtp) for multiple messages
714 using the SMTPUTF8 option on their MAIL FROM commands, in one connection.
715 Previously the "utf8" would be re-prepended for every additional message.
717 JH/19 Reject MAIL FROM commands with SMTPUTF8 when the facility was not advertised.
718 Previously thery were accepted, resulting in issues when attempting to
719 forward messages to a non-supporting MTA.
721 PP/02 Let -n work with printing macros too, not just options.
723 JH/20 Bug 2296: Fix cutthrough for >1 address redirection. Previously only
724 one parent address was copied, and bogus data was used at delivery-logging
725 time. Either a crash (after delivery) or bogus log data could result.
726 Discovery and analysis by Tim Stewart.
728 PP/03 Make ${utf8clean:} expansion operator detect incomplete final character.
729 Previously if the string ended mid-character, we did not insert the
730 promised '?' replacement.
732 PP/04 Documentation: current string operators work on bytes, not codepoints.
734 JH/21 Change as many as possible of the global flags into one-bit bitfields; these
735 should pack well giving a smaller memory footprint so better caching and
736 therefore performance. Group the declarations where this can't be done so
737 that the byte-sized flag variables are not interspersed among pointer
738 variables, giving a better chance of good packing by the compiler.
740 JH/22 Bug 1896: Fix the envelope from for DMARC forensic reports to be possibly
741 non-null, to avoid issues with sites running BATV. Previously reports were
742 sent with an empty envelope sender so looked like bounces.
744 JH/23 Bug 2318: Fix the noerror command within filters. It wasn't working.
745 The ignore_error flag wasn't being returned from the filter subprocess so
746 was not set for later routers. Investigation and fix by Matthias Kurz.
748 JH/24 Bug 2310: Raise a msg:fail:internal event for each undelivered recipient,
749 and a msg:complete for the whole, when a message is manually removed using
750 -Mrm. Developement by Matthias Kurz, hacked on by JH.
752 JH/25 Avoid fixed-size buffers for pathnames in DB access. This required using
753 a "Gnu special" function, asprintf() in the DB utility binary builds; I
754 hope that is portable enough.
756 JH/26 Bug 2311: Fix DANE-TA verification under GnuTLS. Previously it was also
757 requiring a known-CA anchor certificate; make it now rely entirely on the
758 TLSA as an anchor. Checking the name on the leaf cert against the name
759 on the A-record for the host is still done for TA (but not for EE mode).
761 JH/27 Fix logging of proxy address. Previously, a pointless "PRX=[]:0" would be
762 included in delivery lines for non-proxied connections, when compiled with
763 SUPPORT_SOCKS and running with proxy logging enabled.
765 JH/28 Bug 2314: Fire msg:fail:delivery event even when error is being ignored.
766 Developement by Matthias Kurz, tweaked by JH. While in that bit of code,
767 move the existing event to fire before the normal logging of message
768 failure so that custom logging is bracketed by normal logging.
770 JH/29 Bug 2322: A "fail" command in a non-system filter (file) now fires the
771 msg:fail:internal event. Developement by Matthias Kurz.
773 JH/30 Bug 2329: Increase buffer size used for dns lookup from 2k, which was
774 far too small for todays use of crypto signatures stored there. Go all
775 the way to the max DNS message size of 64kB, even though this might be
776 overmuch for IOT constrained device use.
778 JH/31 Fix a bad use of a copy function, which could be used to pointlessly
779 copy a string over itself. The library routine is documented as not
780 supporting overlapping copies, and on MacOS it actually raised a SIGABRT.
782 JH/32 For main options check_spool_space and check_inode_space, where the
783 platform supports 64b integers, support more than the previous 2^31 kB
784 (i.e. more than 2 TB). Accept E, P and T multipliers in addition to
785 the previous G, M, k.
787 JH/33 Bug 2338: Fix the cyrus-sasl authenticator to fill in the
788 $authenticated_fail_id variable on authentication failure. Previously
791 JH/34 Increase RSA keysize of autogen selfsign cert from 1024 to 2048. RHEL 8.0
792 OpenSSL didn't want to use such a weak key. Do for GnuTLS also, and for
793 more-modern GnuTLS move from GNUTLS_SEC_PARAM_LOW to
794 GNUTLS_SEC_PARAM_MEDIUM.
796 JH/35 OpenSSL: fail the handshake when SNI processing hits a problem, server
797 side. Previously we would continue as if no SNI had been received.
799 JH/36 Harden the handling of string-lists. When a list consisted of a sole
800 "<" character, which should be a list-separator specification, we walked
801 off past the nul-terimation.
803 JH/37 Bug 2341: Send "message delayed" warning MDNs (restricted to external
804 causes) even when the retry time is not yet met. Previously they were
805 not, meaning that when (say) an account was over-quota and temp-rejecting,
806 and multiple senders' messages were queued, only one sender would get
807 notified on each configured delay_warning cycle.
809 JH/38 Bug 2351: Log failures to extract envelope addresses from message headers.
811 JH/39 OpenSSL: clear the error stack after an SSL_accept(). With anon-auth
812 cipher-suites, an error can be left on the stack even for a succeeding
813 accept; this results in impossible error messages when a later operation
816 AM/01 Bug 2359: GnuTLS: repeat lowlevel read and write operations while they
817 return error codes indicating retry. Under TLS1.3 this becomes required.
819 JH/40 Fix the feature-cache refresh for EXPERIMENTAL_PIPE_CONNECT. Previously
820 it only wrote the new authenticators, resulting in a lack of tracking of
821 peer changes of ESMTP extensions until the next cache flush.
823 JH/41 Fix the loop reading a message header line to check for integer overflow,
824 and more-often against header_maxsize. Previously a crafted message could
825 induce a crash of the recive process; now the message is cleanly rejected.
827 JH/42 Bug 2366: Fix the behaviour of the dkim_verify_signers option. It had
828 been totally disabled for all of 4.91. Discovery and fix by "Mad Alex".
834 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
835 When redis_servers is set to a list of > 1 element, and the Redis servers
836 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
837 case of MOVED into a DEFER case instead, thus moving the query onto the
838 next server in the list. For a cluster of N elements, all N servers must
839 be defined in redis_servers.
841 GF/02 Catch and remove uninitialized value warning in exiqsumm
842 Check for existence of @ARGV before looking at $ARGV[0]
844 JH/01 Replace the store_release() internal interface with store_newblock(),
845 which internalises the check required to safely use the old one, plus
846 the allocate and data copy operations duplicated in both (!) of the
847 extant use locations.
849 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
850 modifier. This matches the restriction on the commandline.
852 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
853 Previously only the last row was returned.
855 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
856 we assumed that tags in the header were well-formed, and parsed the
857 element content after inspecting only the first char of the tag.
858 Assumptions at that stage could crash the receive process on malformed
861 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
862 While running the DKIM ACL we operate on the Permanent memory pool so that
863 variables created with "set" persist to the DATA ACL. Also (at any time)
864 DNS lookups that fail create cache records using the Permanent pool. But
865 expansions release any allocations made on the current pool - so a dnsdb
866 lookup expansion done in the DKIM ACL releases the memory used for the
867 DNS negative-cache, and bad things result. Solution is to switch to the
868 Main pool for expansions.
869 While we're in that code, add checks on the DNS cache during store_reset,
870 active in the testsuite.
871 Problem spotted, and debugging aided, by Wolfgang Breyha.
873 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
874 When none of the hosts presented to a transport match an already-open
875 connection, close it and proceed with the list. Previously we would
876 queue the message. Spotted by Lena with Yahoo, probably involving
879 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
880 Previously a spurious "250 OK id=" response was appended to the proper
883 JH/08 The "support for" informational output now, which built with Content
884 Scanning support, has a line for the malware scanner interfaces compiled
885 in. Interface can be individually included or not at build time.
887 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
888 by the template makefile "src/EDITME". The "STREAM" support for an older
889 ClamAV interface method is removed.
891 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
892 rows affected is given instead).
894 JH/11 The runtime Berkeley DB library version is now additionally output by
895 "exim -d -bV". Previously only the compile-time version was shown.
897 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
898 SMTP connection. Previously, when one had more recipients than the
899 first, an abortive onward connection was made. Move to full support for
900 multiple onward connections in sequence, handling cutthrough connection
901 for all multi-message initiating connections.
903 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
904 routers. Previously, a multi-recipient message would fail to match the
905 onward-connection opened for the first recipient, and cause its closure.
907 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
908 a timeout on read on a GnuTLS initiating connection, resulting in the
909 initiating connection being dropped. This mattered most when the callout
910 was marked defer_ok. Fix to keep the two timeout-detection methods
913 JH/15 Relax results from ACL control request to enable cutthrough, in
914 unsupported situations, from error to silently (except under debug)
915 ignoring. This covers use with PRDR, frozen messages, queue-only and
918 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
920 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
921 metadata, resulting in a crash in free().
923 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
924 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
925 Broken also in d185889f4, with init system revamp.
927 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
928 Previously we abruptly closed the connection after reading a malware-
929 found indication; now we go on to read the "scan ok" response line,
932 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
933 ACL. Previously, a crash would result.
935 JH/19 Speed up macro lookups during configuration file read, by skipping non-
936 macro text after a replacement (previously it was only once per line) and
937 by skipping builtin macros when searching for an uppercase lead character.
939 JH/20 DANE support moved from Experimental to mainline. The Makefile control
940 for the build is renamed.
942 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
943 was allocated for every new TLS startup, meaning one per message. Fix
944 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
946 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
947 reported the original. Fix to report (as far as possible) the ACL
948 result replacing the original.
950 JH/23 Fix memory leak during multi-message connections using STARTTLS under
951 OpenSSL. Certificate information is loaded for every new TLS startup,
952 and the resources needed to be freed.
954 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
956 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
959 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
960 DATA response info to the (existing) per-recipient response info for
961 the "C=" log element. It can have useful tracking info from the
962 destination system. Patch from Simon Arlott.
964 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
965 length value. Previously this would segfault.
967 HS/02 Support Avast multiline protoocol, this allows passing flags to
968 newer versions of the scanner.
970 JH/28 Ensure that variables possibly set during message acceptance are marked
971 dead before release of memory in the daemon loop. This stops complaints
972 about them when the debug_store option is enabled. Discovered specifically
973 for sender_rate_period, but applies to a whole set of variables.
974 Do the same for the queue-runner and queue-list loops, for variables set
975 from spool message files. Do the same for the SMTP per-message loop, for
976 certain variables indirectly set in ACL operations.
978 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
979 as a multi-recipient message from a mailinglist manager). The coding had
980 an arbitrary cutoff number of characters while checking for more input;
981 enforced by writing a NUL into the buffer. This corrupted long / fast
982 input. The problem was exposed more widely when more pipelineing of SMTP
983 responses was introduced, and one Exim system was feeding another.
984 The symptom is log complaints of SMTP syntax error (NUL chars) on the
985 receiving system, and refused recipients seen by the sending system
986 (propating to people being dropped from mailing lists).
987 Discovered and pinpointed by David Carter.
989 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
990 replaced by the ${authresults } expansion.
992 JH/31 Bug 2257: Fix pipe transport to not use a socket-only syscall.
994 HS/03 Set a handler for SIGTERM and call exit(3) if running as PID 1. This
995 allows proper process termination in container environments.
997 JH/32 Bug 2258: Fix spool_wireformat in combination with LMTP transport.
998 Previously the "final dot" had a newline after it; ensure it is CR,LF.
1000 JH/33 SPF: remove support for the "spf" ACL condition outcome values "err_temp"
1001 and "err_perm", deprecated since 4.83 when the RFC-defined words
1002 "temperror" and "permerror" were introduced.
1004 JH/34 Re-introduce enforcement of no cutthrough delivery on transports having
1005 transport-filters or DKIM-signing. The restriction was lost in the
1006 consolidation of verify-callout and delivery SMTP handling.
1007 Extend the restriction to also cover ARC-signing.
1009 JH/35 Cutthrough: for a final-dot response timeout (and nonunderstood responses)
1010 in defer=pass mode supply a 450 to the initiator. Previously the message
1013 PP/02 DANE: add dane_require_tls_ciphers SMTP Transport option; if unset,
1014 tls_require_ciphers is used as before.
1016 HS/03 Malware Avast: Better match the Avast multiline protocol. Add
1017 "pass_unscanned". Only tmpfails from the scanner are written to
1018 the paniclog, as they may require admin intervention (permission
1019 denied, license issues). Other scanner errors (like decompression
1020 bombs) do not cause a paniclog entry.
1022 JH/36 Fix reinitialisation of DKIM logging variable between messages.
1023 Previously it was possible to log spurious information in receive log
1026 JH/37 Bug 2255: Revert the disable of the OpenSSL session caching. This
1027 triggered odd behaviour from Outlook Express clients.
1029 PP/03 Add util/renew-opendmarc-tlds.sh script for safe renewal of public
1032 JH/38 DKIM: accept Ed25519 pubkeys in SubjectPublicKeyInfo-wrapped form,
1033 since the IETF WG has not yet settled on that versus the original
1034 "bare" representation.
1036 JH/39 Fix syslog logging for syslog_timestamp=no and log_selector +millisec.
1037 Previously the millisecond value corrupted the output.
1038 Fix also for syslog_pid=no and log_selector +pid, for which the pid
1039 corrupted the output.
1045 JH/01 Rework error string handling in TLS interface so that the caller in
1046 more cases is responsible for logging. This permits library-sourced
1047 string to be attached to addresses during delivery, and collapses
1048 pairs of long lines into single ones.
1050 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
1051 during configuration. Wildcards are allowed and expanded.
1053 JH/02 Rework error string handling in DKIM to pass more info back to callers.
1054 This permits better logging.
1056 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
1057 do not close it down and have the child transport start it up again on
1058 the passed-on TCP connection. Instead, proxy the child (and any
1059 subsequent ones) for TLS via a unix-domain socket channel. Logging is
1060 affected: the continued delivery log lines do not have any DNSSEC, TLS
1061 Certificate or OCSP information. TLS cipher information is still logged.
1063 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
1064 identical IP addresses on different listening ports. Will also affect
1067 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
1068 add noisy ifdef guards to special-case this sillyness.
1069 Patch from Bernd Kuhls.
1071 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
1072 than 255 are no longer allowed.
1074 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
1075 Disable the session-cache too, which might reduce our load. Since we
1076 currrectly use a new context for every connection, both as server and
1077 client, there is no benefit for these.
1078 GnuTLS appears to not support tickets server-side by default (we don't
1079 call gnutls_session_ticket_enable_server()) but client side is enabled
1080 by default on recent versions (3.1.3 +) unless the PFS priority string
1083 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
1084 <https://reproducible-builds.org/specs/source-date-epoch/>.
1086 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
1087 the check for any unsuccessful recipients did not notice the limit, and
1088 erroneously found still-pending ones.
1090 JH/08 Pipeline CHUNKING command and data together, on kernels that support
1091 MSG_MORE. Only in-clear (not on TLS connections).
1093 JH/09 Avoid using a temporary file during transport using dkim. Unless a
1094 transport-filter is involved we can buffer the headers in memory for
1095 creating the signature, and read the spool data file once for the
1096 signature and again for transmission.
1098 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
1099 4.77 as the kernel support then wasn't solid, having issues in 64bit
1100 mode. Now, it's been long enough. Add support for FreeBSD also.
1102 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
1103 case where the routing stage had gathered several addresses to send to
1104 a host before calling the transport for the first, we previously failed
1105 to close down TLS in the old transport process before passing the TCP
1106 connection to the new process. The new one sent a STARTTLS command
1107 which naturally failed, giving a failed delivery and bloating the retry
1108 database. Investigation and fix prototype from Wolfgang Breyha.
1110 JH/12 Fix check on SMTP command input synchronisation. Previously there were
1111 false-negatives in the check that the sender had not preempted a response
1112 or prompt from Exim (running as a server), due to that code's lack of
1113 awareness of the SMTP input buffering.
1115 PP/04 Add commandline_checks_require_admin option.
1116 Exim drops privileges sanely, various checks such as -be aren't a
1117 security problem, as long as you trust local users with access to their
1118 own account. When invoked by services which pass untrusted data to
1119 Exim, this might be an issue. Set this option in main configuration
1120 AND make fixes to the calling application, such as using `--` to stop
1123 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
1124 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
1125 responses to those, into a single TLS record each way (this usually means
1126 a single packet). As a side issue, smtp_enforce_sync now works on TLS
1129 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
1130 affects you only if you're dancing at the edge of the param size limits.
1131 If you are, and this message makes sense to you, then: raise the
1132 configured limit or use OpenSSL 1.1. Nothing we can do for older
1135 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
1136 cmdline element to get the documented default one. Previously it was
1139 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
1140 are unsuable for non-message-specific reasons - eg. connection timeout,
1141 banner-time rejection.
1143 JH/16 Fix logging of delivery remote port, when specified by router, under
1146 PP/06 Repair manualroute's ability to take options in any order, even if one
1147 is the name of a transport.
1150 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
1152 JH/17 Change the list-building routines interface to use the expanding-string
1153 triplet model, for better allocation and copying behaviour.
1155 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
1156 Previously it was constructed the first time a possibly-matching string
1157 was met in the configuration file input during startup; now it is done
1160 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
1161 compatible one, to avoid the (poorly documented) possibility of a config
1162 file in the working directory redirecting the DB files, possibly correpting
1163 some existing file. CVE-2017-10140 assigned for BDB.
1165 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
1166 cache-hot. Previously, although the result was properly cached, the
1167 initial verify call returned a defer.
1169 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
1170 the main verify for receipient in uncached-mode.
1172 JH/22 Retire historical build files to an "unsupported" subdir. These are
1173 defined as "ones for which we have no current evidence of testing".
1175 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
1176 if present. Previously it was ignored.
1178 JH/24 Start using specified-initialisers in C structure init coding. This is
1179 a C99 feature (it's 2017, so now considered safe).
1181 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
1182 if was a fixed-sized field and bitmask ops via macros; it is now more
1185 PP/07 GitHub PR 56: Apply MariaDB build fix.
1186 Patch provided by Jaroslav Škarvada.
1188 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
1189 during Coverity cleanups [4.87 JH/47]
1190 Diagnosis and fix provided by Michael Fischer v. Mollard.
1192 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
1193 the right size to place the terminating semicolon on its own folded
1194 line, the header hash was calculated to an incorrect value thanks to
1195 the (relaxed) space the fold became.
1197 HS/02 Fix Bug 2130: large writes from the transport subprocess were chunked
1198 and confused the parent.
1200 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
1201 which could crash as a result. This could lead to undeliverable messages.
1203 JH/28 Logging: "next input sent too soon" now shows where input was truncated
1206 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
1207 matters on fast-turnover and PID-randomising systems, which were getting
1208 out-of-order delivery.
1210 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
1211 a possibly-overlapping copy. The symptom was that "Remote host closed
1212 connection in response to HELO" was logged instead of the actual 4xx
1215 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
1216 Previously only that bufferd was discarded, resulting in SYMTP command
1219 JH/32 DKIM: when a message has multiple signatures matching an identity given
1220 in dkim_verify_signers, run the dkim acl once for each. Previously only
1221 one run was done. Bug 2189.
1223 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
1224 "panic the current process" to "deliberately defer". The panic log is
1225 still written with the problem list name; the mail and reject logs now
1226 get a temp-reject line for the message that was being handled, saying
1227 something like "domains check lookup or other defer". The SMTP 451
1228 message is still "Temporary local problem".
1230 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
1231 A crafted sequence of BDAT commands could result in in-use memory beeing
1232 freed. CVE-2017-16943.
1234 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
1235 from SMTP input. Previously it was always done; now only done for DATA
1236 and not BDAT commands. CVE-2017-16944.
1238 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
1239 to the message (such as an overlong header line). Previously this was
1240 not done and we did not exit BDAT mode. Followon from the previous item
1241 though a different problem.
1247 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
1248 than -2003 did; needs libidn2 in addition to libidn.
1250 JH/02 The path option on a pipe transport is now expanded before use.
1252 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
1253 Patch provided by "Björn", documentation fix added too.
1255 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
1256 missing a wire-to-host endian conversion.
1258 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
1259 close after a BDAT command line could be taken as a following command,
1260 giving a synch failure. Fix by only checking for synch immediately
1261 before acknowledging the chunk.
1263 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
1264 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
1265 macro. Patches provided by Josh Soref.
1267 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
1268 Previously we did not; the RFC seems ambiguous and VRFY is not listed
1269 by IANA as a service extension. However, John Klensin suggests that we
1272 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
1273 the dkim code may be unix-mode line endings rather than smtp wire-format
1274 CRLF, so prepend a CR to any bare LF.
1276 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
1277 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
1279 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
1280 the first were themselves being wrongly included in the feed into dkim
1281 processing; with most chunk sizes in use this resulted in an incorrect
1282 body hash calculated value.
1284 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
1285 DKIM signature block, for verification. Although advised against by
1286 standards it is specifically not ruled illegal.
1288 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
1290 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
1291 missing a body hash (the bh= tag).
1293 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
1294 It seems that HAProxy sends the Proxy Protocol information in clear and
1295 only then does a TLS startup, so do the same.
1297 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
1298 TCP connections (such as for Spamd) unless the daemon successfully set
1299 Fast Open mode on its listening sockets. This fixes breakage seen on
1300 too-old kernels or those not configured for Fast Open, at the cost of
1301 requiring both directions being enabled for TFO, and TFO never being used
1302 by non-daemon-related Exim processes.
1304 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
1305 endings, at least on the first header line. Try to canonify any that get
1306 past that check, despite the cost.
1308 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
1309 now limited to an arbitrary five deep, while parsing addresses with the
1310 strip_excess_angle_brackets option enabled.
1312 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
1313 instead leave the unprompted TLS handshake in socket buffer for the
1314 TLS library to consume.
1316 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
1318 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
1320 JH/16 Drop variables when they go out of scope. Memory management drops a whole
1321 region in one operation, for speed, and this leaves assigned pointers
1322 dangling. Add checks run only under the testsuite which checks all
1323 variables at a store-reset and panics on a dangling pointer; add code
1324 explicitly nulling out all the variables discovered. Fixes one known
1325 bug: a transport crash, where a dangling pointer for $sending_ip_address
1326 originally assigned in a verify callout, is re-used.
1328 PP/06 Drop '.' from @INC in various Perl scripts.
1330 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
1332 PP/08 Reduce a number of compilation warnings under clang; building with
1333 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
1334 should be warning-free.
1336 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
1338 HS/01 Fix portability problems introduced by PP/08 for platforms where
1339 realloc(NULL) is not equivalent to malloc() [SunOS et al].
1341 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
1342 chunk. This allows us to accept broken chunked messages. We need a more
1343 general solution here.
1345 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
1346 already-broken messages in the queue.
1348 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
1350 JH/19 Fix reference counting bug in routing-generated-address tracking.
1356 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
1357 supports it and a size is available (ie. the sending peer gave us one).
1359 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
1360 years of being deprecated). The replacements are the ACLs
1361 acl_smtp_mime and acl_not_smtp_mime.
1363 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
1364 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
1365 or even an in-clear connection were permitted. Now, if the host lookup
1366 was dnssec and dane was requested then the host is only used if the
1367 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
1368 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
1369 if one fails this test.
1370 This means that a poorly-configured remote DNS will make it incommunicado;
1371 but it protects against a DNS-interception attack on it.
1373 JH/04 Bug 1810: make continued-use of an open smtp transport connection
1374 non-noisy when a race steals the message being considered.
1376 JH/05 If main configuration option tls_certificate is unset, generate a
1377 self-signed certificate for inbound TLS connections.
1379 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
1380 in rewrites and routers.
1382 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
1383 and logged a warning sing 4.83; now they are a configuration file error.
1385 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
1386 (lacking @domain). Apply the same qualification processing as RCPT.
1388 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
1390 JH/10 Support ${sha256:} applied to a string (as well as the previous
1393 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
1394 a cutthrough deliver is pending, as we always want to make a connection.
1395 This also avoids re-routing the message when later placing the cutthrough
1396 connection after a verify cache hit.
1397 Do not update it with the verify result either.
1399 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
1400 when routing results in more than one destination address.
1402 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
1403 signing (which inhibits the cutthrough capability). Previously only
1404 the presence of an option was tested; now an expansion evaluating as
1405 empty is permissible (obviously it should depend only on data available
1406 when the cutthrough connection is made).
1408 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
1409 the relevant preceding SMTP command did not note the pipelining mode.
1411 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
1412 Previously they were not counted.
1414 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
1415 as one having no matching records. Previously we deferred the message
1416 that needed the lookup.
1418 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
1419 distinguished as "(=".
1421 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
1422 for missing MX records. Previously it only worked for missing A records.
1424 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
1426 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
1427 after the data-go-ahead and data-ack. Patch from Jason Betts.
1429 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
1430 even for a "none" policy. Patch from Tony Meyer.
1432 JH/22 Fix continued use of a connection for further deliveries. If a port was
1433 specified by a router, it must also match for the delivery to be
1436 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
1437 When one of the recipients of a message was unsuitable for the connection
1438 (has no matching addresses), we lost track of needing to mark it
1439 deferred. As a result mail would be lost.
1441 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
1443 JH/25 Decoding ACL controls is now done using a binary search; the source code
1444 takes up less space and should be simpler to maintain. Merge the ACL
1445 condition decode tables also, with similar effect.
1447 JH/26 Fix problem with one_time used on a redirect router which returned the
1448 parent address unchanged. A retry would see the parent address marked as
1449 delivered, so not attempt the (identical) child. As a result mail would
1452 JH/27 Fix a possible security hole, wherein a process operating with the Exim
1453 UID can gain a root shell. Credit to http://www.halfdog.net/ for
1454 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
1457 JH/28 Enable {spool,log} filesystem space and inode checks as default.
1458 Main config options check_{log,spool}_{inodes,space} are now
1459 100 inodes, 10MB unless set otherwise in the configuration.
1461 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
1462 Previously it only applied to the main-section connection policy
1465 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
1467 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
1468 by me. Added RFC7919 DH primes as an alternative.
1470 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
1471 are not in the system include path.
1473 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
1474 GnuTLS, when a session startup failed (eg because the client disconnected)
1475 Exim did stdio operations after fclose. This was exposed by a recent
1476 change which nulled out the file handle after the fclose.
1478 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
1479 signed directly by the cert-signing cert, rather than an intermediate
1480 OCSP-signing cert. This is the model used by LetsEncrypt.
1482 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
1484 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
1485 an incoming connection.
1487 HS/02 Bug 1802: Do not half-close the connection after sending a request
1490 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
1491 fallback to "prime256v1".
1493 JH/34 SECURITY: Use proper copy of DATA command in error message.
1494 Could leak key material. Remotely exploitable. CVE-2016-9963.
1500 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
1501 and 3.4.4 - once the server is enabled to respond to an OCSP request
1502 it does even when not requested, resulting in a stapling non-aware
1503 client dropping the TLS connection.
1505 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
1506 support variable-length bit vectors. No functional change.
1508 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
1509 The I= interface field on outgoing lines is now after the H= remote
1510 host field, same as incoming lines. There is a separate
1511 outgoing_interface log selector which allows you to disable the
1514 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
1515 If not running log_selector +smtp_connection the mainlog would be held
1516 open indefinitely after a "too many connections" event, including to a
1517 deleted file after a log rotate. Leave the per net connection logging
1518 leaving it open for efficiency as that will be quickly detected by the
1519 check on the next write.
1521 HS/01 Bug 1671: Fix post transport crash.
1522 Processing the wait-<transport> messages could crash the delivery
1523 process if the message IDs didn't exist for some reason. When
1524 using 'split_spool_directory=yes' the construction of the spool
1525 file name failed already, exposing the same netto behaviour.
1527 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
1528 mime_regex ACL conditions.
1530 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
1531 to DSN fail messages (bounces): remote IP, remote greeting, remote response
1532 to HELO, local diagnostic string.
1534 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
1535 log line to debug. Even when configured with a tls authenticator many
1536 client connections are expected to not authenticate in this way, so
1537 an authenticate fail is not an error.
1539 HS/02 Add the Exim version string to the process info. This way exiwhat
1540 gives some more detail about the running daemon.
1542 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
1543 matter for fast-change records such as DNSBLs.
1545 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
1546 retry record, even if constant. There may be multiple transports with
1547 different interface settings and the retry behaviour needs to be kept
1550 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
1552 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
1554 JH/10 Bug 840: fix log_defer_output option of pipe transport
1556 JH/11 Bug 830: use same host for all RCPTS of a message, even under
1557 hosts_randomize. This matters a lot when combined with mua_wrapper.
1559 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
1560 ${quote_pgsql:<string>} operator.
1562 JH/13 Bug 1708: avoid misaligned access in cached lookup.
1564 JH/14 Change header file name for freeradius-client. Relevant if compiling
1565 with Radius support; from the Gentoo tree and checked under Fedora.
1567 JH/15 Bug 1712: Introduce $prdr_requested flag variable
1569 JH/16 Bug 1714: Permit an empty string as expansion result for transport
1570 option transport_filter, meaning no filtering.
1572 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
1574 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
1575 defaults to "*" (all hosts). The variable is now available when not built
1576 with TLS, default unset, mainly to enable keeping the testsuite sane.
1577 If a server certificate is not supplied (via tls_certificate) an error is
1578 logged, and clients will find TLS connections fail on startup. Presumably
1579 they will retry in-clear.
1580 Packagers of Exim are strongly encouraged to create a server certificate
1581 at installation time.
1583 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
1584 with the $config_file variable.
1586 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
1587 in transport context, after the attempt, and per-recipient. The latter type
1588 is per host attempted. The event data is the error message, and the errno
1589 information encodes the lookup type (A vs. MX) used for the (first) host,
1590 and the trailing two digits of the smtp 4xx response.
1592 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
1593 to write to mainlog (or rejectlog, paniclog) in the window between file
1594 creation and permissions/ownership being changed. Particularly affects
1595 installations where exicyclog is run as root, rather than exim user;
1596 result is that the running daemon panics and dies.
1598 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
1600 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
1601 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
1602 "pri" and "weight". Note that the previous implicit priority given by the
1603 list order is no longer honoured.
1605 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
1606 for DKIM processing.
1608 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
1609 by defining SUPPORT_SOCKS.
1611 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
1612 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
1613 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
1614 variables are renamed to proxy_{local,external}_{address,port}.
1616 JH/27 Move Internationalisation support from Experimental to mainline, enabled
1617 for a build by defining SUPPORT_I18N
1619 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
1620 of the query string, and make ${quote_redis:} do that quoting.
1622 JH/29 Move Events support from Experimental to mainline, enabled by default
1623 and removable for a build by defining DISABLE_EVENT.
1625 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
1627 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
1628 cached by the daemon.
1630 JH/32 Move Redis support from Experimental to mainline, enabled for a build
1631 by defining LOOKUP_REDIS. The libhiredis library is required.
1633 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
1634 keys are given for lookup.
1636 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
1637 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
1638 only supported when built with TLS support. The PolarSSL SHA routines
1639 are still used when the TLS library is too old for convenient support.
1641 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
1642 openssl_options), for security. OpenSSL forces this from version 1.1.0
1643 server-side so match that on older versions.
1645 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
1646 allocation for $value could be released as the expansion processing
1647 concluded, but leaving the global pointer active for it.
1649 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
1650 and to use the domains and local_parts ACL conditions.
1652 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
1653 incorrectly not doubled on cutthrough transmission, hence seen as a
1654 body-termination at the receiving system - resulting in truncated mails.
1655 Commonly the sender saw a TCP-level error, and retransmitted the message
1656 via the normal store-and-forward channel. This could result in duplicates
1657 received - but deduplicating mailstores were liable to retain only the
1658 initial truncated version.
1660 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
1662 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
1664 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
1665 we're in there, support oversigning also; bug 1309.
1667 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
1669 HS/04 Add support for keep_environment and add_environment options.
1671 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
1672 either intentional arithmetic overflow during PRNG, or testing config-
1675 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
1676 delivery resulted in actual delivery. Cancel cutthrough before DATA
1679 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
1680 rejects a recipient: pass the reject to the originator.
1682 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
1683 Many were false-positives and ignorable, but it's worth fixing the
1686 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
1687 for the new environment-manipulation done at startup. Move the routines
1688 from being local to tls.c to being global via the os.c file.
1690 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
1691 an extract embedded as result-arg for a map, the first arg for extract
1692 is unavailable so we cannot tell if this is a numbered or keyed
1693 extraction. Accept either.
1699 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
1702 JH/02 The smtp transport option "multi_domain" is now expanded.
1704 JH/03 The smtp transport now requests PRDR by default, if the server offers
1707 JH/04 Certificate name checking on server certificates, when exim is a client,
1708 is now done by default. The transport option tls_verify_cert_hostnames
1709 can be used to disable this per-host. The build option
1710 EXPERIMENTAL_CERTNAMES is withdrawn.
1712 JH/05 The value of the tls_verify_certificates smtp transport and main options
1713 default to the word "system" to access the system default CA bundle.
1714 For GnuTLS, only version 3.0.20 or later.
1716 JH/06 Verification of the server certificate for a TLS connection is now tried
1717 (but not required) by default. The verification status is now logged by
1718 default, for both outbound TLS and client-certificate supplying inbound
1721 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
1724 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
1725 Status Notification (bounce) messages are now MIME format per RFC 3464.
1726 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
1727 under the control of the dsn_advertise_hosts option, and routers may
1728 have a dsn_lasthop option.
1730 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
1731 default, modifiable by a malware= option. The list separator for
1732 the options can now be changed in the usual way. Bug 68.
1734 JH/10 The smtp_receive_timeout main option is now expanded before use.
1736 JH/11 The incoming_interface log option now also enables logging of the
1737 local interface on delivery outgoing connections.
1739 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
1740 if the interface and destination host and port all match.
1742 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
1745 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
1746 Patch from Andrew Lewis.
1748 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
1749 now supports optional time-restrictions, weighting, and priority
1750 modifiers per server. Patch originally by <rommer@active.by>.
1752 JH/16 The spamd_address main option now supports a mixed list of local
1753 and remote servers. Remote servers can be IPv6 addresses, and
1754 specify a port-range.
1756 JH/17 Bug 68: The spamd_address main option now supports an optional
1757 timeout value per server.
1759 JH/18 Bug 1581: Router and transport options headers_add/remove can
1760 now have the list separator specified.
1762 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
1765 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
1768 JH/21 Support for the A6 type of dns record is withdrawn.
1770 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
1771 rather than the verbs used.
1773 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
1774 from 255 to 1024 chars.
1776 JH/24 Verification callouts now attempt to use TLS by default.
1778 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
1779 are generic router options now. The defaults didn't change.
1781 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
1782 Original patch from Alexander Shikoff, worked over by JH.
1784 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
1785 files if ZCAT_COMMAND is not executable.
1787 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
1789 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
1791 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
1792 Normally benign, it bites when the pair was led to by a CNAME;
1793 modern usage is to not canonicalize the domain to a CNAME target
1794 (and we were inconsistent anyway for A-only vs AAAA+A).
1796 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
1798 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
1799 when evaluating $sender_host_dnssec.
1801 JH/31 Check the HELO verification lookup for DNSSEC, adding new
1802 $sender_helo_dnssec variable.
1804 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
1806 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
1808 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
1810 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
1811 documented as working, but never had. Support all but $spam_report.
1813 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
1814 added for tls authenticator.
1816 HS/03 Add perl_taintmode main config option
1822 TL/01 When running the test suite, the README says that variables such as
1823 no_msglog_check are global and can be placed anywhere in a specific
1824 test's script, however it was observed that placement needed to be near
1825 the beginning for it to behave that way. Changed the runtest perl
1826 script to read through the entire script once to detect and set these
1827 variables, reset to the beginning of the script, and then run through
1828 the script parsing/test process like normal.
1830 TL/02 The BSD's have an arc4random API. One of the functions to induce
1831 adding randomness was arc4random_stir(), but it has been removed in
1832 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
1833 function when detected.
1835 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
1836 cause callback expansion.
1838 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
1839 syntax errors in an expansion can be treated as a string instead of
1840 logging or causing an error, due to the internal use of bool_lax
1841 instead of bool when processing it.
1843 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
1844 server certificates when making smtp deliveries.
1846 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
1848 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
1850 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
1852 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
1853 Merged patch from Sebastian Wiedenroth.
1855 JH/05 Fix results-pipe from transport process. Several recipients, combined
1856 with certificate use, exposed issues where response data items split
1857 over buffer boundaries were not parsed properly. This eventually
1858 resulted in duplicates being sent. This issue only became common enough
1859 to notice due to the introduction of connection certificate information,
1860 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1862 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1863 size buffer was used, resulting in syntax errors when an expansion
1866 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1867 version 3.3.6 or later.
1869 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1870 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1871 both become "event_action", the variables become $event_name, $event_data
1872 and $event_defer_errno. There is a new variable $verify_mode, usable in
1873 routers, transports and related events. The tls:cert event is now also
1874 raised for inbound connections, if the main configuration event_action
1877 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1878 early OCSP support, but no stapling (appears to be less than 1.0.0).
1880 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1881 server certificate names available under the smtp transport option
1882 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1885 JH/10 Time-related extraction expansions from certificates now use the main
1886 option "timezone" setting for output formatting, and are consistent
1887 between OpenSSL and GnuTLS compilations. Bug 1541.
1889 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1890 encoded parameter in the incoming message. Bug 1558.
1892 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1893 include certificate info, eximon was claiming there were spoolfile
1896 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1898 JH/14 Log delivery-related information more consistently, using the sequence
1899 "H=<name> [<ip>]" wherever possible.
1901 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1902 are problematic for Debian distribution, omit them from the release
1905 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1907 JH/16 Fix string representation of time values on 64bit time_t architectures.
1910 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1911 output list separator was used.
1916 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1917 checkers that were complaining about end of non-void function with no
1920 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1921 This was a regression introduced in 4.83 by another bugfix.
1923 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1925 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1926 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1932 TF/01 Correctly close the server side of TLS when forking for delivery.
1934 When a message was received over SMTP with TLS, Exim failed to clear up
1935 the incoming connection properly after forking off the child process to
1936 deliver the message. In some situations the subsequent outgoing
1937 delivery connection happened to have the same fd number as the incoming
1938 connection previously had. Exim would try to use TLS and fail, logging
1939 a "Bad file descriptor" error.
1941 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1942 utilities have not been installed.
1944 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1945 temporary space as the ACL may create new global variables.
1947 TL/01 LDAP support uses per connection or global context settings, depending
1948 upon the detected version of the libraries at build time.
1950 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1951 to extract and use the src ip:port in logging and expansions as if it
1952 were a direct connection from the outside internet. PPv2 support was
1953 updated based on HAProxy spec change in May 2014.
1955 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1957 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1958 Properly escape header and check for NULL return.
1960 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1963 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1965 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1966 characters in header names, implemented as a verify condition.
1967 Contributed by Michael Fischer v. Mollard.
1969 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1970 results permerror and temperror. Previous values are deprecated but
1971 still accepted. In a future release, err_perm and err_temp will be
1972 completely removed, which will be a backward incompatibility if the
1973 ACL tests for either of these two old results. Patch contributed by
1974 user bes-internal on the mailing list.
1976 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1978 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1979 selectors, in both main and reject logs.
1981 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1984 JH/07 Add malware type "sock" for talking to simple daemon.
1986 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1988 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1989 routers/transports under cutthrough routing.
1991 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1992 numbers. Touch up "bool" conditional to keep the same definition.
1994 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1996 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1998 JH/12 Expand items in router/transport headers_add or headers_remove lists
1999 individually rather than the list as a whole. Bug 1452.
2001 Required for reasonable handling of multiple headers_ options when
2002 they may be empty; requires that headers_remove items with embedded
2003 colons must have them doubled (or the list-separator changed).
2005 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
2006 view the policy declared in the DMARC record. Currently, $dmarc_status
2007 is a combined value of both the record presence and the result of the
2010 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
2012 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
2013 dnslookup router and the smtp transport (applying to the forward
2016 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
2017 of ldap servers used for a specific lookup. Patch provided by Heiko
2020 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
2021 New variable $lookup_dnssec_authenticated for observability.
2023 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
2024 Patch submitted by Lars Timman.
2026 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
2028 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
2029 Requires trusted mode and valid format message id, aborts otherwise.
2030 Patch contributed by Heiko Schlichting.
2032 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
2033 certextract with support for various fields. Bug 1358.
2035 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
2036 is requested by default, modifiable by smtp transport option
2039 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
2040 operate on certificate variables to give certificate fingerprints
2041 Also new ${sha256:cert_variable}.
2043 JH/23 The PRDR feature is moved from being Experimental into the mainline.
2045 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
2046 Christian Aistleitner.
2048 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
2050 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
2051 file. Patch from Wolfgang Breyha.
2053 JH/25 Expand the coverage of the delivery $host and $host_address to
2054 client authenticators run in verify callout. Bug 1476.
2056 JH/26 Port service names are now accepted for tls_on_connect_ports, to
2057 align with daemon_smtp_ports. Bug 72.
2059 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
2060 support and error reporting did not work properly.
2062 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
2063 and is readable. Patch from Andrew Colin Kissa.
2065 TL/14 Enhance documentation of ${run expansion and how it parses the
2066 commandline after expansion, particularly in the case when an
2067 unquoted variable expansion results in an empty value.
2069 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
2071 PP/02 Fix internal collision of T_APL on systems which support RFC3123
2072 by renaming away from it. Addresses GH issue 15, reported by
2075 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
2077 TL/15 SECURITY: prevent double expansion in math comparison functions
2078 (can expand unsanitized data). Not remotely exploitable.
2085 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
2087 PP/02 Make -n do something, by making it not do something.
2088 When combined with -bP, the name of an option is not output.
2090 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
2093 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
2094 $sender_host_name and config options to manage this, and basic check
2097 PP/05 DSCP support for outbound connections and control modifier for inbound.
2099 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
2100 (Only plugin which currently uses this is kerberos4, which nobody should
2101 be using, but we should make it available and other future plugins might
2102 conceivably use it, even though it would break NAT; stuff *should* be
2103 using channel bindings instead).
2105 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
2106 name; added for Sendmail compatibility; requires admin caller.
2107 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
2108 just ignore it); requires trusted caller.
2109 Also parse but ignore: -Ac -Am -X<logfile>
2112 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
2114 TL/02 Add +smtp_confirmation as a default logging option.
2116 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
2117 Patch by Magnus Holmgren from 2007-02-20.
2119 TL/04 Bugzilla 1281 - Spec typo.
2120 Bugzilla 1283 - Spec typo.
2121 Bugzilla 1290 - Spec grammar fixes.
2123 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
2125 TL/06 Add Experimental DMARC support using libopendmarc libraries.
2127 TL/07 Fix an out of order global option causing a segfault. Reported to dev
2128 mailing list by by Dmitry Isaikin.
2130 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
2132 JH/02 Support "G" suffix to numbers in ${if comparisons.
2134 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
2136 NM/01 Bugzilla 1197 - Spec typo
2137 Bugzilla 1196 - Spec examples corrections
2139 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
2141 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
2142 gnutls_enable_pkcs11, but renamed to more accurately indicate its
2145 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
2146 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
2148 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
2149 "acl {{name}{arg}...}", and optional args on acl condition
2152 JH/05 Permit multiple router/transport headers_add/remove lines.
2154 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
2156 JH/07 Avoid using a waiting database for a single-message-only transport.
2157 Performance patch from Paul Fisher. Bugzilla 1262.
2159 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
2162 JH/09 Add $headers_added variable, with content from use of ACL modifier
2163 add_header (but not yet added to the message). Bugzilla 199.
2165 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
2166 Pulled from Bugzilla 817 by Wolfgang Breyha.
2168 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
2170 (nb: this is the same fix as in Exim 4.80.1)
2172 JH/11 Add A= logging on delivery lines, and a client_set_id option on
2175 JH/12 Add optional authenticated_sender logging to A= and a log_selector
2178 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
2180 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
2181 advertise SMTP AUTH mechanism to us, instead of a generic
2182 protocol violation error. Also, make Exim more robust to bad
2183 data from the Dovecot auth socket.
2185 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
2187 When a queue runner is handling a message, Exim first routes the
2188 recipient addresses, during which it prunes them based on the retry
2189 hints database. After that it attempts to deliver the message to
2190 any remaining recipients. It then updates the hints database using
2193 So if a recipient address works intermittently, it can get repeatedly
2194 deferred at routing time. The retry hints record remains fresh so the
2195 address never reaches the final cutoff time.
2197 This is a fairly common occurrence when a user is bumping up against
2198 their storage quota. Exim had some logic in its local delivery code
2199 to deal with this. However it did not apply to per-recipient defers
2200 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
2202 This change adds a proper retry rule check during routing so that the
2203 final cutoff time is checked against the message's age. We only do
2204 this check if there is an address retry record and there is not a
2205 domain retry record; this implies that previous attempts to handle
2206 the address had the retry_use_local_parts option turned on. We use
2207 this as an approximation for the destination being like a local
2208 delivery, as in LMTP.
2210 I suspect this new check makes the old local delivery cutoff check
2211 redundant, but I have not verified this so I left the code in place.
2213 TF/02 Correct gecos expansion when From: is a prefix of the username.
2215 Test 0254 submits a message to Exim with the header
2219 When I ran the test suite under the user fanf2, Exim expanded
2220 the header to contain my full name, whereas it should have added
2221 a Resent-Sender: header. It erroneously treats any prefix of the
2222 username as equal to the username.
2224 This change corrects that bug.
2226 GF/01 DCC debug and logging tidyup
2227 Error conditions log to paniclog rather than rejectlog.
2228 Debug lines prefixed by "DCC: " to remove any ambiguity.
2230 TF/03 Avoid unnecessary rebuilds of lookup-related code.
2232 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
2233 Bug spotted by Jeremy Harris; was flawed since initial commit.
2234 Would have resulted in OCSP responses post-SNI triggering an Exim
2235 NULL dereference and crash.
2237 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
2239 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
2240 Bug detection, analysis and fix by Samuel Thibault.
2241 Bugzilla 1331, Debian bug #698092.
2243 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
2245 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
2246 Server implementation by Todd Lyons, client by JH.
2247 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
2248 config variable "prdr_enable" controls whether the server
2249 advertises the facility. If the client requests PRDR a new
2250 acl_data_smtp_prdr ACL is called once for each recipient, after
2251 the body content is received and before the acl_smtp_data ACL.
2252 The client is controlled by both of: a hosts_try_prdr option
2253 on the smtp transport, and the server advertisement.
2254 Default client logging of deliveries and rejections involving
2255 PRDR are flagged with the string "PRDR".
2257 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
2258 fclose(). Diagnosis by Todd Lyons.
2260 PP/17 Update configure.default to handle IPv6 localhost better.
2261 Patch by Alain Williams (plus minor tweaks).
2264 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
2265 This is now consistent with GnuTLS, and is now documented: the
2266 previous undocumented portable approach to treating the option as
2267 unset was to force an expansion failure. That still works, and
2268 an empty string is now equivalent.
2270 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
2271 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
2272 not performing validation itself.
2274 PP/20 Added force_command boolean option to pipe transport.
2275 Patch from Nick Koston, of cPanel Inc.
2277 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
2280 TF/04 Added udpsend ACL modifier and hexquote expansion operator
2282 PP/21 Fix eximon continuous updating with timestamped log-files.
2283 Broken in a format-string cleanup in 4.80, missed when I repaired the
2284 other false fix of the same issue.
2285 Report and fix from Heiko Schlichting.
2288 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
2289 Report from Prashanth Katuri.
2291 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
2292 It's SecureTransport, so affects any MacOS clients which use the
2293 system-integrated TLS libraries, including email clients.
2295 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
2296 using a MIME ACL for non-SMTP local injection.
2297 Report and assistance in diagnosis by Warren Baker.
2299 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
2301 JH/16 Fix comparisons for 64b. Bugzilla 1385.
2303 TL/09 Add expansion variable $authenticated_fail_id to keep track of
2304 last id that failed so it may be referenced in subsequent ACL's.
2306 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
2309 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
2310 ldap library initialization, allowing self-signed CA's to be
2311 used. Also properly sets require_cert option later in code by
2312 using NULL (global ldap config) instead of ldap handle (per
2313 session). Bug diagnosis and testing by alxgomz.
2315 TL/12 Enhanced documentation in the ratelimit.pl script provided in
2316 the src/util/ subdirectory.
2318 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
2319 renamed to Transport Post Delivery Action by Jeremy Harris, as
2322 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
2323 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
2324 redis_servers = needs to be configured which will be used by the redis
2325 lookup. Patch from Warren Baker, of The Packet Hub.
2327 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
2329 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
2330 hostname or reverse DNS when processing a host list. Used suggestions
2331 from multiple comments on this bug.
2333 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
2335 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
2336 Missed a few lines, added it to make the runtest require no keyboard
2339 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
2340 contains upper case chars. Make router use caseful_local_part.
2342 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
2343 support when GnuTLS has been built with p11-kit.
2349 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
2351 This, or similar/improved, will also be change PP/11 of 4.82.
2357 PP/01 Handle short writes when writing local log-files.
2358 In practice, only affects FreeBSD (8 onwards).
2359 Bugzilla 1053, with thanks to Dmitry Isaikin.
2361 NM/01 Bugzilla 949 - Documentation tweak
2363 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
2366 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
2368 PP/02 Implemented gsasl authenticator.
2370 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
2372 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
2373 `pkg-config foo` for cflags/libs.
2375 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
2376 with rest of GSASL and with heimdal_gssapi.
2378 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
2379 `pkg-config foo` for cflags/libs for the TLS implementation.
2381 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
2382 properties get this fed in as external SSF. A number of robustness
2383 and debugging improvements to the cyrus_sasl authenticator.
2385 PP/08 cyrus_sasl server now expands the server_realm option.
2387 PP/09 Bugzilla 1214 - Log authentication information in reject log.
2388 Patch by Jeremy Harris.
2390 PP/10 Added dbmjz lookup type.
2392 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
2394 PP/12 MAIL args handles TAB as well as SP, for better interop with
2395 non-compliant senders.
2396 Analysis and variant patch by Todd Lyons.
2398 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
2399 Bug report from Lars Müller <lars@samba.org> (via SUSE),
2400 Patch from Dirk Mueller <dmueller@suse.com>
2402 PP/13 tls_peerdn now print-escaped for spool files.
2403 Observed some $tls_peerdn in wild which contained \n, which resulted
2404 in spool file corruption.
2406 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
2407 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
2408 or write after TLS renegotiation, which otherwise led to messages
2411 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
2412 as a tracking header (ie: a signed header comes before the signature).
2413 Patch from Wolfgang Breyha.
2415 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
2416 comma-sep list; embedded commas doubled.
2418 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
2420 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
2422 Report and patch from Dmitry Banschikov.
2424 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
2425 Removed SSL_clear() after SSL_new() which led to protocol negotiation
2426 failures. We appear to now support TLS1.1+ with Exim.
2428 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
2429 lets Exim select keys and certificates based upon TLS SNI from client.
2430 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
2431 before an outbound SMTP session. New log_selector, +tls_sni.
2433 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
2434 NULL dereference. Report and patch from Alun Jones.
2436 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
2437 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
2438 Not seeing resolver debug output on NetBSD, but suspect this is a
2439 resolver implementation change.
2441 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
2442 Left warnings. Added "eximon gdb" invocation mode.
2444 PP/21 Defaulting "accept_8bitmime" to true, not false.
2446 PP/22 Added -bw for inetd wait mode support.
2448 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
2449 locate the relevant includes and libraries. Made this the default.
2451 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
2452 Bugzilla 1246, report and most of solution from Tomasz Kusy.
2454 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
2455 This may cause build issues on older platforms.
2457 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
2458 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
2459 gnutls_require_mac & gnutls_require_protocols (no longer supported).
2460 Added SNI support via GnuTLS too.
2461 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
2463 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
2465 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
2466 Applied second patch from Janne, implementing suggestion to default
2467 multiple-strings-in-record handling to match SPF spec.
2469 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
2471 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
2472 read-only, out of scope).
2473 Patch from Wolfgang Breyha, report from Stuart Northfield.
2475 PP/29 Fix three issues highlighted by clang analyser static analysis.
2476 Only crash-plausible issue would require the Cambridge-specific
2477 iplookup router and a misconfiguration.
2478 Report from Marcin Mirosław.
2480 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
2482 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
2483 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
2484 As part of this, removing so much warning spew let me fix some minor
2485 real issues in debug logging.
2487 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
2488 assignment on my part. Fixed.
2490 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
2491 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
2492 Janne Snabb (who went above and beyond: thank you).
2494 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
2495 string otherwise requires a connection and a bunch more work and it's
2496 relatively easy to get wrong. Should also expose TLS library linkage
2499 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
2500 64-bit ${eval} (JH/03).
2502 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
2503 GNU libc to support some of the 64-bit stuff, should not lead to
2504 conflicts. Defined before os.h is pulled in, so if a given platform
2505 needs to override this, it can.
2507 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
2508 protection layer was required, which is not implemented.
2509 Bugzilla 1254, patch from Wolfgang Breyha.
2511 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
2512 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
2513 tls_dhparam take prime identifiers. Also unbreak combination of
2514 OpenSSL+DH_params+TLSSNI.
2516 PP/39 Disable SSLv2 by default in OpenSSL support.
2522 PP/01 Solaris build fix for Oracle's LDAP libraries.
2523 Bugzilla 1109, patch from Stephen Usher.
2525 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
2527 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
2530 TF/02 Fix a couple more cases where we did not log the error message
2531 when unlink() failed. See also change 4.74-TF/03.
2533 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
2534 lock up or crash if it happened to be inside a call to libc when it
2535 got a SIGUSR1 from exiwhat.
2537 The SIGUSR1 handler appends the current process status to the process
2538 log which is later printed by exiwhat. It used to use the general
2539 purpose logging code to do this, but several functions it calls are
2540 not safe for signals.
2542 The new output code in the SIGUSR1 handler is specific to the process
2543 log, and simple enough that it's easy to inspect for signal safety.
2544 Removing some special cases also simplifies the general logging code.
2545 Removing the spurious timestamps from the process log simplifies
2548 TF/04 Improved ratelimit ACL condition.
2550 The /noupdate option has been deprecated in favour of /readonly which
2551 has clearer semantics. The /leaky, /strict, and /readonly update modes
2552 are mutually exclusive. The update mode is no longer included in the
2553 database key; it just determines when the database is updated. (This
2554 means that when you upgrade Exim will forget old rate measurements.)
2556 Exim now checks that the per_* options are used with an update mode that
2557 makes sense for the current ACL. For example, when Exim is processing a
2558 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
2559 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
2560 must specify per_mail/readonly. If you omit the update mode it defaults to
2561 /leaky where that makes sense (as before) or /readonly where required.
2563 The /noupdate option is now undocumented but still supported for
2564 backwards compatibility. It is equivalent to /readonly except that in
2565 ACLs where /readonly is required you may specify /leaky/noupdate or
2566 /strict/noupdate which are treated the same as /readonly.
2568 A useful new feature is the /count= option. This is a generalization
2569 of the per_byte option, so that you can measure the throughput of other
2570 aggregate values. For example, the per_byte option is now equivalent
2571 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
2573 The per_rcpt option has been generalized using the /count= mechanism
2574 (though it's more complicated than the per_byte equivalence). When it is
2575 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
2576 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
2577 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
2578 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
2579 Note that using per_rcpt with a non-readonly update mode in more than
2580 one ACL will cause the recipients to be double-counted. (The per_mail
2581 and per_byte options don't have this problem.)
2583 The handling of very low rates has changed slightly. If the computed rate
2584 is less than the event's count (usually one) then this event is the first
2585 after a long gap. In this case the rate is set to the same as this event's
2586 count, so that the first message of a spam run is counted properly.
2588 The major new feature is a mechanism for counting the rate of unique
2589 events. The new per_addr option counts the number of different
2590 recipients that someone has sent messages to in the last time period. It
2591 behaves like per_rcpt if all the recipient addresses are different, but
2592 duplicate recipient addresses do not increase the measured rate. Like
2593 the /count= option this is a general mechanism, so the per_addr option
2594 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
2595 example, measure the rate that a client uses different sender addresses
2596 with the options per_mail/unique=$sender_address. There are further
2597 details in the main documentation.
2599 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
2601 TF/06 Removed a few PCRE remnants.
2603 TF/07 Automatically extract Exim's version number from tags in the git
2604 repository when doing development or release builds.
2606 PP/02 Raise smtp_cmd_buffer_size to 16kB.
2607 Bugzilla 879. Patch from Paul Fisher.
2609 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
2610 Heavily based on revision 40f9a89a from Simon Arlott's tree.
2613 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
2615 PP/05 Variable $av_failed, true if the AV scanner deferred.
2616 Bugzilla 1078. Patch from John Horne.
2618 PP/06 Stop make process more reliably on build failure.
2619 Bugzilla 1087. Patch from Heiko Schlittermann.
2621 PP/07 Make maildir_use_size_file an _expandable_ boolean.
2622 Bugzilla 1089. Patch from Heiko Schlittermann.
2624 PP/08 Handle ${run} returning more data than OS pipe buffer size.
2625 Bugzilla 1131. Patch from Holger Weiß.
2627 PP/09 Handle IPv6 addresses with SPF.
2628 Bugzilla 860. Patch from Wolfgang Breyha.
2630 PP/10 GnuTLS: support TLS 1.2 & 1.1.
2632 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
2635 PP/11 match_* no longer expand right-hand-side by default.
2636 New compile-time build option, EXPAND_LISTMATCH_RHS.
2637 New expansion conditions, "inlist", "inlisti".
2639 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
2641 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
2643 PP/14 fix log_write() format string regression from TF/03.
2644 Bugzilla 1152. Patch from Dmitry Isaikin.
2650 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
2652 PP/02 Harmonised TLS library version reporting; only show if debugging.
2653 Layout now matches that introduced for other libraries in 4.74 PP/03.
2655 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
2657 PP/04 New "dns_use_edns0" global option.
2659 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
2662 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
2663 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
2665 TK/01 Updated PolarSSL code to 0.14.2.
2666 Bugzilla 1097. Patch from Andreas Metzler.
2668 PP/07 Catch divide-by-zero in ${eval:...}.
2669 Fixes bugzilla 1102.
2671 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
2674 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
2675 format-string attack -- SECURITY: remote arbitrary code execution.
2677 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
2678 time unintentionally subject to list matching rules, letting the header
2679 cause arbitrary Exim lookups (of items which can occur in lists, *not*
2680 arbitrary string expansion). This allowed for information disclosure.
2682 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
2683 INT_MIN/-1 -- value coerced to INT_MAX.
2689 NM/01 Workaround for PCRE version dependency in version reporting
2692 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
2693 This fixes portability to compilers other than gcc, notably
2694 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
2696 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
2697 makefiles for portability to HP-UX and POSIX correctness.
2699 PP/01 Permit LOOKUP_foo enabling on the make command-line.
2700 Also via indented variable definition in the Makefile.
2701 (Debugging by Oliver Heesakkers).
2703 PP/02 Restore caching of spamd results with expanded spamd_address.
2704 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
2706 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
2707 Improves build reliability. Fix from: Frank Elsner
2709 NM/02 Fix wide character breakage in the rfc2047 coding
2710 Fixes bug 1064. Patch from Andrey N. Oktyabrski
2712 NM/03 Allow underscore in dnslist lookups
2713 Fixes bug 1026. Patch from Graeme Fowler
2715 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
2716 Code patches from Adam Ciarcinski of NetBSD.
2718 NM/04 Fixed exiqgrep to cope with mailq missing size issue
2721 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
2722 is logged, to avoid truncation. Patch from John Horne.
2724 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
2725 Patch from Jakob Hirsch.
2727 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
2728 SQL string expansion failure details.
2729 Patch from Andrey Oktyabrski.
2731 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
2732 Patch from Simon Arlott.
2734 PP/09 New lookups functionality failed to compile on old gcc which rejects
2735 extern declarations in function scope.
2736 Patch from Oliver Fleischmann
2738 PP/10 Use sig_atomic_t for flags set from signal handlers.
2739 Check getgroups() return and improve debugging.
2740 Fixed developed for diagnosis in bug 927 (which turned out to be
2743 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
2744 Patch from Mark Zealey.
2746 PP/12 Bugzilla 1056: Improved spamd server selection.
2747 Patch from Mark Zealey.
2749 PP/13 Bugzilla 1086: Deal with maildir quota file races.
2750 Based on patch from Heiko Schlittermann.
2752 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
2753 Patch from Uwe Doering, sign-off by Michael Haardt.
2755 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
2756 variable declaration deep within a block. Bug and patch from
2759 PP/15 lookups-Makefile IRIX compatibility coercion.
2761 PP/16 Make DISABLE_DKIM build knob functional.
2763 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
2764 Patch by Simon Arlott
2766 TF/03 Fix valgrind.h portability to C89 compilers that do not support
2767 variable argument macros. Our copy now differs from upstream.
2773 TF/01 Failure to get a lock on a hints database can have serious
2774 consequences so log it to the panic log.
2776 TF/02 Log LMTP confirmation messages in the same way as SMTP,
2777 controlled using the smtp_confirmation log selector.
2779 TF/03 Include the error message when we fail to unlink a spool file.
2781 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
2782 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
2783 for maintaining out-of-tree patches for some time.
2785 PP/01 Bugzilla 139: Documentation and portability issues.
2786 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
2787 Handle per-OS dynamic-module compilation flags.
2789 PP/02 Let /dev/null have normal permissions.
2790 The 4.73 fixes were a little too stringent and complained about the
2791 permissions on /dev/null. Exempt it from some checks.
2792 Reported by Andreas M. Kirchwitz.
2794 PP/03 Report version information for many libraries, including
2795 Exim version information for dynamically loaded libraries. Created
2796 version.h, now support a version extension string for distributors
2797 who patch heavily. Dynamic module ABI change.
2799 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
2800 privilege escalation vulnerability whereby the Exim run-time user
2801 can cause root to append content of the attacker's choosing to
2804 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
2807 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
2808 If dropping privileges for untrusted macros, we disabled normal logging
2809 on the basis that it would fail; for the Exim run-time user, this is not
2810 the case, and it resulted in successful deliveries going unlogged.
2811 Fixed. Reported by Andreas Metzler.
2817 PP/01 Date: & Message-Id: revert to normally being appended to a message,
2818 only prepend for the Resent-* case. Fixes regression introduced in
2819 Exim 4.70 by NM/22 for Bugzilla 607.
2821 PP/02 Include check_rfc2047_length in configure.default because we're seeing
2822 increasing numbers of administrators be bitten by this.
2824 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
2826 PP/03 Bugzilla 994: added openssl_options main configuration option.
2828 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
2830 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
2832 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
2834 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
2835 without a peer certificate, leading to a segfault because of an
2836 assumption that peers always have certificates. Be a little more
2837 paranoid. Problem reported by Martin Tscholak.
2839 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
2840 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
2841 NB: ClamAV planning to remove STREAM in "middle of 2010".
2842 CL also introduces -bmalware, various -d+acl logging additions and
2843 more caution in buffer sizes.
2845 PP/09 Implemented reverse_ip expansion operator.
2847 PP/10 Bugzilla 937: provide a "debug" ACL control.
2849 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
2851 PP/12 Bugzilla 973: Implement --version.
2853 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2855 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2857 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2859 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2860 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2861 ignore trailing whitespace.
2863 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2865 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2868 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2869 Notification from Dr Andrew Aitchison.
2871 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2872 ExtendedDetectionInfo response format.
2873 Notification from John Horne.
2875 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2878 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2879 XSL and documented dependency on system catalogs, with examples of how
2882 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2885 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2886 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2887 configuration file which is writeable by the Exim user or group.
2889 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2890 of configuration files to cover files specified with the -C option if
2891 they are going to be used with root privileges, not just the default
2894 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2895 option (effectively making it always true).
2897 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2898 files to be used while preserving root privileges.
2900 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2901 that rogue child processes cannot use them.
2903 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2904 run-time user, instead of root.
2906 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2907 Exim run-time user without dropping privileges.
2909 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2910 result string, instead of calling string_vformat() twice with the same
2913 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2914 for other users. Others should always drop root privileges if they use
2915 -C on the command line, even for a whitelisted configure file.
2917 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2919 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2925 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2926 $data_path, and $header_path variables; fixed documentation bugs and
2929 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2930 exipick to access non-standard spools, including the "frozen" queue
2933 NM/01 Bugzilla 965: Support mysql stored procedures.
2934 Patch from Alain Williams
2936 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2938 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2939 Patch from Andreas Metzler
2941 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2942 Patch from Kirill Miazine
2944 NM/05 Bugzilla 671: Added umask to procmail example.
2946 JJ/03 installed exipick 20100323.0, fixing doc bug
2948 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2949 directory. Notification and patch from Dan Rosenberg.
2951 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2953 TK/02 Improve log output when DKIM signing operation fails.
2955 MH/01 Treat the transport option dkim_domain as a colon separated
2956 list, not as a single string, and sign the message with each element,
2957 omitting multiple occurences of the same signer.
2959 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2960 Bugzilla 985, 986. Patch by Simon Arlott
2962 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2963 Patch by Simon Arlott
2965 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2966 MBX locking. Notification from Dan Rosenberg.
2972 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2974 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2976 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2978 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2980 NM/04 Bugzilla 915: Flag broken perl installation during build.
2986 TK/01 Added patch by Johannes Berg that expands the main option
2987 "spamd_address" if it starts with a dollar sign.
2989 TK/02 Write list of recipients to X-Envelope-Sender header when building
2990 the mbox-format spool file for content scanning (suggested by Jakob
2993 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2994 (http://www.dcc-servers.net/) support via dccifd. Activated by
2995 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2997 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2998 by Mark Daniel Reidel <mr@df.eu>.
3000 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
3001 When building exim an external PCRE library is now needed -
3002 PCRE is a system library on the majority of modern systems.
3003 See entry on PCRE_LIBS in EDITME file.
3005 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
3006 conversation. Added nologin parameter to request.
3007 Patch contributed by Kirill Miazine.
3009 TF/01 Do not log submission mode rewrites if they do not change the address.
3011 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
3013 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
3014 log files in place. Contributed by Roberto Lima.
3016 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
3018 TF/03 Bugzilla 615: When checking the local_parts router precondition
3019 after a local_part_suffix or local_part_prefix option, Exim now
3020 does not use the address's named list lookup cache, since this
3021 contains cached lookups for the whole local part.
3023 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
3024 Robert Millan. Documentation is in experimental-spec.txt.
3026 TF/04 Bugzilla 668: Fix parallel build (make -j).
3028 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
3030 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
3031 Patch provided by Jan Srzednicki.
3033 TF/05 Leading white space used to be stripped from $spam_report which
3034 wrecked the formatting. Now it is preserved.
3036 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
3037 that they are available at delivery time.
3039 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
3041 TF/08 TLS error reporting now respects the incoming_interface and
3042 incoming_port log selectors.
3044 TF/09 Produce a more useful error message if an SMTP transport's hosts
3045 setting expands to an empty string.
3047 NM/06 Bugzilla 744: EXPN did not work under TLS.
3048 Patch provided by Phil Pennock.
3050 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
3051 Patch provided by Richard Godbee.
3053 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
3054 acl_smtp_notquit, added index entry.
3056 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
3057 Patch provided by Eugene Bujak.
3059 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
3060 accept(). Patch provided by Maxim Dounin.
3062 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
3063 Patch provided by Phil Pennock.
3065 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
3067 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
3068 Patch provided by Brad "anomie" Jorsch.
3070 NM/14 Bugzilla 622: Added timeout setting to transport filter.
3071 Patch provided by Dean Brooks.
3073 TK/05 Add native DKIM support (does not depend on external libraries).
3075 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
3076 Patch provided by Graeme Fowler.
3078 NM/16 Bugzilla 851: Documentation example syntax fix.
3080 NM/17 Changed NOTICE file to remove references to embedded PCRE.
3082 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
3085 NM/19 Bugzilla 745: TLS version reporting.
3086 Patch provided by Phil Pennock.
3088 NM/20 Bugzilla 167: bool: condition support.
3089 Patch provided by Phil Pennock.
3091 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
3092 clients. Patch provided by Phil Pennock.
3094 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
3095 Patch provided by Brad "anomie" Jorsch.
3097 NM/23 Bugzilla 687: Fix misparses in eximstats.
3098 Patch provided by Heiko Schlittermann.
3100 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
3101 Patch provided by Heiko Schlittermann.
3103 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
3104 plus update to original patch.
3106 NM/26 Bugzilla 799: Documentation correction for ratelimit.
3108 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
3109 Patch provided by David Brownlee.
3111 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
3113 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
3115 NM/30 Bugzilla 888: TLS documentation bugfixes.
3117 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
3119 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
3120 Unlike the original bugzilla I have changed all shell scripts in src tree.
3122 NM/33 Bugzilla 898: Transport filter timeout fix.
3123 Patch by Todd Rinaldo.
3125 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
3126 Patch by Serge Demonchaux.
3128 NM/35 Bugzilla 39: Base64 decode bug fixes.
3129 Patch by Jakob Hirsch.
3131 NM/36 Bugzilla 909: Correct connect() call in dcc code.
3133 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
3135 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
3137 NM/39 Bugzilla 911: Fixed MakeLinks build script.
3143 TK/01 Add preliminary DKIM support. Currently requires a forked version of
3144 ALT-N's libdkim that I have put here:
3145 http://duncanthrax.net/exim-experimental/
3147 Note to Michael Haardt: I had to rename some vars in sieve.c. They
3148 were called 'true' and it seems that C99 defines that as a reserved
3149 keyword to be used with 'bool' variable types. That means you could
3150 not include C99-style headers which use bools without triggering
3151 build errors in sieve.c.
3153 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
3154 as mailq or other aliases. Changed the --help handling significantly
3155 to do whats expected. exim_usage() emits usage/help information.
3157 SC/01 Added the -bylocaldomain option to eximstats.
3159 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
3161 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
3163 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
3169 PH/01 Another patch from the Sieve maintainer.
3171 PH/02 When an IPv6 address is converted to a string for single-key lookup
3172 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
3173 dots are used instead of colons so that keys in lsearch files need not
3174 contain colons. This was done some time before quoting was made available
3175 in lsearch files. However, iplsearch files do require colons in IPv6 keys
3176 (notated using the quote facility) so as to distinguish them from IPv4
3177 keys. This meant that lookups for IP addresses in host lists did not work
3178 for iplsearch lookups.
3180 This has been fixed by arranging for IPv6 addresses to be expressed with
3181 colons if the lookup type is iplsearch. This is not incompatible, because
3182 previously such lookups could never work.
3184 The situation is now rather anomalous, since one *can* have colons in
3185 ordinary lsearch keys. However, making the change in all cases is
3186 incompatible and would probably break a number of configurations.
3188 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
3191 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
3192 conversion specification without a maximum field width, thereby enabling
3193 a rogue spamd server to cause a buffer overflow. While nobody in their
3194 right mind would setup Exim to query an untrusted spamd server, an
3195 attacker that gains access to a server running spamd could potentially
3196 exploit this vulnerability to run arbitrary code as the Exim user.
3198 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
3199 $primary_hostname instead of what libspf2 thinks the hosts name is.
3201 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
3202 a directory entry by the name of the lookup key. Previously, if a
3203 symlink pointed to a non-existing file or a file in a directory that
3204 Exim lacked permissions to read, a lookup for a key matching that
3205 symlink would fail. Now it is enough that a matching directory entry
3206 exists, symlink or not. (Bugzilla 503.)
3208 PH/03 The body_linecount and body_zerocount variables are now exported in the
3211 PH/04 Added the $dnslist_matched variable.
3213 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
3214 This means they are set thereafter only if the connection becomes
3217 PH/06 Added the client_condition to authenticators so that some can be skipped
3218 by clients under certain conditions.
3220 PH/07 The error message for a badly-placed control=no_multiline_responses left
3221 "_responses" off the end of the name.
3223 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
3225 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
3226 (without spaces) instead of just copying the configuration text.
3228 PH/10 Added the /noupdate option to the ratelimit ACL condition.
3230 PH/11 Added $max_received_linelength.
3232 PH/12 Added +ignore_defer and +include_defer to host lists.
3234 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
3235 way in which PCRE > 7.0 is built.
3237 PH/14 Implemented queue_only_load_latch.
3239 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
3240 MAIL command. The effect was to mangle the value on 64-bit systems.
3242 PH/16 Another patch from the Sieve maintainer.
3244 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
3246 PH/18 If a system quota error occurred while trying to create the file for
3247 a maildir delivery, the message "Mailbox is full" was not appended to the
3248 bounce if the delivery eventually timed out. Change 4.67/27 below applied
3249 only to a quota excession during the actual writing of the file.
3251 PH/19 It seems that peer DN values may contain newlines (and other non-printing
3252 characters?) which causes problems in log lines. The DN values are now
3253 passed through string_printing() before being added to log lines.
3255 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
3256 and InterBase are left for another time.)
3258 PH/21 Added message_body_newlines option.
3260 PH/22 Guard against possible overflow in moan_check_errorcopy().
3262 PH/23 POSIX allows open() to be a macro; guard against that.
3264 PH/24 If the recipient of an error message contained an @ in the local part
3265 (suitably quoted, of course), incorrect values were put in $domain and
3266 $local_part during the evaluation of errors_copy.
3272 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
3273 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
3276 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
3277 issue a MAIL command.
3279 PH/02 In an ACL statement such as
3281 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
3283 if a client was not listed at all, or was listed with a value other than
3284 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
3285 the condition was not true (as it should be), so access was not denied.
3286 The bug was that the ! inversion was incorrectly passed on to the second
3287 item. This has been fixed.
3289 PH/03 Added additional dnslists conditions == and =& which are different from
3290 = and & when the dns lookup returns more than one IP address.
3292 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
3293 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
3295 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
3296 FSYNC, which compiles an option called disable_fsync that allows for
3297 bypassing fsync(). The documentation is heavily laced with warnings.
3299 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
3301 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
3302 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
3303 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
3304 including adding "make clean"; (3) Added -fPIC when compiling the test
3305 dynamically loaded module, to get rid of a warning.
3307 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
3308 message fails, move_frozen_messages = true and ignore_bounce_errors_after
3309 = 0s. The bug is otherwise harmless.
3311 PH/07 There was a bug in the dovecot authenticator such that the value of
3312 $auth1 could be overwritten, and so not correctly preserved, after a
3313 successful authentication. This usually meant that the value preserved by
3314 the server_setid option was incorrect.
3316 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
3318 PH/09 Installed PCRE release 7.0.
3320 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
3321 run for batched SMTP input. It is now run at the start of every message
3322 in the batch. While fixing this I discovered that the process information
3323 (output by running exiwhat) was not always getting set for -bs and -bS
3324 input. This is fixed, and it now also says "batched" for BSMTP.
3326 PH/11 Added control=no_pipelining.
3328 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
3329 patch, slightly modified), and move the expansion of helo_data till after
3330 the connection is made in the smtp transport (so it can use these
3333 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
3335 PH/14 Added log_selector = +pid.
3337 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
3339 PH/16 Add ${if forany and ${if forall.
3341 PH/17 Added dsn_from option to vary the From: line in DSNs.
3343 PH/18 Flush SMTP output before performing a callout, unless control =
3344 no_callout_flush is set.
3346 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
3347 was true (the default) a successful delivery failed to delete the retry
3348 item, thus causing premature timeout of the address. The bug is now
3351 PH/20 Added hosts_avoid_pipelining to the smtp transport.
3353 PH/21 Long custom messages for fakedefer and fakereject are now split up
3354 into multiline responses in the same way that messages for "deny" and
3355 other ACL rejections are.
3357 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
3358 with slight modification.
3360 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
3361 draft, changing the syntax and factoring some duplicate code".
3363 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
3364 for deliveries of the second and subsequent messages over the same SMTP
3367 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
3368 ${reduce, with only minor "tidies".
3370 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
3372 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
3373 expansion side effects.
3375 PH/27 When a message times out after an over-quota error from an Exim-imposed
3376 quota, the bounce message says "mailbox is full". This message was not
3377 being given when it was a system quota that was exceeded. It now should
3380 MH/03 Made $recipients available in local_scan(). local_scan() already has
3381 better access to the recipient list through recipients_list[], but
3382 $recipients can be useful in postmaster-provided expansion strings.
3384 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
3385 in the case of a MAIL command with additional options following the
3386 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
3387 were accidentally chopped off.
3389 PH/29 SMTP synchronization checks are implemented when a command is read -
3390 there is a check that no more input is waiting when there shouldn't be
3391 any. However, for some commands, a delay in an ACL can mean that it is
3392 some time before the response is written. In this time, more input might
3393 arrive, invalidly. So now there are extra checks after an ACL has run for
3394 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
3395 pipelining has not been advertised.
3397 PH/30 MH's patch to allow iscntrl() characters to be list separators.
3399 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
3400 returned in the SMTP response when smtp_return_error_details was false.
3401 This has been fixed.
3403 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
3404 instead of the C I/O that was originally supplied, because problems were
3405 reported on Solaris.
3407 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
3408 Exim which did not show up earlier: it was assuming that a call to
3409 SSL_CTX_set_info_callback() might give an error value. In fact, there is
3410 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
3411 was a macro that became an assignment, so it seemed to work. This has
3412 changed to a proper function call with a void return, hence the compile
3413 error. Exim's code has been fixed.
3415 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
3418 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
3420 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
3422 PH/37 If a message is not accepted after it has had an id assigned (e.g.
3423 because it turns out to be too big or there is a timeout) there is no
3424 "Completed" line in the log. When some messages of this type were
3425 selected by exigrep, they were listed as "not completed". Others were
3426 picked up by some special patterns. I have improved the selection
3427 criteria to be more general.
3429 PH/38 The host_find_failed option in the manualroute router can now be set
3430 to "ignore", to completely ignore a host whose IP address cannot be
3431 found. If all hosts are ignored, the behaviour is controlled by the new
3432 host_all_ignored option.
3434 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
3435 homing or because of multiple MX records with /mx) generated more than
3436 one IP address, and the following item turned out to be the local host,
3437 all the secondary addresses of the first item were incorrectly removed
3438 from the list, along with the local host and any following hosts (which
3439 is what is supposed to happen).
3441 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
3442 whoever called Exim into the -H file. In the case of the daemon it was
3443 behaving confusingly. When first started, it used values for whoever
3444 started the daemon, but after a SIGHUP it used the Exim user (because it
3445 calls itself on a restart). I have changed the code so that it now always
3448 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
3449 message are rejected with the same error (e.g. no authentication or bad
3450 sender address), and a DATA command is nevertheless sent (as can happen
3451 with PIPELINING or a stupid MUA), the error message that was given to the
3452 RCPT commands is included in the rejection of the DATA command. This is
3453 intended to be helpful for MUAs that show only the final error to their
3456 PH/42 Another patch from the Sieve maintainer.
3458 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
3459 Eximstats - Fixed some broken HTML links and added missing column headers
3461 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
3462 columns for Rejects, Temp Rejects, Ham, and Spam rows.
3464 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
3466 PH/43 Yet another patch from the Sieve maintainer.
3468 PH/44 I found a way to check for a TCP/IP connection going away before sending
3469 the response to the final '.' that terminates a message, but only in the
3470 case where the client has not sent further data following the '.'
3471 (unfortunately, this is allowed). However, in many cases there won't be
3472 any further data because there won't be any more messages to send. A call
3473 to select() can be used: if it shows that the input is "ready", there is
3474 either input waiting, or the socket has been closed. An attempt to read
3475 the next input character can distinguish the two cases. Previously, Exim
3476 would have sent an OK response which the client would never have see.
3477 This could lead to message repetition. This fix should cure that, at
3478 least in a lot of common cases.
3480 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
3481 advertised in response to EHLO.
3487 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
3488 fixed by 4.65/MH/01 (is this a record?) are fixed:
3490 (i) An empty string was always treated as zero by the numeric comparison
3491 operators. This behaviour has been restored.
3493 (ii) It is documented that the numeric comparison operators always treat
3494 their arguments as decimal numbers. This was broken in that numbers
3495 starting with 0 were being interpreted as octal.
3497 While fixing these problems I realized that there was another issue that
3498 hadn't been noticed. Values of message_size_limit (both the global option
3499 and the transport option) were treated as octal if they started with 0.
3500 The documentation was vague. These values are now always treated as
3501 decimal, and I will make that clear in the documentation.
3507 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
3508 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
3511 MH/01 Don't check that the operands of numeric comparison operators are
3512 integers when their expansion is in "skipping" mode (fixes bug
3513 introduced by 4.64-PH/07).
3515 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
3516 child addresses, Exim now panics and dies. Previously, because the count
3517 is held in a short int, deliveries were likely to be lost. As such a
3518 large number of recipients for a single message is ridiculous
3519 (performance will be very, very poor), I have chosen to impose a limit
3520 rather than extend the field.
3526 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
3527 leftover -K file (the existence of which was triggered by #402).
3528 While we were at it, introduced process PID as part of the -K
3529 filename. This should rule out race conditions when creating
3532 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
3533 processing considerably. Previous code took too long for large mails,
3534 triggering a timeout which in turn triggers #401.
3536 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
3537 in the DK code in transports.c. sendfile() is not really portable,
3538 hence the _LINUX specificness.
3540 TF/01 In the add_headers option to the mail command in an Exim filter,
3541 there was a bug that Exim would claim a syntax error in any
3542 header after the first one which had an odd number of characters
3545 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
3546 callout verification, Exim cached a "reject" for the entire domain. This
3547 is correct for most verifications, but it is not correct for a recipient
3548 verification with use_sender or use_postmaster set, because in that case
3549 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
3550 case of MAIL FROM:<> rejection from other early rejections (e.g.
3551 rejection of HELO). When verifying a recipient using a non-null MAIL
3552 address, the cache is ignored if it shows MAIL FROM:<> rejection.
3553 Whatever the result of the callout, the value of the domain cache is
3554 left unchanged (for any other kind of callout, getting as far as trying
3555 RCPT means that the domain itself is ok).
3557 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
3560 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
3561 manifest itself as EPIPE rather than ECONNECT. When tidying away a
3562 session, the daemon ignores ECONNECT errors and logs others; it now
3563 ignores EPIPE as well.
3565 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
3566 (quoted-printable decoding).
3568 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
3569 later the small subsequent patch to fix an introduced bug.
3571 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
3573 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
3575 PH/08 An error is now given if message_size_limit is specified negative.
3577 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
3578 to be given (somewhat) arbitrary names.
3580 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
3583 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
3584 miscellaneous code fixes
3586 PH/10 Added the log_reject_target ACL modifier to specify where to log
3589 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
3590 hostname. This is wrong, because it relates to the incoming message (and
3591 probably the interface on which it is arriving) and not to the outgoing
3592 callout (which could be using a different interface). This has been
3593 changed to use the value of the helo_data option from the smtp transport
3594 instead - this is what is used when a message is actually being sent. If
3595 there is no remote transport (possible with a router that sets up host
3596 addresses), $smtp_active_hostname is used.
3598 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
3599 tweaks were necessary in order to get it to work (see also 21 below):
3600 (a) The code assumed that strncpy() returns a negative number on buffer
3601 overflow, which isn't the case. Replaced with Exim's string_format()
3603 (b) There were several signed/unsigned issues. I just did the minimum
3604 hacking in of casts. There is scope for a larger refactoring.
3605 (c) The code used strcasecmp() which is not a standard C function.
3606 Replaced with Exim's strcmpic() function.
3607 (d) The code set only $1; it now sets $auth1 as well.
3608 (e) A simple test gave the error "authentication client didn't specify
3609 service in request". It would seem that Dovecot has changed its
3610 interface. Fortunately there's a specification; I followed it and
3611 changed what the client sends and it appears to be working now.
3613 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
3616 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
3617 address A is aliased to B and C, where B exists and C does not. Without
3618 -v the output is "A verified" because verification stops after a
3619 successful redirection if more than one address is generated. However,
3620 with -v the child addresses are also verified. Exim was outputting "A
3621 failed to verify" and then showing the successful verification for C,
3622 with its parentage. It now outputs "B failed to verify", showing B's
3623 parentage before showing the successful verification of C.
3625 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
3626 look up a TXT record in a specific list after matching in a combined
3629 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
3630 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
3631 they consult the DNS. I had assumed they would set it the way they
3632 wanted; and indeed my experiments on Linux seem to show that in some
3633 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
3634 To be on the safe side, however, I have now made the interface to
3635 host_find_byname() similar to host_find_bydns(), with an argument
3636 containing the DNS resolver options. The host_find_byname() function now
3637 sets these options at its start, just as host_find_bydns() does. The smtp
3638 transport options dns_qualify_single and dns_search_parents are passed to
3639 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
3640 of host_find_byname() use the default settings of RES_DEFNAMES
3641 (qualify_single) but not RES_DNSRCH (search_parents).
3643 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
3644 spool_read_header() do less string testing, by means of a preliminary
3645 switch on the second character of optional "-foo" lines. (This is
3646 overdue, caused by the large number of possibilities that now exist.
3647 Originally there were few.) While I was there, I also converted the
3648 str(n)cmp tests so they don't re-test the leading "-" and the first
3649 character, in the hope this might squeeze out yet more improvement.
3651 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
3652 flag allowing group syntax was set by the header_syntax check but not
3653 turned off, possible causing trouble later; (2) The flag was not being
3654 set at all for the header_verify test, causing "group"-style headers to
3655 be rejected. I have now set it in this case, and also caused header_
3656 verify to ignore an empty address taken from a group. While doing this, I
3657 came across some other cases where the code for allowing group syntax
3658 while scanning a header line wasn't quite right (mostly, not resetting
3659 the flag correctly in the right place). These bugs could have caused
3660 trouble for malformed header lines. I hope it is now all correct.
3662 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
3663 with the "reply" argument non-NULL. The code, however (which originally
3664 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
3665 but it didn't always do it. This confused somebody who was copying the
3666 code for some other use. I have removed all the tests.
3668 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
3669 feature that was used to support insecure browsers during the U.S. crypto
3670 embargo. It requires special client support, and Exim is probably the
3671 only MTA that supported it -- and would never use it because real RSA is
3672 always available. This code has been removed, because it had the bad
3673 effect of slowing Exim down by computing (never used) parameters for the
3674 RSA_EXPORT functionality.
3676 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
3677 authenticator to fail if there's a tab character in the incoming data
3678 (there should never be unless someone is messing about, as it's supposed
3679 to be base64-encoded). Also added, on Timo's advice, the "secured" option
3680 if the connection is using TLS or if the remote IP is the same as the
3681 local IP, and the "valid-client-cert option" if a client certificate has
3684 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
3685 authenticators. This can be used for authorization after authentication
3686 succeeds. (In the case of plaintext, it servers for both authentication
3689 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
3690 if any retry times were supplied.
3692 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
3693 connection, where there is no client IP address to check. In this
3694 situation, the verify now always succeeds.
3696 PH/25 Applied John Jetmore's -Mset patch.
3698 PH/26 Added -bem to be like -Mset, but loading a message from a file.
3700 PH/27 In a string expansion for a processed (not raw) header when multiple
3701 headers of the same name were present, leading whitespace was being
3702 removed from all of them, but trailing whitespace was being removed only
3703 from the last one. Now trailing whitespace is removed from each header
3704 before concatenation. Completely empty headers in a concatenation (as
3705 before) are ignored.
3707 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
3708 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
3710 PH/29 [Removed. This was a change that I later backed out, and forgot to
3711 correct the ChangeLog entry (that I had efficiently created) before
3712 committing the later change.]
3714 PH/30 Exim was sometimes attempting to deliver messages that had suffered
3715 address errors (4xx response to RCPT) over the same connection as other
3716 messages routed to the same hosts. Such deliveries are always "forced",
3717 so retry times are not inspected. This resulted in far too many retries
3718 for the affected addresses. The effect occurred only when there were more
3719 hosts than the hosts_max_try setting in the smtp transport when it had
3720 the 4xx errors. Those hosts that it had tried were not added to the list
3721 of hosts for which the message was waiting, so if all were tried, there
3722 was no problem. Two fixes have been applied:
3724 (i) If there are any address or message errors in an SMTP delivery, none
3725 of the hosts (tried or untried) are now added to the list of hosts
3726 for which the message is waiting, so the message should not be a
3727 candidate for sending over the same connection that was used for a
3728 successful delivery of some other message. This seems entirely
3729 reasonable: after all the message is NOT "waiting for some host".
3730 This is so "obvious" that I'm not sure why it wasn't done
3731 previously. Hope I haven't missed anything, but it can't do any
3732 harm, as the worst effect is to miss an optimization.
3734 (ii) If, despite (i), such a delivery is accidentally attempted, the
3735 routing retry time is respected, so at least it doesn't keep
3736 hammering the server.
3738 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
3739 in ${readsocket because some servers need this prod.
3741 PH/32 Added some extra debug output when updating a wait-xxx database.
3743 PH/33 The hint "could be header name not terminated by colon", which has been
3744 given for certain expansion errors for a long time, was not being given
3745 for the ${if def:h_colon_omitted{... case.
3747 PH/34 The spec says: "With one important exception, whenever a domain list is
3748 being scanned, $domain contains the subject domain." There was at least
3749 one case where this was not true.
3751 PH/35 The error "getsockname() failed: connection reset by peer" was being
3752 written to the panic log as well as the main log, but it isn't really
3753 panic-worthy as it just means the connection died rather early on. I have
3754 removed the panic log writing for the ECONNRESET error when getsockname()
3757 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
3758 runs only) independently of the message's sender address. This meant
3759 that, if the 4xx error was in fact related to the sender, a different
3760 message to the same recipient with a different sender could confuse
3761 things. In particular, this can happen when sending to a greylisting
3762 server, but other circumstances could also provoke similar problems.
3763 I have changed the default so that the retry time for these errors is now
3764 based a combination of the sender and recipient addresses. This change
3765 can be overridden by setting address_retry_include_sender=false in the
3768 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
3769 remote server are returned as part of bounce messages. This was not
3770 happening for LMTP over a pipe (the lmtp transport), but now it is the
3771 same for both kinds of LMTP.
3773 PH/38 Despite being documented as not happening, Exim was rewriting addresses
3774 in header lines that were in fact CNAMEs. This is no longer the case.
3776 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
3777 and queue runs started by the daemon processed all messages. This has
3778 been fixed so that -R and -S can now usefully be given with -q<time>.
3780 PH/40 Import PCRE release 6.7 (fixes some bugs).
3782 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
3784 PH/42 Give an error if -q is specified more than once.
3786 PH/43 Renamed the variables $interface_address and $interface_port as
3787 $received_ip_address and $received_port, to make it clear that these
3788 values apply to message reception, and not to the outgoing interface when
3789 a message is delivered. (The old names remain recognized, of course.)
3791 PH/44 There was no timeout on the connect() call when using a Unix domain
3792 socket in the ${readsocket expansion. There now is.
3794 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
3795 be meaningful with "accept".
3797 SC/01 Eximstats V1.43
3798 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
3800 SC/02 Eximstats V1.44
3801 Use a glob alias rather than an array ref in the generated
3802 parser. This improves both readability and performance.
3804 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
3805 Collect SpamAssassin and rejection statistics.
3806 Don't display local sender or destination tables unless
3807 there is data to show.
3808 Added average volumes into the top table text output.
3810 SC/04 Eximstats V1.46
3811 Collect data on the number of addresses (recipients)
3812 as well as the number of messages.
3814 SC/05 Eximstats V1.47
3815 Added 'Message too big' to the list of mail rejection
3816 reasons (thanks to Marco Gaiarin).
3818 SC/06 Eximstats V1.48
3819 Mainlog lines which have GMT offsets and are too short to
3820 have a flag are now skipped.
3822 SC/07 Eximstats V1.49 (Alain Williams)
3823 Added the -emptyok flag.
3825 SC/08 Eximstats V1.50
3826 Fixes for obtaining the IP address from reject messages.
3828 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
3829 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
3830 whitespace changes from 4.64-PH/27
3832 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
3835 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
3836 are found, allow negative numbers in numeric criteria)
3838 JJ/06 exipick.20061117.2, added new $message_body_missing variable
3840 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
3841 to match changes made in 4.64-PH/43
3843 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
3845 PH/47 Put in an explicit test for a DNS lookup of an address record where the
3846 "domain" is actually an IP address, and force a failure. This locks out
3847 those revolvers/nameservers that support "A-for-A" lookups, in
3848 contravention of the specifications.
3850 PH/48 When a host name was looked up from an IP address, and the subsequent
3851 forward lookup of the name timed out, the host name was left in
3852 $sender_host_name, contrary to the specification.
3854 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3855 restricted to single-key lookups, Exim was not diagnosing an error if
3856 * or *@ was used with a query-style lookup.
3858 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3860 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3861 long ago, but noone interested enough thought of it. Let's just say that
3862 the "1.1" means that there are some new functions that weren't there at
3863 some point in the past.
3865 PH/51 Error processing for expansion failure of helo_data from an smtp
3866 transport during callout processing was broken.
3868 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3869 tested/used via the -bh/-bhc/-bs options.
3871 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3872 bug, fixed in subsequent PCRE releases).
3874 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3875 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3877 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3883 SC/01 Use a glob alias rather than an array ref in eximstats generated
3884 parser. This improves both readability and performance.
3886 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3887 Don't display local sender or destination tables in eximstats unless
3888 there is data to show.
3889 Added average volumes into the eximstats top table text output.
3891 SC/03 Collect data on the number of addresses (recipients) as well
3892 as the number of messages in eximstats.
3894 TF/01 Correct an error in the documentation for the redirect router. Exim
3895 does (usually) call initgroups() when daemonizing.
3897 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3898 with consistent privilege compared to when running as a daemon.
3900 TF/03 Note in the spec that $authenticated_id is not set for local
3901 submissions from trusted users.
3903 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3904 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3906 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3907 by adding some example configuration directives to the default
3908 configuration file. A little bit of work is required to uncomment the
3909 directives and define how usernames and passwords are checked, but
3910 there is now a framework to start from.
3912 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3913 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3914 without this. I don't know how relevant this is to other LDAP libraries.
3916 PH/02 Add the verb name to the "unknown ACL verb" error.
3918 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3920 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3922 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3923 directory not expanded when it should be if an expanded home directory
3924 was set for the address (which is overridden by the transport).
3926 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3929 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3930 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3931 because it is too late at that time, and has no effect.
3933 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3934 security issue with \' (bugzilla #107). I could not use the
3935 PQescapeStringConn() function, because it needs a PGconn value as one of
3938 PH/08 When testing addresses using -bt, indicate those final addresses that
3939 are duplicates that would not cause an additional delivery. At least one
3940 person was confused, thinking that -bt output corresponded to deliveries.
3941 (Suppressing duplicates isn't a good idea as you lose the information
3942 about possibly different redirections that led to the duplicates.)
3944 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3945 systems where poll() doesn't work, in particular OS X.
3947 PH/10 Added more information to debugging output for retry time not reached.
3949 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3950 operations in malware.c.
3952 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3955 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3956 syslog_duplication set false, log lines that would normally be written
3957 both the the main log and to the reject log were not written to syslog at
3960 PH/14 In the default configuration, change the use of "message" in ACL warn
3961 statements to "add_header".
3963 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3964 not followed by a command (e.g. "seen endif").
3966 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3967 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3970 PH/17 Added extra conditions to the default value of delay_warning_condition
3974 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3975 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3976 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3979 The Auto-Submitted: and various List- headers are standardised, whereas I
3980 don't think Precedence: ever was.
3982 PH/18 Refactored debugging code in route_finduser() to show more information,
3983 in particular, the error code if getpwnam() issues one.
3985 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3986 This is apparently needed in addition to the PH/07 change above to avoid
3987 any possible encoding problems.
3989 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3990 but not after initializing Perl.
3992 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3993 output them only if debugging. By default they are written stderr,
3994 apparently, which is not desirable.
3996 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3999 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
4002 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
4004 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
4005 authenticated or an ident call has been made. Suppress the default
4006 values for $authenticated_id and $authenticated_sender (but permit -oMai
4007 and -oMas) when testing with -bh.
4009 PH/24 Re-jigged the order of the tests in the default configuration so that the
4010 tests for valid domains and recipients precede the DNS black list and CSA
4011 tests, on the grounds that those ones are more expensive.
4013 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
4014 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
4015 This bug exists in every version of Exim that I still have, right back to
4018 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
4019 However, an attempt to turn on case-sensitivity in a regex key by
4020 including (?-i) didn't work because the subject string was already
4021 lowercased, and the effects were non-intuitive. It turns out that a
4022 one-line patch can be used to allow (?-i) to work as expected.
4028 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
4029 other effects) broke the use of negated acl sub-conditions.
4031 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
4034 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
4035 "Deny" causes Exim to reject the incoming connection with a 554 error.
4036 Unfortunately, if there is a major crisis, such as a disk failure,
4037 tcp-wrappers gives "deny", whereas what one would like would be some
4038 kind of temporary error. A kludge has been added to help with this.
4039 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
4040 554 error is used if errno is still zero or contains ENOENT (which occurs
4041 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
4044 PH/03 Add -lutil to the default FreeBSD LIBS setting.
4046 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
4047 errors. Otherwise a message that provokes a temporary error (when other
4048 messages do not) can cause a whole host to time out.
4050 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
4051 addresses were routed directly to files or pipes from a redirect router.
4052 File deliveries just didn't batch; pipe deliveries might have suffered
4055 PH/06 A failure to get a lock for a hints database would erroneously always say
4056 "Failed to get write lock", even when it was really a read lock.
4058 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
4059 of 0600. This has been changed to use the value of the lockfile_mode
4060 option (which defaults to 0600).
4062 PH/08 Applied small patch from the Sieve maintainer.
4064 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
4065 folder from quota calculations, a direct delivery into this folder messed
4066 up the contents of the maildirsize file. This was because the regex was
4067 used only to exclude .Trash (or whatever) when the size of the mailbox
4068 was calculated. There was no check that a delivery was happening into an
4069 excluded directory. This bug has been fixed by ignoring all quota
4070 processing for deliveries into excluded directories.
4072 PH/10 Added the maildirfolder_create_regex option to appendfile.
4078 PH/01 The code for finding all the local interface addresses on a FreeBSD
4079 system running IPv6 was broken. This may well have applied to all BSD
4080 systems, as well as to others that have similar system calls. The broken
4081 code found IPv4 interfaces correctly, but gave incorrect values for the
4082 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
4083 that it would not match correctly against @[] and not recognize the IPv6
4086 PH/02 The ipliteral router was not recognizing addresses of the form user@
4087 [ipv6:....] because it didn't know about the "ipv6:" prefix.
4089 PH/03 Added disable_ipv6.
4091 PH/04 Changed $reply_address to use the raw form of the headers instead of the
4092 decoded form, because it is most often used to construct To: headers
4093 lines in autoreplies, and the decoded form may well be syntactically
4094 invalid. However, $reply_address has leading white space removed, and all
4095 newlines turned into spaces so that the autoreply transport does not
4098 PH/05 If group was specified without a user on a router, and no group or user
4099 was specified on a transport, the group from the router was ignored.
4101 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
4102 for visible compile-time settings that can be used to change these
4103 numbers, for those that want even more. Backwards compatibility with old
4104 spool files has been maintained. However, going back to a previous Exim
4105 release will lost any variables that are in spool files.
4107 PH/07 Two small changes when running in the test harness: increase delay when
4108 passing a TCP/IP connection to a new process, in case the original
4109 process has to generate a bounce, and remove special handling of
4110 127.0.0.2 (sic), which is no longer necessary.
4112 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
4113 be the same on different OS.
4115 PH/09 Moved a debug statement in filter processing to avoid a race problem when
4118 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
4119 whether --show-vars was specified or not
4121 JJ/02 exipick: Added support for new ACL variable spool format introduced
4124 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
4125 syntactically invalid From: or Reply-to: line, and a filter used this to
4126 generate an autoreply, and therefore failed to obtain an address for the
4127 autoreply, Exim could try to deliver to a non-existent relative file
4128 name, causing unrelated and misleading errors. What now happens is that
4129 it logs this as a hard delivery error, but does not attempt to create a
4132 PH/11 The exinext utility has a -C option for testing purposes, but although
4133 the given file was scanned by exinext itself; it wasn't being passed on
4134 when Exim was called.
4136 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
4137 an end-of-file indication when reading a command response.
4139 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
4140 compiled. In many other places in Exim, IPv6 addresses are always
4141 recognized, so I have changed this. It also means that IPv4 domain
4142 literals of the form [IPV4:n.n.n.n] are now always recognized.
4144 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
4145 used if the router is not running as root, for example, when verifying at
4146 ACL time, or when using -bh. The debugging output from this situation was
4147 non-existent - all you got was a failure to exec. I have made two
4150 (a) Failures to set uid/gid, the current directory, or a process leader
4151 in a subprocess such as that created by queryprogram now generate
4152 suitable debugging output when -d is set.
4154 (b) The queryprogram router detects when it is not running as root,
4155 outputs suitable debugging information if -d is set, and then runs
4156 the subprocess without attempting to change uid/gid.
4158 PH/15 Minor change to Makefile for building test_host (undocumented testing
4161 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
4162 additional section of a DNS packet that returns MX or SRV records.
4163 Instead, it always explicitly searches for A/AAAA records. This avoids
4164 major problems that occur when a DNS server includes only records of one
4165 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
4166 fixed another bug: if SRV records were looked up and the corresponding
4167 address records were *not* found in the additional section, the port
4168 values from the SRV records were lost.
4170 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
4171 using the correct key (the original address) when searching the retry
4172 rules in order to find which one to use for generating the retry hint.
4174 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
4175 adding the default one. Similarly, if it contains a Reply-To: header, the
4176 errors_reply_to option, if set, is not used.
4178 PH/19 When calculating a retry time, Exim used to measure the "time since
4179 failure" by looking at the "first failed" field in the retry record. Now
4180 it does not use this if it is later than than the arrival time of the
4181 message. Instead it uses the arrival time. This makes for better
4182 behaviour in cases where some deliveries succeed, thus re-setting the
4183 "first failed" field. An example is a quota failure for a huge message
4184 when small messages continue to be delivered. Without this change, the
4185 "time since failure" will always be short, possible causing more frequent
4186 delivery attempts for the huge message than are intended.
4187 [Note: This change was subsequently modified - see PH/04 for 4.62.]
4189 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
4190 $1, $2, $3) because the numerical variables can be reset during some
4191 expansion items (e.g. "match"), thereby losing the authentication data.
4193 PH/21 Make -bV show the size of off_t variables so that the test suite can
4194 decide whether to run tests for quotas > 2G.
4196 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
4197 mailbox_size, and mailbox_filecount in the appendfile transport. If a
4198 filecount value is greater than 2G or if a quota value is greater than 2G
4199 on a system where the size of off_t is not greater than 4, a panic error
4202 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
4203 never match. The debug and -bh output now contains an explicit error
4204 message indicating a malformed IPv4 address or mask.
4206 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
4207 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
4208 PH/23 above applies.
4210 PH/25 Do not write to syslog when running in the test harness. The only
4211 occasion when this arises is a failure to open the main or panic logs
4212 (for which there is an explicit test).
4214 PH/26 Added the /no_tell option to "control=freeze".
4216 PH/27 If a host name lookup failed very early in a connection, for example, if
4217 the IP address matched host_lookup and the reverse lookup yielded a name
4218 that did not have a forward lookup, an error message of the form "no IP
4219 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
4220 could be logged. Now it outputs the IP address instead of "NULL".
4222 PH/28 An enabling patch from MH: add new function child_open_exim2() which
4223 allows the sender and the authenticated sender to be set when
4224 submitting a message from within Exim. Since child_open_exim() is
4225 documented for local_scan(), the new function should be too.
4227 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
4228 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
4229 results in an empty string is now treated as unset.
4231 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
4233 PH/31 Added sender_verify_fail logging option.
4235 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
4236 needed by RFC 822 but not by RFC 2822 was commented out. I have now
4237 tidied the source and removed it altogether.
4239 PH/33 When a queue run was abandoned because the load average was too high, a
4240 log line was always written; now it is written only if the queue_run log
4241 selector is set. In addition, the log line for abandonment now contains
4242 information about the queue run such as the pid. This is always present
4243 in "start" and "stop" lines but was omitted from the "abandon" line.
4245 PH/34 Omit spaces between a header name and the colon in the error message that
4246 is given when verify = headers_syntax fails (if there are lots of them,
4247 the message gets confusing).
4249 PH/35 Change the default for dns_check_names_pattern to allow slashes within
4250 names, as there are now some PTR records that contain slashes. This check
4251 is only to protect against broken name servers that fall over on strange
4252 characters, so the fact that it applies to all lookups doesn't matter.
4254 PH/36 Now that the new test suite is complete, we can remove some of the
4255 special code in Exim that was needed for the old test suite. For example,
4256 sorting DNS records because real resolvers return them in an arbitrary
4257 order. The new test suite's fake resolver always returns records in the
4260 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
4261 bounces) except when queue_only is set, to avoid logging races between
4262 the different processes.
4264 PH/38 Panic-die if .include specifies a non-absolute path.
4266 PH/39 A tweak to the "H" retry rule from its user.
4268 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
4269 a label. They prevented compilation on older perls.
4271 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
4272 a warning to be raised on newish perls.
4274 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
4275 on queue. Changes to match documented behaviour of showing count of
4276 messages matching specified criteria.
4278 PH/40 Changed the default ident timeout from 30s to 5s.
4280 PH/41 Added support for the use of login_cap features, on those BSD systems
4281 that have them, for controlling the resources used by pipe deliveries.
4283 PH/42 The content-scanning code uses fopen() to create files in which to put
4284 message data. Previously it was not paying any attention to the mode of
4285 the files. Exim runs with umask(0) because the rest of the code creates
4286 files with open(), and sets the required mode explicitly. Thus, these
4287 files were ending up world-writeable. This was not a big issue, because,
4288 being within the spool directory, they were not world-accessible. I have
4289 created a function called modefopen, which takes an additional mode
4290 argument. It sets umask(777), creates the file, chmods it to the required
4291 mode, then resets the umask. All the relevant calls to fopen() in the
4292 content scanning code have been changed to use this function.
4294 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
4295 to 24 hours. This avoids potential overflow problems when processing G
4296 and H retry rules. I suspect nobody ever tinkers with this value.
4298 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
4300 PH/45 When the plaintext authenticator is running as a client, the server's
4301 challenges are checked to ensure they are valid base64 strings. By
4302 default, the authentication attempt is cancelled if an invalid string is
4303 received. Setting client_ignore_invalid_base64 true ignores these errors.
4304 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
4305 they are received. Thus, the responses can be made to depend on the
4306 challenges. If an invalid string is ignored, an empty string is placed in
4309 PH/46 Messages that are created by the autoreply transport now contains a
4310 References: header, in accordance with RFCs 2822 and 3834.
4312 PH/47 Added authenticated_sender_force to the smtp transport.
4314 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
4316 PH/49 Installed latest patch from the Sieve maintainer.
4318 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
4319 was also set, the appendfile transport was unnecessarily scanning a
4320 directory of message files (e.g. for maildir delivery) to find the count
4321 of files (along with the size), even though it did not need this
4322 information. It now does the scan only if it needs to find either the
4323 size of the count of files.
4325 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
4327 PH/52 Two bugs concerned with error handling when the smtp transport is
4330 (i) Exim was not creating retry information for temporary errors given
4331 for individual recipients after the DATA command when the smtp transport
4332 was used in LMTP mode. This meant that they could be retried too
4333 frequently, and not timed out correctly.
4335 (ii) Exim was setting the flag that allows error details to be returned
4336 for LMTP errors on RCPT commands, but not for LMTP errors for individual
4337 recipients that were returned after the DATA command.
4339 PH/53 This is related to PH/52, but is more general: for any failing address,
4340 when detailed error information was permitted to be returned to the
4341 sender, but the error was temporary, then after the final timeout, only
4342 "retry timeout exceeded" was returned. Now it returns the full error as
4343 well as "retry timeout exceeded".
4345 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
4346 do this, and (what is worse) MTAs that accept it.
4348 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
4349 will now be deprecated.
4351 PH/56 New os.c-cygwin from the Cygwin maintainer.
4353 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
4354 formats (previously only available in exim formats via -bpr, -bpru,
4355 and -bpra. Now also available in native and exiqgrep formats)
4357 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
4358 with very large, slow to parse queues
4360 JJ/08 exipick: added ! as generic prefix to negate any criteria format
4362 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
4364 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
4365 responses to authentication challenges, though it was showing the
4366 challenges; (ii) I've removed the CR characters from the debug output for
4369 PH/58 Allow for the insertion of a newline as well as a space when a string
4370 is turned into more than one encoded-word during RFC 2047 encoding. The
4371 Sieve code now uses this.
4373 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
4374 data_4xx, lost_connection, tls_required.
4376 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
4377 message was being sent as an SMTP response.
4379 PH/61 Add -l and -k options to exicyclog.
4381 PH/62 When verifying, if an address was redirected to one new address, so that
4382 verification continued, and the new address failed or deferred after
4383 having set something in $address_data, the value of $address_data was not
4384 passed back to the ACL. This was different to the case when no
4385 redirection occurred. The value is now passed back in both cases.
4387 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
4388 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
4389 use login_cap.h, so on its own it isn't the distinguishing feature. The
4390 new name refers directly to the setclassresources() function.
4392 PH/65 Added configuration files for NetBSD3.
4394 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
4396 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
4397 is preferred over IPv4.
4399 PH/68 The bounce_return_message and bounce_return_body options were not being
4400 honoured for bounces generated during the reception of non-SMTP messages.
4401 In particular, this applied to messages rejected by the ACL. This bug has
4402 been fixed. However, if bounce_return_message is true and bounce_return_
4403 body is false, the headers that are returned for a non-SMTP message
4404 include only those that have been read before the error was detected.
4405 (In the case of an ACL rejection, they have all been read.)
4407 PH/69 The HTML version of the specification is now built in a directory called
4408 spec_html instead of spec.html, because the latter looks like a path with
4409 a MIME-type, and this confuses some software.
4411 PH/70 Catch two compiler warnings in sieve.c.
4413 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
4414 function verify_get_ident() calls ip_connect() to connect a socket, but
4415 if the "connect()" function timed out, ip_connect() used to close the
4416 socket. However, verify_get_ident() also closes the socket later, and in
4417 between Exim writes to the log, which may get opened at this point. When
4418 the socket was closed in ip_connect(), the log could get the same file
4419 descriptor number as the socket. This naturally causes chaos. The fix is
4420 not to close the socket in ip_connect(); the socket should be closed by
4421 the function that creates it. There was only one place in the code where
4422 this was missing, in the iplookup router, which I don't think anybody now
4423 uses, but I've fixed it anyway.
4425 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
4426 well as to direct DNS lookups. Otherwise the handling of names in host
4427 lists is inconsistent and therefore confusing.
4433 PH/01 Two changes to the default runtime configuration:
4435 (1) Move the checks for relay_from_hosts and authenticated clients from
4436 after to before the (commented out) DNS black list checks.
4438 (2) Add control=submission to the relay_from_hosts and authenticated
4439 clients checks, on the grounds that messages accepted by these
4440 statements are most likely to be submissions.
4442 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
4444 (1) Generate an error if the third argument for the ${prvs expansion is
4447 (2) Treat a missing third argument of ${prvscheck as if it were an empty
4450 (3) Reset the variables that are obtained from the first argument of
4451 ${prvscheck and used in the second argument before leaving the code,
4452 because their memory is reclaimed, so using them afterwards may do
4455 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
4456 one (it's much easier than Tom thought :-).
4458 (5) Because of (4), we can now allow for the use of $prvscheck_result
4459 inside the third argument.
4461 PH/03 For some reason, the default setting of PATH when running a command from
4462 a pipe transport was just "/usr/bin". I have changed it to
4465 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
4466 anything to be listed in the output from -bV.
4468 PH/05 When a filter generated an autoreply, the entire To: header line was
4469 quoted in the delivery log line, like this:
4471 => >A.N.Other <ano@some.domain> <original@ddress> ...
4473 This has been changed so that it extracts the operative address. There
4474 may be more than one such address. If so, they are comma-separated, like
4477 => >ano@some.domain,ona@other.domain <original@ddress> ...
4479 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
4480 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
4481 not being looked up in the rDNS to get a host name, Exim was showing the
4482 IP address twice in Received: lines, even though the IP addresses were
4483 identical. For example:
4485 Received: from [1.2.3.4] (helo=[1.2.3.4])
4487 However, if the real host name was known, it was omitting the HELO data
4488 if it matched the actual IP address. This has been tidied up so that it
4489 doesn't show the same IP address twice.
4491 PH/07 When both +timestamp and +memory debugging was on, the value given by
4492 $tod_xxx expansions could be wrong, because the tod_stamp() function was
4493 called by the debug printing, thereby overwriting the timestamp buffer.
4494 Debugging no longer uses the tod_stamp() function when +timestamp is set.
4496 PH/08 When the original message was included in an autoreply transport, it
4497 always said "this is a copy of the message, including all the headers",
4498 even if body_only or headers_only was set. It now gives an appropriate
4501 PH/09 Applied a patch from the Sieve maintainer which:
4503 o fixes some comments
4504 o adds the (disabled) notify extension core
4505 o adds some debug output for the result of if/elsif tests
4506 o points to the current vacation draft in the documentation
4507 and documents the missing references header update
4511 o fixes a bug in processing the envelope test (when testing
4512 multiple envelope elements, the last element determined the
4515 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
4516 Electronic Mail") by including:
4518 Auto-submitted: auto-generated
4520 in the messages that it generates (bounce messages and others, such as
4521 warnings). In the case of bounce messages for non-SMTP messages, there was
4522 also a typo: it was using "Auto_submitted" (underscore instead of
4523 hyphen). Since every message generated by Exim is necessarily in response
4524 to another message, thes have all been changed to:
4526 Auto-Submitted: auto-replied
4528 in accordance with these statements in the RFC:
4530 The auto-replied keyword:
4532 - SHOULD be used on messages sent in direct response to another
4533 message by an automatic process,
4535 - MUST NOT be used on manually-generated messages,
4537 - MAY be used on Delivery Status Notifications (DSNs) and Message
4538 Disposition Notifications (MDNs),
4540 - MUST NOT be used on messages generated by automatic or periodic
4541 processes, except for messages which are automatic responses to
4544 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
4545 to the default Received: header definition.
4547 PH/12 Added log selector acl_warn_skipped (default on).
4549 PH/13 After a successful wildlsearch lookup, discard the values of numeric
4550 variables because (a) they are in the wrong storage pool and (b) even if
4551 they were copied, it wouldn't work properly because of the caching.
4553 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
4554 checking when decoding. Apparently there are clients that generate
4555 overlong encoded strings. Why am I not surprised?
4557 PH/15 If the first argument of "${if match_address" was not empty, but did not
4558 contain an "@" character, Exim crashed. Now it writes a panic log message
4559 and treats the condition as false.
4561 PH/16 In autoreply, treat an empty string for "once" the same as unset.
4563 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
4564 extension "envelope-auth". The code is finished and in agreement with
4565 other implementations, but there is no documentation so far and in fact,
4566 nobody wrote the draft yet. This extension is currently #undef'ed, thus
4567 not changing the active code.
4569 Print executed "if" and "elsif" statements when debugging is used. This
4570 helps a great deal to understand what a filter does.
4572 Document more things not specified clearly in RFC3028. I had all this
4573 sorted out, when out of a sudden new issues came to my mind. Oops."
4575 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
4578 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
4579 canonical form (as documented). However, after a host name lookup from
4580 the IP address, check_host() was doing a simple string comparison with
4581 addresses acquired from the DNS when checking that the found name did
4582 have the original IP as one of its addresses. Since any found IPv6
4583 addresses are likely to be in abbreviated form, the comparison could
4584 fail. Luckily, there already exists a function for doing the comparison
4585 by converting both addresses to binary, so now that is used instead of
4586 the text comparison.
4588 PH/20 There was another similar case to PH/19, when a complete host name was
4589 given in a host list; looking up its IP address could give an abbreviated
4590 form, whereas the current host's name might or might not be abbreviated.
4591 The same fix has been applied.
4597 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
4598 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
4601 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
4602 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
4604 PH/03 Typo: missing ".o" in src/pcre/Makefile.
4606 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
4607 header line, restrict the check to what is listed in RFCs 2369 and 2929.
4608 Also, for "Auto-Submitted", treat anything other than "no" as
4609 non-personal, in accordance with RFC 3834. (Previously it treated
4610 anything starting "auto-" as non-personal.)
4612 TF/01 The control=submission/name=... option had a problem with syntax
4613 errors if the name included a slash character. The /name= option
4614 now slurps the rest of the string, so it can include any characters
4615 but it must come last in the list of options (after /sender_retain
4618 PH/05 Some modifications to the interface to the fake nameserver for the new
4626 TK/01 Added the "success_on_redirect" address verification option. See
4627 NewStuff for rationale and an example.
4629 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
4631 PH/02 Patch to exigrep to allow it to work on syslog lines.
4633 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
4634 fread() to skip over the body file's header line, because in Cygwin the
4635 header line is locked and is inaccessible.
4637 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
4638 co-exist for some time) to make it clear that it is the Exim ID that is
4639 referenced, not the Message-ID: header line.
4641 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
4642 string_format() function, because snprintf() does not exist on all
4645 PH/06 The use of forbid_filter_existstest now also locks out the use of the
4646 ${stat: expansion item.
4648 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
4649 protocol synchronization error", to keep the pedants happy.
4651 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
4652 well as for IRIX systems, when gcc is being used. See the host.c source
4655 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
4657 PH/10 Named domain lists were not working if used in a queue_smtp_domains
4660 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
4661 transport and to the smtp transport in LMTP mode.
4663 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
4665 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
4666 run a filter in a subprocess. This could lead to confusion in subsequent
4667 lookups in the parent process. There should also be a search_tidyup() at
4668 the end of the subprocess.
4670 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
4671 only if the host matched helo_try_verify_hosts, which caused the
4672 verification to occur when the EHLO/HELO command was issued. The ACL just
4673 tested the remembered result. Now, if a previous verification attempt has
4674 not happened, "verify = helo" does it there and then.
4676 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
4678 TK/03 Fix log output including CR from clamd.
4680 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
4681 exist provoked a memory error which could cause a segfault.
4683 PH/15 Installed PCRE 6.2
4685 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
4687 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
4688 of the problem. Specifically, suggested +O2 rather than +O1 for the
4691 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
4693 PH/20 If a delivery was routed to a non-standard port by means of an SRV
4694 record, the port was not correctly logged when the outgoing_port log
4695 selector was set (it logged the transort's default port).
4697 PH/21 Added support for host-specific ports to manualroute, queryprogram,
4698 fallback_hosts, and "hosts" in the smtp transport.
4700 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
4701 host errors such as "Connection refused".
4703 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
4704 authentication with radiusclient 0.4.9:
4706 - Error returned from rc_read_config was caught wrongly
4707 - Username/password not passed on to radius server due to wrong length.
4709 The presumption is that some radiusclient API changes for 4.51/PH/17
4710 were not taken care of correctly. The code is still untested by me (my
4711 Linux distribution still has 0.3.2 of radiusclient), but it was
4712 contributed by a Radius user.
4714 PH/24 When doing a callout, the value of $domain wasn't set correctly when
4715 expanding the "port" option of the smtp transport.
4717 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
4718 while reading a MIME header. Thanks to Tom Hughes for a patch.
4720 PH/24 Include config.h inside local_scan.h so that configuration settings are
4723 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
4724 that in an ACL for RCPT (for example), you can examine exactly what was
4727 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
4728 commands, but it was not correctly comparing the address with the actual
4729 client host address. Thus, it would show the EHLO address in Received:
4730 header lines when this was not necessary.
4732 PH/27 Added the % operator to ${eval:}.
4734 PH/28 Exim tries to create and chdir to its spool directory when it starts;
4735 it should be ignoring failures (because with -C, for example, it has lost
4736 privilege). It wasn't ignoring creation failures other than "already
4739 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
4742 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
4743 because an input line was too long, either on its own, or by virtue of
4744 too many continuations, the temporary file was not being removed, and the
4745 return code was incorrect.
4747 PH/31 Missing "BOOL" in function definition in filtertest.c.
4749 PH/32 Applied Sieve patches from the maintainer.
4751 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
4753 PH/33 Added "verify = not_blind".
4755 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
4756 Local/Makefile (with some defaults set). These are used in built scripts
4757 such as exicyclog, but they have never been used in the exim_install
4758 script (though there are many overriding facilities there). I have
4759 arranged that the exim_install script now takes note of these two
4762 PH/35 Installed configuration files for Dragonfly.
4764 PH/36 When a locally submitted message by a trusted user did not contain a
4765 From: header, and the sender address was obtained from -f or from an SMTP
4766 MAIL command, and the trusted user did not use -F to supply a sender
4767 name, $originator_name was incorrectly used when constructing a From:
4768 header. Furthermore, $originator_name was used for submission mode
4769 messages from external hosts without From: headers in a similar way,
4770 which is clearly wrong.
4772 PH/37 Added control=suppress_local_fixups.
4774 PH/38 When log_selector = +received_sender was set, and the addition of the
4775 sender made the log line's construction buffer exactly full, or one byte
4776 less than full, an overflow happened when the terminating "\n" was
4779 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
4780 when the result of a list match is failure because a DNS lookup failed.
4782 PH/40 RM_COMMAND is now used in the building process.
4784 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
4785 the "build-* directories that it finds.
4787 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
4788 address in a domain literal was a prefix of an interface address.
4790 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
4791 when verifying a sender address, unless rewrite_headers is false.
4793 PH/44 Wrote a long comment about why errors_to addresses are verified as
4794 recipients, not senders.
4796 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
4797 the ratelimit ACL was added.
4799 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
4801 PH/46 Added extra information about PostgreSQL errors to the error string.
4803 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
4804 avoiding the need to install special zones in a real server. This is
4805 backwards compatible; if it can't find the fake resolver, it drops back.
4806 Thus, both old and new test suites can be run.
4808 TF/02 Added util/ratelimit.pl
4810 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
4811 clock is set back in time.
4813 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
4814 Candler <B.Candler@pobox.com>.
4816 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
4817 OK for addresses that are the result of redirections.
4819 PH/48 A number of further additions for the benefit of the new test suite,
4820 including a fake gethostbyname() that interfaces to the fake DNS resolver
4823 TF/06 The fix for widen_domains has also been applied to qualify_single and
4824 search_parents which are the other dnslookup options that can cause
4827 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
4830 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
4832 TF/07 Exim produced the error message "an SRV record indicated no SMTP
4833 service" if it encountered an MX record with an empty target hostname.
4834 The message is now "an MX or SRV record indicated no SMTP service".
4836 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
4837 if the DNS of the sending site is misconfigured. This is quite a
4838 common situation. This change restores the behaviour of treating a
4839 helo verification defer as a failure.
4841 PH/51 If self=fail was set on a router, the bounce message did not include the
4842 actual error message.
4848 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
4850 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
4851 command itself ended in error, the underlying message about the transport
4852 filter timeout was being overwritten with the pipe command error. Now the
4853 underlying error message should be appended to the second error message.
4855 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4857 PH/02 Reduce the amount of output that "make" produces by default. Full output
4858 can still be requested.
4860 PH/03 The warning log line about a condition test deferring for a "warn" verb
4861 was being output only once per connection, rather than after each
4862 occurrence (because it was using the same function as for successful
4863 "warn" verbs). This seems wrong, so I have changed it.
4865 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4866 it should not have, which might have caused a crash in the right
4867 circumstances, but probably never did.
4869 PH/04 Installed a modified version of Tony Finch's patch to make submission
4870 mode fix the return path as well as the Sender: header line, and to
4871 add a /name= option so that you can make the user's friendly name appear
4874 TF/03 Added the control = fakedefer ACL modifier.
4876 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4877 Mark Lowes for thorough testing.
4879 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4881 TK/03 Merged latest SRS patch from Miles Wilton.
4883 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4884 with the definition in sysexits.h (which is #included earlier).
4885 Fortunately, Exim does not actually use EX_OK. The code used to try to
4886 preserve the sysexits.h value, by assuming that macro definitions were
4887 scanned for macro replacements. I have been disabused of this notion,
4888 so now the code just undefines EX_OK before #including unistd.h.
4890 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4891 in the smtp transport. When a block could not be written in a single
4892 write() function, the timeout was being re-applied to each part-write.
4893 This seems wrong - if the receiver was accepting one byte at a time it
4894 would take for ever. The timeout is now adjusted when this happens. It
4895 doesn't have to be particularly precise.
4897 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4898 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4900 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4901 without a domain if the check to <postmaster@domain> fails.
4903 SC/01 Eximstats: added -xls and the ability to specify output files
4904 (patch written by Frank Heydlauf).
4906 SC/02 Eximstats: use FileHandles for outputting results.
4908 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4910 SC/04 Eximstats: fixed display of large numbers with -nvr option
4912 SC/05 Eximstats: fixed merging of reports with empty tables.
4914 SC/06 Eximstats: added the -include_original_destination flag
4916 SC/07 Eximstats: removed tabs and trailing whitespace.
4918 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4920 TK/06 MBOX spool code: Add real "From " MBOX separator line
4921 so the .eml file is really in mbox format (even though
4922 most programs do not really care). Patch from Alex Miller.
4924 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4925 The latter is generated from $received_to and is only set if the
4926 message has one envelope recipient. SA can use these headers,
4927 obviously out-of-the-box. Patch from Alex Miller.
4929 PH/08 The ${def test on a variable was returning false if the variable's
4930 value was "0", contrary to what the specification has always said!
4931 The result should be true unless the variable is empty.
4933 PH/09 The syntax error of a character other than { following "${if
4934 def:variable_name" (after optional whitespace) was not being diagnosed.
4935 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4936 accidental colon was present, for example, could give incorrect results.
4938 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4939 result is used (not including appendfile, where other changes are about
4942 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4943 This involved changing a lot of size variables from int to off_t. It
4944 should work with maildirs and everything.
4946 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4947 spamd dying while we are connected to it.
4949 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4952 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4953 function for BDB 4.3.
4955 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4957 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4958 the sources for PCRE have been re-arranged and more files are now
4961 PH/15 The code I had for printing potentially long long variables in PH/11
4962 above was not the best (it lost precision). The length of off_t variables
4963 is now inspected at build time, and an appropriate printing format (%ld
4964 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4965 to be "long long int" or "long int". This is needed for the internal
4966 formatting function string_vformat().
4968 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4969 the configuration file to be ":syslog", then the script "guesses" where
4970 the logs files are, rather than using the compiled in default. In our
4971 case the guess is not the same as the compiled default, so the script
4972 suddenly stopped working when I started to use syslog. The patch checks
4973 to see if log_file_path is "". If so, it attempts to read it from exim
4974 with no configuration file to get the compiled in version, before it
4975 falls back to the previous guessing code."
4977 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4978 implementing BATV in an Exim configuration. See NewStuff for the gory
4981 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4982 Makefile that are specific to HP-UX.
4984 PH/18 If the "use_postmaster" option was set for a recipient callout together
4985 with the "random" option, the postmaster address was used as the MAIL
4986 FROM address for the random test, but not for the subsequent recipient
4987 test. It is now used for both.
4989 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4990 patch removes a few documentation additions to RFC 3028, because the
4991 latest draft now contains them. It adds the new en;ascii-case comparator
4992 and a new error check for 8bit text in MIME parts. Comparator and
4993 require names are now matched exactly. I enabled the subaddress
4994 extension, but it is not well tested yet (read: it works for me)."
4996 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4997 rework some of the code of TK/09 above to avoid the hardwired use of
4998 "%lld" and "long long". Replaced the call to snprintf() with a call to
5001 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
5002 records point to non-existent hosts", "retry timeout exceeded", and
5003 "retry time not reached for any host after a long failure period".
5005 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
5006 experimental DomainKeys support:
5008 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
5009 (2) On an error such as an illegally used "control", the wrong name for
5010 the control was given.
5012 These problems did NOT occur unless DomainKeys support was compiled.
5014 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
5016 PH/24 Added ${if match_ip condition.
5018 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
5019 databases so that it will be absolutely obvious if a crash occurs in the
5020 DB library. This is a regular occurrence (often caused by mis-matched
5023 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
5024 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
5025 picked up on a user's system that detects such things. There doesn't seem
5026 to be a gcc warning option for this - only an attribute that has to be
5027 put on the function's prototype. It seems that in Fedora Core 4 they have
5028 set this on a number of new functions. No doubt there will be more in due
5031 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
5032 specify a transport. However, if an address that was verified by such a
5033 router was the subject of a callout, Exim crashed because it tried to
5034 read the rcpt_include_affixes from the non-existent transport. Now it
5035 just assumes that the setting of that option is false. This bug was
5036 introduced by 4.51/PH/31.
5038 PH/28 Changed -d+all to exclude +memory, because that information is very
5039 rarely of interest, but it makes the output a lot bigger. People tend to
5040 do -d+all out of habit.
5042 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
5043 code in os-type was giving problems when libc.so lives in lib64, like on
5046 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
5047 aren't the modern standard, and it seems that some systems' include files
5048 don't always have them. Exim was already checking for some of the newer
5049 ones like T_AAAA, and defining it itself. I've added checks for all the
5050 record types that Exim uses.
5052 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
5053 not automatically generating a new one, as it is supposed to. This
5054 prevented TLS from working. If the file did exist, but contained invalid
5055 data, a new version was generated, as expected. It was only the case of a
5056 non-existent file that was broken.
5058 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
5059 with a change in libdomainkeys > 0.64.
5061 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
5062 from DNS. If the selector record carries the flag, it now has
5063 precedence over the domain-wide flag.
5065 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
5067 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
5068 the use of a port name that isn't defined in /etc/services) occurred, the
5069 message was deferred as in a normal delivery, and thus remained on the
5070 spool, instead of being failed because of the mua_wrapper setting. This
5071 is now fixed, and I tidied up some of the mua_wrapper messages at the
5074 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
5075 the messages in a hash of arrays rather than using individual hashes.
5076 This is a bit cleaner and results in dramatic memory savings, albeit
5077 at a slight CPU cost.
5079 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
5080 as requested by Marc Sherman.
5082 SC/10 Eximstats: added histograms for user specified patterns as requested
5085 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
5087 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
5088 fopen() in the content-scanning modules that did not already have it.
5094 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
5095 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
5097 TK/02 Fix ACL "control" statement not being available in MIME ACL.
5099 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
5101 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
5102 to test Sieve filters that use "vacation".
5104 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
5105 that changes the way the GnuTLS parameters are stored in the cache file.
5106 The new format can be generated externally. For backward compatibility,
5107 if the data in the cache doesn't make sense, Exim assumes it has read an
5108 old-format file, and it generates new data and writes a new file. This
5109 means that you can't go back to an older release without removing the
5112 PH/03 A redirect router that has both "unseen" and "one_time" set does not
5113 work if there are any delivery delays because "one_time" forces the
5114 parent to be marked "delivered", so its unseen clone is never tried
5115 again. For this reason, Exim now forbids the simultaneous setting of
5118 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
5119 redirected to themselves ("homonym" addresses). Read the long ChangeLog
5120 entry if you want to know the details. The fix, however, neglected to
5121 consider the case when local delivery batching is involved. The test for
5122 "previously delivered" was not happening when checking to see if an
5123 address could be batched with a previous (undelivered) one; under
5124 certain circumstances this could lead to multiple deliveries to the same
5127 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
5128 in its include files, and this causes problems building Exim.
5130 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
5131 header_syntax) but Exim was just ignoring anything given after a slash.
5132 In particular, this caused confusion with an attempt to use "verify =
5133 reverse_host_lookup/defer_ok". An error is now given when options are
5134 supplied for verify items that do not have them. (Maybe reverse_host_
5135 lookup should have a defer_ok option, but that's a different point.)
5137 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
5138 defined by RFC 821) to 2048, because there were problems with some AUTH
5139 commands, and RFC 1869 says the size should be increased for extended
5140 SMTP commands that take arguments.
5142 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
5145 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
5146 "unknown" error; now it says that the functionality isn't in the binary.
5148 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
5149 an address' error message when a string expansion fails (syntax or
5150 whatever). Otherwise the password may appear in the log. Following change
5151 PH/42 below, there is no longer a chance of it appearing in a bounce
5154 PH/11 Installed exipick version 20050225.0 from John Jetmore.
5156 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
5157 of its addresses was ever tried. (Bugzilla bug #2.)
5159 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
5160 the result incorrectly in the debug output. (It correctly added a newline
5161 to what was transported.)
5163 TF/01 Added $received_time.
5165 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
5166 commented out examples of how to interface to a virus scanner and to
5167 SpamAssassin. Also added commented examples of av_scanner and
5168 spamd_address settings.
5170 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
5171 and controls are allowed in which ACLs. There were a couple of minor
5172 errors. Some of the entries in the conditions table (which is a table of
5173 where they are NOT allowed) were getting very unwieldy; rewrote them as a
5174 negation of where the condition IS allowed.
5176 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
5178 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
5179 header file does not have a version number, so I've had to invent a new
5180 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
5181 API. The code is untested by me (my Linux distribution still has 0.3.2 of
5182 radiusclient), but it was contributed by a Radius user.
5184 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
5185 files or directories, for OpenSSL.
5187 PH/19 When an Exim process that is running as root has to create an Exim log
5188 file, it does so in a subprocess that runs as exim:exim so as to get the
5189 ownership right at creation (otherwise, other Exim processes might see
5190 the file with the wrong ownership). There was no test for failure of this
5191 fork() call, which would lead to the process getting stuck as it waited
5192 for a non-existent subprocess. Forks do occasionally fail when resources
5193 run out. I reviewed all the other calls to fork(); they all seem to check
5196 PH/20 When checking for unexpected SMTP input at connect time (before writing
5197 the banner), Exim was not dealing correctly with a non-positive return
5198 from the read() function. If the client had disconnected by this time,
5199 the result was a log entry for a synchronization error with an empty
5200 string after "input=" when read() returned zero. If read() returned -1
5201 (an event I could not check), uninitialized data bytes were printed.
5202 There were reports of junk text (parts of files, etc) appearing after
5205 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
5207 PH/22 Added support for macro redefinition, and (re)definition in between
5208 driver and ACL definitions.
5210 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
5211 forgetting to use the resulting value; it was using the unexpanded value.
5213 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
5214 hadn't been configured. The fix is from Juergen Kreileder, who
5215 understands it better than I do:
5217 "Here's what I see happening with three configured cyrus_sasl
5218 authenticators configured (plain, login, cram-md5):
5220 On startup auth_cyrus_sasl_init() gets called for each of these.
5221 This means three calls to sasl_listmech() without a specified mech_list.
5222 => SASL tests which mechs of all available mechs actually work
5223 => three warnings about OTP not working
5224 => the returned list contains: plain, login, cram-md5, digest-md5, ...
5226 With the patch, sasl_listmech() also gets called three times. But now
5227 SASL's mech_list option is set to the server_mech specified in the the
5228 authenticator. Or in other words, the answer from sasl_listmech()
5229 gets limited to just the mech you're testing for (which is different
5231 => the return list contains just 'plain' or 'login', 'cram-md5' or
5232 nothing depending on the value of ob->server_mech.
5234 I've just tested the patch: Authentication still works fine,
5235 unavailable mechs specified in the exim configuration are still
5236 caught, and the auth.log warnings about OTP are gone."
5238 PH/25 When debugging is enabled, the contents of the command line are added
5239 to the debugging output, even when log_selector=+arguments is not
5242 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
5243 answer is "GNU", and only if the return is "GNU/something" is the answer
5246 PH/27 $acl_verify_message is now set immediately after the failure of a
5247 verification in an ACL, and so is available in subsequent modifiers. In
5248 particular, the message can be preserved by coding like this:
5250 warn !verify = sender
5251 set acl_m0 = $acl_verify_message
5253 Previously, $acl_verify_message was set only while expanding "message"
5254 and "log_message" when a very denied access.
5256 PH/28 Modified OS/os.c-Linux with
5258 -#ifndef OS_LOAD_AVERAGE
5259 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
5261 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
5262 nomenclature these days.)
5264 PH/29 Installed patch from the Sieve maintainer that adds the options
5265 sieve_useraddress and sieve_subaddress to the redirect router.
5267 PH/30 In these circumstances:
5268 . Two addresses routed to the same list of hosts;
5269 . First host does not offer TLS;
5270 . First host accepts first address;
5271 . First host gives temporary error to second address;
5272 . Second host offers TLS and a TLS session is established;
5273 . Second host accepts second address.
5274 Exim incorrectly logged both deliveries with the TLS parameters (cipher
5275 and peerdn, if requested) that were in fact used only for the second
5278 PH/31 When doing a callout as part of verifying an address, Exim was not paying
5279 attention to any local part prefix or suffix that was matched by the
5280 router that accepted the address. It now behaves in the same way as it
5281 does for delivery: the affixes are removed from the local part unless
5282 rcpt_include_affixes is set on the transport.
5284 PH/32 Add the sender address, as F=<...>, to the log line when logging a
5285 timeout during the DATA phase of an incoming message.
5287 PH/33 Sieve envelope tests were broken for match types other than :is. I have
5288 applied a patch sanctioned by the Sieve maintainer.
5290 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
5291 the uid or gid is negative. A case of a negative gid caused this to be
5292 noticed. The fix allows for either to be negative.
5294 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
5295 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
5298 PH/36 The change PH/12 above was broken. Fixed it.
5300 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
5301 the grounds that routing the same address twice would always produce the
5302 same answer. This might have been true once, but it is certainly no
5303 longer true now. Routing a child address may depend on the previous
5304 routing that produced that child. Some complicated redirection strategies
5305 went wrong when messages had multiple recipients, and made Exim's
5306 behaviour dependent on the order in which the addresses were given.
5308 I have moved the duplicate checking until after the routing is complete.
5309 Exim scans the addresses that are assigned to local and remote
5310 transports, and removes any duplicates. This means that more work will be
5311 done, as duplicates will always all be routed, but duplicates are
5312 presumably rare, so I don't expect this is of any significance.
5314 For deliveries to pipes, files, and autoreplies, the duplicate checking
5315 still happens during the routing process, since they are not going to be
5318 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
5319 It corrects a timeout issue with spamd. This is Ian's comment: "The
5320 background is that sometimes spamd either never reads data from a
5321 connection it has accepted, or it never writes response data. The exiscan
5322 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
5323 blindly assumes that writes won't block so it may never time out."
5325 PH/39 Allow G after quota size as well as K and M.
5327 PH/40 The value set for $authenticated_id in an authenticator may not contain
5328 binary zeroes or newlines because the value is written to log lines and
5329 to spool files. There was no check on this. Now the value is run through
5330 the string_printing() function so that such characters are converted to
5331 printable escape sequences.
5333 PH/41 $message_linecount is a new variable that contains the total number of
5334 lines in the message. Compare $body_linecount, which is the count for the
5337 PH/42 Exim no longer gives details of delivery errors for specific addresses in
5338 bounce and delay warning messages, except in certain special cases, which
5341 (a) An SMTP error message from a remote host;
5342 (b) A message specified in a :fail: redirection;
5343 (c) A message specified in a "fail" command in a system filter;
5344 (d) A message specified in a FAIL return from the queryprogram router;
5345 (e) A message specified by the cannot_route_message router option.
5347 In these cases only, Exim does include the error details in bounce and
5348 warning messages. There are also a few cases where bland messages such
5349 as "unrouteable address" or "local delivery error" are given.
5351 PH/43 $value is now also set for the "else" part of a ${run expansion.
5353 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
5354 being worked on, but at least Exim now implements the latest version to
5357 PH/45 In a pipe transport, although a timeout while waiting for the pipe
5358 process to complete was treated as a delivery failure, a timeout while
5359 writing the message to the pipe was logged, but erroneously treated as a
5360 successful delivery. Such timeouts include transport filter timeouts. For
5361 consistency with the overall process timeout, these timeouts are now
5362 treated as errors, giving rise to delivery failures by default. However,
5363 there is now a new Boolean option for the pipe transport called
5364 timeout_defer, which, if set TRUE, converts the failures into defers for
5365 both kinds of timeout. A transport filter timeout is now identified in
5368 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
5369 systems where "make" and "gmake" are different, calling "gmake" at top
5370 level broke things. I've arranged for the value of $(MAKE) to be passed
5371 from the Makefile to this script so that it can call the same version of
5375 A note about Exim versions 4.44 and 4.50
5376 ----------------------------------------
5378 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
5379 changes of various kinds. As a consequence, a big documentation update was
5380 needed. This delayed the release for rather longer than seemed good, especially
5381 in the light of a couple of (minor) security issues. Therefore, the changes
5382 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
5383 release. So 4.44 and 4.50 are in effect two different branches that both start
5386 I have left the 4.50 change log unchanged; it contains all the changes since
5387 4.43. The change log for 4.44 is below; many of its items are identical to
5388 those for 4.50. This seems to be the most sensible way to preserve the
5389 historical information.
5395 1. Minor wording change to the doc/README.SIEVE file.
5397 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
5398 computation of the current number of files was incorrect.
5400 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5401 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5404 4. Give more explanation in the error message when the command for a transport
5405 filter fails to execute.
5407 5. There are several places where Exim runs a non-Exim command in a
5408 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5409 was being done only for the command run by the queryprogram router. It is
5410 now done for all such subprocesses. The other cases are: ${run, transport
5411 filters, and the commands run by the lmtp and pipe transports.
5413 6. Added CONFIGURE_GROUP build-time option.
5415 7. Some older OS have a limit of 256 on the maximum number of file
5416 descriptors. Exim was using setrlimit() to set 1000 as a large value
5417 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5418 systems. I've change it so that if it can't get 1000, it tries for 256.
5420 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5421 was an oversight, and furthermore, ever since the addition of extra
5422 controls (e.g. 4.43/32), the checks on when to allow different forms of
5423 "control" were broken. There should now be diagnostics for all cases when a
5424 control that does not make sense is encountered.
5426 9. Added the /retain_sender option to "control=submission".
5428 10. $recipients is now available in the predata ACL (oversight).
5430 11. Tidy the search cache before the fork to do a delivery from a message
5431 received from the command line. Otherwise the child will trigger a lookup
5432 failure and thereby defer the delivery if it tries to use (for example) a
5433 cached ldap connection that the parent has called unbind on.
5435 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5436 of $address_data from the recipient verification was clobbered by the
5437 sender verification.
5439 13. The value of address_data from a sender verification is now available in
5440 $sender_address_data in subsequent conditions in the ACL statement.
5442 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
5444 15. Added a new option "connect=<time>" to callout options, to set a different
5447 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5448 was its contents. (It was OK if the option was not defined at all.)
5450 17. A "Completed" log line is now written for messages that are removed from
5451 the spool by the -Mrm option.
5453 18. New variables $sender_verify_failure and $recipient_verify_failure contain
5454 information about exactly what failed.
5456 19. Added -dd to debug only the daemon process.
5458 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
5459 handles timeouts, both on the server side and network timeouts. Renamed the
5460 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
5462 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
5463 It is now set to "smtps".
5465 22. $host_address is now set to the target address during the checking of
5466 ignore_target_hosts.
5468 23. When checking ignore_target_hosts for an ipliteral router, no host name was
5469 being passed; this would have caused $sender_host_name to have been used if
5470 matching the list had actually called for a host name (not very likely,
5471 since this list is usually IP addresses). A host name is now passed as
5474 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5475 code that specifies a non-restarting handler (typically sigaction() in
5476 modern systems) in an attempt to fix a rare and obscure crash bug.
5478 25. Narrowed the window for a race in the daemon that could cause it to ignore
5479 SIGCHLD signals. This is not a major problem, because they are used only to
5480 wake it up if nothing else does.
5482 26. A malformed maildirsize file could cause Exim to calculate negative values
5483 for the mailbox size or file count. Odd effects could occur as a result.
5484 The maildirsize information is now recalculated if the size or filecount
5487 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5488 support for a long time. Removed HAVE_SYS_VFS_H.
5490 28. Installed the latest version of exipick from John Jetmore.
5492 29. In an address list, if the pattern was not a regular expression, an empty
5493 subject address (from a bounce message) matched only if the pattern was an
5494 empty string. Non-empty patterns were not even tested. This was the wrong
5495 because it is perfectly reasonable to use an empty address as part of a
5496 database query. An empty address is now tested by patterns that are
5497 lookups. However, all the other forms of pattern expect the subject to
5498 contain a local part and a domain, and therefore, for them, an empty
5499 address still always fails if the pattern is not itself empty.
5501 30. Exim went into a mad DNS loop when attempting to do a callout where the
5502 host was specified on an smtp transport, and looking it up yielded more
5503 than one IP address.
5505 31. Re-factored the code for checking spool and log partition space into a
5506 function that finds that data and another that does the check. The former
5507 is then used to implement four new variables: $spool_space, $log_space,
5508 $spool_inodes, and $log_inodes.
5510 32. The RFC2047 encoding function was originally intended for short strings
5511 such as real names; it was not keeping to the 75-character limit for
5512 encoded words that the RFC imposes. It now respects the limit, and
5513 generates multiple encoded words if necessary. To be on the safe side, I
5514 have increased the buffer size for the ${rfc2047: expansion operator from
5517 33. It is now permitted to omit both strings after an "if" condition; if the
5518 condition is true, the result is "true". As before, when the second string
5519 is omitted, a false condition yields an empty string. This makes it less
5520 cumbersome to write custom ACL and router conditions.
5522 34. Failure to deliver a bounce message always caused it to be frozen, even if
5523 there was an errors_to setting on the router. The errors_to setting is now
5526 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5527 canonical form (fully expanded) before being placed in
5528 $sender_host_address.
5530 36. The table in the code that translates DNS record types into text (T_A to
5531 "A" for instance) was missing entries for NS and CNAME. It is just possible
5532 that this could have caused confusion if both these types were looked up
5533 for the same domain, because the text type is used as part of Exim's
5534 per-process caching. But the chance of anyone hitting this buglet seems
5537 37. The dnsdb lookup has been extended in a number of ways.
5539 (1) There is a new type, "zns", which walks up the domain tree until it
5540 finds some nameserver records. It should be used with care.
5542 (2) There is a new type, "mxh", which is like "mx" except that it returns
5543 just the host names, not the priorities.
5545 (3) It is now possible to give a list of domains (or IP addresses) to be
5546 looked up. The behaviour when one of the lookups defers can be
5547 controlled by a keyword.
5549 (4) It is now possible to specify the separator character for use when
5550 multiple records are returned.
5552 38. The dnslists ACL condition has been extended: it is now possible to supply
5553 a list of IP addresses and/or domains to be looked up in a particular DNS
5556 39. Added log_selector=+queue_time_overall.
5558 40. When running the queue in the test harness, wait just a tad after forking a
5559 delivery process, to get repeatability of debugging output.
5561 41. Include certificate and key file names in error message when GnuTLS fails
5562 to set them up, because the GnuTLS error message doesn't include the name
5563 of the failing file when there is a problem reading it.
5565 42. Allow both -bf and -bF in the same test run.
5567 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
5569 44. The "Exiscan patch" is now merged into the mainline Exim source.
5571 45. Sometimes the final signoff response after QUIT could fail to get
5572 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5573 < 0 before doing a fflush(). This bug looks as though it goes back to the
5574 introduction of TLS in release 3.20, but "sometimes" must have been rare
5575 because the tests only now provoked it.
5577 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
5578 (this can affect the format of dates).
5580 47. exim_tidydb, when checking for the continued existence of a message for
5581 which it has found a message-specific retry record, was not finding
5582 messages that were in split spool directories. Consequently, it was
5583 deleting retry records that should have stayed in existence.
5585 48. Steve fixed some bugs in eximstats.
5587 49. The SPA authentication driver was not abandoning authentication and moving
5588 on to the next authenticator when an expansion was forced to fail,
5589 contradicting the general specification for all authenticators. Instead it
5590 was generating a temporary error. It now behaves as specified.
5592 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
5593 (the order specifies the preference for clients). The order is now AES256,
5594 AES128, 3DES, ARCFOUR128.
5596 51. Small patch to Sieve code - explicitly set From: when generating an
5599 52. Exim crashed if a remote delivery caused a very long error message to be
5600 recorded - for instance if somebody sent an entire SpamAssassin report back
5601 as a large number of 550 error lines. This bug was coincidentally fixed by
5602 increasing the size of one of Exim's internal buffers (big_buffer) that
5603 happened as part of the Exiscan merge. However, to be on the safe side, I
5604 have made the code more robust (and fixed the comments that describe what
5607 53. Now that there can be additional text after "Completed" in log lines (if
5608 the queue_time_overall log selector is set), a one-byte patch to exigrep
5609 was needed to allow it to recognize "Completed" as not the last thing in
5612 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
5613 patch that reportedly fixes this has been added. I am not expert enough to
5614 create a test for it. This is what the patch creator wrote:
5616 "I found a little strange behaviour of ldap code when working with
5617 Windows 2003 AD Domain, where users was placed in more than one
5618 Organization Units. When I tried to give exim partial DN, the exit code
5619 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
5620 But simultaneously result of request was absolutely normal ldap result,
5621 so I produce this patch..."
5623 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
5624 REFERENCE, so I have modified the code to exclude the patch when that macro
5627 55. Some experimental protocols are using DNS PTR records for new purposes. The
5628 keys for these records are domain names, not reversed IP addresses. The
5629 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5630 leaves it alone. Component reversal etc. now happens only for IP addresses.
5633 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
5635 57. Double the size of the debug message buffer (to 2048) so that more of very
5636 long debug lines gets shown.
5638 58. The exicyclog utility now does better if the number of log files to keep
5639 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
5641 59. Two changes related to the smtp_active_hostname option:
5643 (1) $smtp_active_hostname is now available as a variable.
5644 (2) The default for smtp_banner uses $smtp_active_hostname instead
5645 of $primary_hostname.
5647 60. The host_aton() function is supposed to be passed a string that is known
5648 to be a valid IP address. However, in the case of IPv6 addresses, it was
5649 not checking this. This is a hostage to fortune. Exim now panics and dies
5650 if the condition is not met. A case was found where this could be provoked
5651 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5652 components; fortuitously, this particular loophole had already been fixed
5653 by change 4.50/55 above.
5655 If there are any other similar loopholes, the new check in host_aton()
5656 itself should stop them being exploited. The report I received stated that
5657 data on the command line could provoke the exploit when Exim was running as
5658 exim, but did not say which command line option was involved. All I could
5659 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5660 running as the user.
5663 61. There was a buffer overflow vulnerability in the SPA authentication code
5664 (which came originally from the Samba project). I have added a test to the
5665 spa_base64_to_bits() function which I hope fixes it.
5668 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
5669 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
5671 63. The daemon start-up calls getloadavg() while still root for those OS that
5672 need the first call to be done as root, but it missed one case: when
5673 deliver_queue_load_max is set with deliver_drop_privilege. This is
5674 necessary for the benefit of the queue runner, because there is no re-exec
5675 when deliver_drop_privilege is set.
5677 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
5678 This has been fixed.
5680 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5681 was in use, was not putting the data itself into the right store pool;
5682 consequently, it could be overwritten for a subsequent message in the same
5683 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5686 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
5688 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
5689 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
5690 the function were treating the return as a boolean value, which happened to
5691 work because 0=false and not-0=true, but is not correct code.
5693 68. The host_aton() function was not handling scoped IPv6 addresses (those
5694 with, for example, "%eth0" on the end) correctly.
5696 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
5697 negated items (that is, ~something) in unsigned ints. Some compilers
5698 apparently mutter when there is no cast.
5700 70. If an address verification called from an ACL failed, and did not produce a
5701 user-specific message (i.e. there was only a "system" message), nothing was
5702 put in $acl_verify_message. In this situation, it now puts the system
5705 71. Change 4.23/11 added synchronization checking at the start of an SMTP
5706 session; change 4.31/43 added the unwanted input to the log line - except
5707 that it did not do this in the start of session case. It now does.
5709 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
5710 This is wrong and can cause the other end to generate a synchronization
5711 error if it is another Exim or anything else that does the synchronization
5712 check. A QUIT command is no longer sent after a timeout.
5714 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
5715 during host lookups.
5717 74. The defer_ok option of callout verification was not working if it was used
5718 when verifying addresses in header lines, that is, for this case:
5720 verify = header_sender/callout=defer_ok
5722 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
5723 those file descriptors could be used for SMTP connections. If anything
5724 wrote to stderr (the example that came up was "warn" in embedded Perl), it
5725 could be sent to the SMTP client, causing chaos. The daemon now opens
5726 stdin, stdout, and stderr to /dev/null when it puts itself into the
5729 76. Arrange for output from Perl's "warn" command to be written to Exim's main
5730 log by default. The user can override this with suitable Perl magic.
5732 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
5733 the log message when discard triggers, was not working for the DATA ACL or
5734 for the non-SMTP ACL.
5736 78. Error message wording change in sieve.c.
5738 79. If smtp_accept_max_per_host was set, the number of connections could be
5739 restricted to fewer than expected, because the daemon was trying to set up
5740 a new connection before checking whether the processes handling previous
5741 connections had finished. The check for completed processes is now done
5742 earlier. On busy systems, this bug wouldn't be noticed because something
5743 else would have woken the daemon, and it would have reaped the completed
5746 80. If a message was submitted locally by a user whose login name contained one
5747 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
5748 It caused a spool format error. I have fixed the spool reading code. A
5749 related problem was that the "from" clause in the Received: line became
5750 illegal because of the space(s). It is now covered by ${quote_local_part.
5752 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
5755 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
5757 83. Patch from Sieve maintainer to fix unterminated string problem in
5758 "vacation" handling.
5760 84. Some minor changes to the Linux configuration files to help with other
5761 OS variants using glibc.
5763 85. One more patch for Sieve to update vacation handling to latest spec.
5766 ----------------------------------------------------
5767 See the note above about the 4.44 and 4.50 releases.
5768 ----------------------------------------------------
5774 1. Change 4.43/35 introduced a bug that caused file counts to be
5775 incorrectly computed when quota_filecount was set in an appendfile
5778 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5779 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5782 3. Give more explanation in the error message when the command for a transport
5783 filter fails to execute.
5785 4. There are several places where Exim runs a non-Exim command in a
5786 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5787 was being done only for the command run by the queryprogram router. It is
5788 now done for all such subprocesses. The other cases are: ${run, transport
5789 filters, and the commands run by the lmtp and pipe transports.
5791 5. Some older OS have a limit of 256 on the maximum number of file
5792 descriptors. Exim was using setrlimit() to set 1000 as a large value
5793 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5794 systems. I've change it so that if it can't get 1000, it tries for 256.
5796 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5797 was an oversight, and furthermore, ever since the addition of extra
5798 controls (e.g. 4.43/32), the checks on when to allow different forms of
5799 "control" were broken. There should now be diagnostics for all cases when a
5800 control that does not make sense is encountered.
5802 7. $recipients is now available in the predata ACL (oversight).
5804 8. Tidy the search cache before the fork to do a delivery from a message
5805 received from the command line. Otherwise the child will trigger a lookup
5806 failure and thereby defer the delivery if it tries to use (for example) a
5807 cached ldap connection that the parent has called unbind on.
5809 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5810 of $address_data from the recipient verification was clobbered by the
5811 sender verification.
5813 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5814 was its contents. (It was OK if the option was not defined at all.)
5816 11. A "Completed" log line is now written for messages that are removed from
5817 the spool by the -Mrm option.
5819 12. $host_address is now set to the target address during the checking of
5820 ignore_target_hosts.
5822 13. When checking ignore_target_hosts for an ipliteral router, no host name was
5823 being passed; this would have caused $sender_host_name to have been used if
5824 matching the list had actually called for a host name (not very likely,
5825 since this list is usually IP addresses). A host name is now passed as
5828 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5829 code that specifies a non-restarting handler (typically sigaction() in
5830 modern systems) in an attempt to fix a rare and obscure crash bug.
5832 15. Narrowed the window for a race in the daemon that could cause it to ignore
5833 SIGCHLD signals. This is not a major problem, because they are used only to
5834 wake it up if nothing else does.
5836 16. A malformed maildirsize file could cause Exim to calculate negative values
5837 for the mailbox size or file count. Odd effects could occur as a result.
5838 The maildirsize information is now recalculated if the size or filecount
5841 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5842 support for a long time. Removed HAVE_SYS_VFS_H.
5844 18. Updated exipick to current release from John Jetmore.
5846 19. Allow an empty sender to be matched against a lookup in an address list.
5847 Previously the only cases considered were a regular expression, or an
5850 20. Exim went into a mad DNS lookup loop when doing a callout where the
5851 host was specified on the transport, if the DNS lookup yielded more than
5854 21. The RFC2047 encoding function was originally intended for short strings
5855 such as real names; it was not keeping to the 75-character limit for
5856 encoded words that the RFC imposes. It now respects the limit, and
5857 generates multiple encoded words if necessary. To be on the safe side, I
5858 have increased the buffer size for the ${rfc2047: expansion operator from
5861 22. Failure to deliver a bounce message always caused it to be frozen, even if
5862 there was an errors_to setting on the router. The errors_to setting is now
5865 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5866 canonical form (fully expanded) before being placed in
5867 $sender_host_address.
5869 24. Updated eximstats to version 1.33
5871 25. Include certificate and key file names in error message when GnuTLS fails
5872 to set them up, because the GnuTLS error message doesn't include the name
5873 of the failing file when there is a problem reading it.
5875 26. Expand error message when OpenSSL has problems setting up cert/key files.
5878 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5879 (this can affect the format of dates).
5881 28. exim_tidydb, when checking for the continued existence of a message for
5882 which it has found a message-specific retry record, was not finding
5883 messages that were in split spool directories. Consequently, it was
5884 deleting retry records that should have stayed in existence.
5886 29. eximstats updated to version 1.35
5887 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5888 1.35 - bugfix such that pie charts by volume are generated correctly
5890 30. The SPA authentication driver was not abandoning authentication and moving
5891 on to the next authenticator when an expansion was forced to fail,
5892 contradicting the general specification for all authenticators. Instead it
5893 was generating a temporary error. It now behaves as specified.
5895 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5896 (the order specifies the preference for clients). The order is now AES256,
5897 AES128, 3DES, ARCFOUR128.
5899 31. Small patch to Sieve code - explicitly set From: when generating an
5902 32. Exim crashed if a remote delivery caused a very long error message to be
5903 recorded - for instance if somebody sent an entire SpamAssassin report back
5904 as a large number of 550 error lines. This bug was coincidentally fixed by
5905 increasing the size of one of Exim's internal buffers (big_buffer) that
5906 happened as part of the Exiscan merge. However, to be on the safe side, I
5907 have made the code more robust (and fixed the comments that describe what
5910 33. Some experimental protocols are using DNS PTR records for new purposes. The
5911 keys for these records are domain names, not reversed IP addresses. The
5912 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5913 leaves it alone. Component reversal etc. now happens only for IP addresses.
5916 34. The host_aton() function is supposed to be passed a string that is known
5917 to be a valid IP address. However, in the case of IPv6 addresses, it was
5918 not checking this. This is a hostage to fortune. Exim now panics and dies
5919 if the condition is not met. A case was found where this could be provoked
5920 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5921 components; fortuitously, this particular loophole had already been fixed
5922 by change 4.50/55 or 4.44/33 above.
5924 If there are any other similar loopholes, the new check in host_aton()
5925 itself should stop them being exploited. The report I received stated that
5926 data on the command line could provoke the exploit when Exim was running as
5927 exim, but did not say which command line option was involved. All I could
5928 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5929 running as the user.
5932 35. There was a buffer overflow vulnerability in the SPA authentication code
5933 (which came originally from the Samba project). I have added a test to the
5934 spa_base64_to_bits() function which I hope fixes it.
5937 36. The daemon start-up calls getloadavg() while still root for those OS that
5938 need the first call to be done as root, but it missed one case: when
5939 deliver_queue_load_max is set with deliver_drop_privilege. This is
5940 necessary for the benefit of the queue runner, because there is no re-exec
5941 when deliver_drop_privilege is set.
5943 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5944 was in use, was not putting the data itself into the right store pool;
5945 consequently, it could be overwritten for a subsequent message in the same
5946 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5949 38. Sometimes the final signoff response after QUIT could fail to get
5950 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5951 < 0 before doing a fflush(). This bug looks as though it goes back to the
5952 introduction of TLS in release 3.20, but "sometimes" must have been rare
5953 because the tests only now provoked it.
5959 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5960 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5961 Now it returns an integer. A number of calls were still expecting a T/F
5962 return. Fortuitously, in all cases, the tests worked in OK situations,
5963 which is the norm. However, things would have gone wrong on any write
5964 failures on the smtp file descriptor. This function is used when sending
5965 messages over SMTP and also when doing verify callouts.
5967 2. When Exim is called to do synchronous delivery of a locally submitted
5968 message (the -odf or -odi options), it no longer closes stderr before doing
5971 3. Implemented the mua_wrapper option.
5973 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5975 5. Implemented the functions header_remove(), header_testname(),
5976 header_add_at_position(), and receive_remove_recipient(), and exported them
5979 6. If an ACL "warn" statement specified the addition of headers, Exim already
5980 inserted X-ACL-Warn: at the start if there was no header name. However, it
5981 was not making this test for the second and subsequent header lines if
5982 there were newlines in the string. This meant that an invalid header could
5983 be inserted if Exim was badly configured.
5985 7. Allow an ACL "warn" statement to add header lines at the start or after all
5986 the Received: headers, as well as at the end.
5988 8. Added the rcpt_4xx retry error code.
5990 9. Added postmaster_mailfrom=xxx to callout verification option.
5992 10. Added mailfrom=xxxx to the callout verification option, for verify=
5995 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5996 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5998 12. Inserted some casts to stop certain compilers warning when using pointer
5999 differences as field lengths or precisions in printf-type calls (mostly
6000 affecting debugging statements).
6002 13. Added optional readline() support for -be (dynamically loaded).
6004 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
6005 same clock tick as a message's arrival, so that its received time was the
6006 same as the "first fail" time on the retry record, and that message
6007 remained on the queue past the ultimate address timeout, every queue runner
6008 would try a delivery (because it was past the ultimate address timeout) but
6009 after another failure, the ultimate address timeout, which should have then
6010 bounced the address, did not kick in. This was a "< instead of <=" error;
6011 in most cases the first failure would have been in the next clock tick
6012 after the received time, and all would be well.
6014 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
6015 being recognized when the domain list was tested by the match_domain
6016 condition in an expansion string.
6018 16. Added the ${str2b64: operator.
6020 17. Exim was always calling setrlimit() to set a large limit for the number of
6021 processes, without checking whether the existing limit was already
6022 adequate. (It did check for the limit on file descriptors.) Furthermore,
6023 errors from getrlimit() and setrlimit() were being ignored. Now they are
6024 logged to the main and panic logs, but Exim does carry on, to try to do its
6025 job under whatever limits there are.
6027 18. Imported PCRE 5.0.
6029 19. Trivial typo in log message " temporarily refused connection" (the leading
6032 20. If the log selector return_path_on_delivery was set and an address was
6033 redirected to /dev/null, the delivery process crashed because it assumed
6034 that a return path would always be set for a "successful" delivery. In this
6035 case, the whole delivery is bypassed as an optimization, and therefore no
6038 21. Internal re-arrangement: the function for sending a challenge and reading
6039 a response while authentication was assuming a zero-terminated challenge
6040 string. It's now changed to take a pointer and a length, to allow for
6041 binary data in such strings.
6043 22. Added the cyrus_sasl authenticator (code supplied by MBM).
6045 23. Exim was not respecting finduser_retries when seeking the login of the
6046 uid under which it was called; it was always trying 10 times. (The default
6047 setting of finduser_retries is zero.) Also, it was sleeping after the final
6048 failure, which is pointless.
6050 24. Implemented tls_on_connect_ports.
6052 25. Implemented acl_smtp_predata.
6054 26. If the domain in control=submission is set empty, Exim assumes that the
6055 authenticated id is a complete email address when it generates From: or
6056 Sender: header lines.
6058 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
6059 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
6060 chown and chgrp in /bin and hostname in /usr/bin.
6062 28. Exim was keeping the "process log" file open after each use, just as it
6063 does for the main log. This opens the possibility of it remaining open for
6064 long periods when the USR1 signal hits a daemon. Occasional processlog
6065 errors were reported, that could have been caused by this. Anyway, it seems
6066 much more sensible not to leave this file open at all, so that is what now
6069 29. The long-running daemon process does not normally write to the log once it
6070 has entered its main loop, and it closes the log before doing so. This is
6071 so that log files can straightforwardly be renamed and moved. However,
6072 there are a couple of unusual error situations where the daemon does write
6073 log entries, and I had neglected to close the log afterwards.
6075 30. The text of an SMTP error response that was received during a remote
6076 delivery was being truncated at 512 bytes. This is too short for some of
6077 the long messages that one sometimes sees. I've increased the limit to
6080 31. It is now possible to make retry rules that apply only when a message has a
6081 specific sender, in particular, an empty sender.
6083 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
6084 it possible to be selective about when SMTP synchronization is enforced.
6086 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
6088 32. Implemented hosts_connection_nolog.
6090 33. Added an ACL for QUIT.
6092 34. Setting "delay_warning=" to disable warnings was not working; it gave a
6095 35. Added mailbox_size and mailbox_filecount to appendfile.
6097 36. Added control = no_multiline_responses to ACLs.
6099 37. There was a bug in the logic of the code that waits for the clock to tick
6100 in the case where the clock went backwards by a substantial amount such
6101 that the microsecond fraction of "now" was more than the microsecond
6102 fraction of "then" (but the whole seconds number was less).
6104 38. Added support for the libradius Radius client library this is found on
6105 FreeBSD (previously only the radiusclient library was supported).
6111 1. When certain lookups returned multiple values in the form name=value, the
6112 quoting of the values was not always being done properly. Specifically:
6113 (a) If the value started with a double quote, but contained no whitespace,
6115 (b) If the value contained whitespace other than a space character (i.e.
6116 tabs or newlines or carriage returns) it was not quoted.
6117 This fix has been applied to the mysql and pgsql lookups by writing a
6118 separate quoting function and calling it from the lookup code. The fix
6119 should probably also be applied to nisplus, ibase and oracle lookups, but
6120 since I cannot test any of those, I have not disturbed their existing code.
6122 2. A hit in the callout cache for a specific address caused a log line with no
6123 reason for rejecting RCPT. Now it says "Previous (cached) callout
6124 verification failure".
6126 3. There was an off-by-one bug in the queryprogram router. An over-long
6127 return line was truncated at 256 instead of 255 characters, thereby
6128 overflowing its buffer with the terminating zero. As well as fixing this, I
6129 have increased the buffer size to 1024 (and made a note to document this).
6131 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
6132 when Exim is waiting for an SMTP response from a remote server, Exim
6133 restarts its select() call on the socket, thereby resetting its timeout.
6134 This is not a problem when such interrupts are rare. Somebody set up a cron
6135 job to run exiwhat every 2 minutes, which is less than the normal select()
6136 timeout (5 or 10 minutes). This meant that the select() timeout never
6137 kicked in because it was always reset. I have fixed this by comparing the
6138 time when an interrupt arrives with the time at the start of the first call
6139 to select(). If more time than the timeout has elapsed, the interrupt is
6140 treated as a timeout.
6142 5. Some internal re-factoring in preparation for the addition of Sieve
6143 extensions (by MH). In particular, the "personal" test is moved to a
6144 separate function, and given an option for scanning Cc: and Bcc: (which is
6145 not set for Exim filters).
6147 6. When Exim created an email address using the login of the caller as the
6148 local part (e.g. when creating a From: or Sender: header line), it was not
6149 quoting the local part when it contained special characters such as @.
6151 7. Installed new OpenBSD configuration files.
6153 8. Reworded some messages for syntax errors in "and" and "or" conditions to
6154 try to make them clearer.
6156 9. Callout options, other than the timeout value, were being ignored when
6157 verifying sender addresses in header lines. For example, when using
6159 verify = header_sender/callout=no_cache
6161 the cache was (incorrectly) being used.
6163 10. Added a missing instance of ${EXE} to the exim_install script; this affects
6164 only the Cygwin environment.
6166 11. When return_path_on_delivery was set as a log selector, if different remote
6167 addresses in the same message used different return paths and parallel
6168 remote delivery occurred, the wrong values would sometimes be logged.
6169 (Whenever a remote delivery process finished, the return path value from
6170 the most recently started remote delivery process was logged.)
6172 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
6173 lines when AUTH and/or TLS are in use. This is the "received protocol"
6174 field. Exim used to use "asmtp" for authenticated SMTP, without any
6175 indication (in the protocol name) for TLS use. Now it follows the RFC and
6176 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
6177 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
6178 names appear in log lines as well as in Received: header lines.
6180 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
6181 extensions, and comparison tests, and to fix some bugs.
6183 14. Changes to the "personal" filter test:
6185 (1) The test was buggy in that it was just doing the equivalent of
6186 "contains" tests on header lines. For example, if a user's address was
6187 anne@some.where, the "personal" test would incorrectly be true for
6189 To: susanne@some.where
6191 This test is now done by extracting each address from the header in turn,
6192 and checking the entire address. Other tests that are part of "personal"
6193 are now done using regular expressions (for example, to check local parts
6194 of addresses in From: header lines).
6196 (2) The list of non-personal local parts in From: addresses has been
6197 extended to include "listserv", "majordomo", "*-request", and "owner-*",
6198 taken from the Sieve specification recommendations.
6200 (3) If the message contains any header line starting with "List-" it is
6201 treated as non-personal.
6203 (4) The test for "circular" in the Subject: header line has been removed
6204 because it now seems ill-conceived.
6206 15. Minor typos in src/EDITME comments corrected.
6208 16. Installed latest exipick from John Jetmore.
6210 17. If headers_add on a router specified a text string that was too long for
6211 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
6212 of string_sprintf() is now avoided.
6214 18. $message_body_size was not set (it was always zero) when running the DATA
6215 ACL and the local_scan() function.
6217 19. For the "mail" command in an Exim filter, no default was being set for
6218 the once_repeat time, causing a random time value to be used if "once" was
6219 specified. (If the value happened to be <= 0, no repeat happened.) The
6220 default is now 0s, meaning "never repeat". The "vacation" command was OK
6221 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
6222 (I found it when inspecting the code).
6224 20. There is now an overall timeout for performing a callout verification. It
6225 defaults to 4 times the callout timeout, which applies to individual SMTP
6226 commands during the callout. The overall timeout applies when there is more
6227 than one host that can be tried. The timeout is checked before trying the
6228 next host. This prevents very long delays if there are a large number of
6229 hosts and all are timing out (e.g. when the network connections are timing
6230 out). The value of the overall timeout can be changed by specifying an
6231 additional sub-option for "callout", called "maxwait". For example:
6233 verify = sender/callout=5s,maxwait=20s
6235 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
6236 to the end before writing, but this should make it even safer).
6238 22. Exim was forgetting that it had advertised PIPELINING for the second and
6239 subsequent messages on an SMTP connection. It was also not resetting its
6240 memory on STARTTLS and an internal HELO.
6242 23. When Exim logs an SMTP synchronization error within a session, it now
6243 records whether PIPELINING has been advertised or not.
6245 24. Added 3 instances of "(long int)" casts to time_t variables that were being
6246 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
6247 rather than long int.
6249 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
6251 26. Added the never_mail option to autoreply.
6257 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
6258 crash if the getsockname() call failed; this can happen if a connection is
6259 closed very soon after it is established. The problem was simply in the
6260 order in which certain operations were done, causing Exim to try to write
6261 to the SMTP stream before it had set up the file descriptor. The bug has
6262 been fixed by making things happen in the correct order.
6268 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
6269 before the connection was closed, thus losing the rejection response.
6271 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
6272 some early Solaris releases, but causes trouble in current releases where
6273 socklen_t is defined.
6275 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
6278 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
6281 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
6282 that releasing the top of it at the end releases what was used for sub-
6283 expansions (unless the block got too big). However, discard this block if
6284 the first thing is a variable or header, so that we can use its block when
6285 it is dynamic (useful for very large $message_headers, for example).
6287 6. Lookups now cache *every* query, not just the most recent. A new, separate
6288 store pool is used for this. It can be recovered when all lookup caches are
6289 flushed. Lookups now release memory at the end of their result strings.
6290 This has involved some general refactoring of the lookup sources.
6292 7. Some code has been added to the store_xxx() functions to reduce the amount
6293 of flapping under certain conditions.
6295 8. log_incoming_interface used to affect only the <= reception log lines. Now
6296 it causes the local interface and port to be added to several more SMTP log
6297 lines, for example "SMTP connection from", and rejection lines.
6299 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
6301 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
6303 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
6304 could be overwritten at the end of the current message (or the start of a
6305 new message if it was set in a HELO ACL). The value is now preserved for
6306 the duration of the SMTP connection.
6308 12. If a transport had a headers_rewrite setting, and a matching header line
6309 contained an unqualified address, that address was qualified, even if it
6310 did not match any rewriting rules. The underlying bug was that the values
6311 of the flags that permit the existence of unqualified sender and recipient
6312 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
6313 non-local messages, and by -bnq for local messages) were not being
6314 preserved with the message after it was received.
6316 13. When Exim was logging an SMTP synchronization error, it could sometimes log
6317 "next input=" as part of the text comprising the host identity instead of
6318 the correct text. The code was using the same buffer for two different
6319 strings. However, depending on which order the printing function evaluated
6320 its arguments, the bug did not always show up. Under Linux, for example, my
6321 test suite worked just fine.
6323 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
6324 doesn't work with some older versions of Perl. It has been changed to "my",
6325 which in any case is probably the better facility to use.
6327 15. A really picky compiler found some instances of statements for creating
6328 error messages that either had too many or two few arguments for the format
6331 16. The size of the buffer for calls to the DNS resolver has been increased
6332 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
6333 for addresses that have a lot of PTR records. This alleviates a problem; it
6334 does not fully solve it.
6336 17. A dnsdb lookup for PTR records that receives more data than will fit in the
6337 buffer now truncates the list and logs the incident, which is the same
6338 action as happens when Exim is looking up a host name and its aliases.
6339 Previously in this situation something unpredictable would happen;
6340 sometimes it was "internal error: store_reset failed".
6342 18. If a server dropped the connection unexpectedly when an Exim client was
6343 using GnuTLS and trying to read a response, the client delivery process
6344 crashed while trying to generate an error log message.
6346 19. If a "warn" verb in an ACL added multiple headers to a message in a single
6347 string, for example:
6349 warn message = H1: something\nH2: something
6351 the text was added as a single header line from Exim's point of view
6352 though it ended up OK in the delivered message. However, searching for the
6353 second and subsequent header lines using $h_h2: did not work. This has been
6354 fixed. Similarly, if a system filter added multiple headers in this way,
6355 the routers could not see them.
6357 20. Expanded the error message when iplsearch is called with an invalid key to
6358 suggest using net-iplsearch in a host list.
6360 21. When running tests using -bh, any delays imposed by "delay" modifiers in
6361 ACLs are no longer actually imposed (and a message to that effect is
6364 22. If a "gecos" field in a passwd entry contained escaped characters, in
6365 particular, if it contained a \" sequence, Exim got it wrong when building
6366 a From: or a Sender: header from that name. A second bug also caused
6367 incorrect handling when an unquoted " was present following a character
6368 that needed quoting.
6370 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
6371 was not being matched caselessly.
6373 24. Arranged for all hyphens in the exim.8 source to be escaped with
6376 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
6377 properly. Recipient callout cache records were still being keyed to include
6378 the sender, even when use_sender was set false. This led to far more
6379 callouts that were necessary. The sender is no longer included in the key
6380 when use_sender is false.
6382 26. Added "control = submission" modifier to ACLs.
6384 27. Added the ${base62d: operator to decode base 62 numbers.
6386 28. dnsdb lookups can now access SRV records.
6388 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
6389 the configuration file.
6391 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
6392 (-v) mode. This makes the output for a verbose queue run more intelligible.
6394 31. Added a use_postmaster feature to recipient callouts.
6396 32. Added the $body_zerocount variable, containing the number of binary zero
6397 bytes in the message body.
6399 33. The time of last modification of the "new" subdirectory is now used as the
6400 "mailbox time last read" when there is a quota error for a maildir
6403 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
6405 35. Added +ignore_unknown as a special item in host lists.
6407 36. Code for decoding IPv6 addresses in host lists is now included, even if
6408 IPv6 support is not being compiled. This fixes a bug in which an IPv6
6409 address was recognized as an IP address, but was then not correctly decoded
6410 into binary, causing unexpected and incorrect effects when compared with
6417 1. Very minor rewording of debugging text in manualroute to say "list of
6418 hosts" instead of "hostlist".
6420 2. If verify=header_syntax was set, and a header line with an unqualified
6421 address (no domain) and a large number of spaces between the end of the
6422 name and the colon was received, the reception process suffered a buffer
6423 overflow, and (when I tested it) crashed. This was caused by some obsolete
6424 code that should have been removed. The fix is to remove it!
6426 3. When running in the test harness, delay a bit after writing a bounce
6427 message to get a bit more predictability in the log output.
6429 4. Added a call to search_tidyup() just before forking a reception process. In
6430 theory, someone could use a lookup in the expansion of smtp_accept_max_
6431 per_host which, without the tidyup, could leave open a database connection.
6433 5. Added the variables $recipient_data and $sender_data which get set from a
6434 lookup success in an ACL "recipients" or "senders" condition, or a router
6435 "senders" option, similar to $domain_data and $local_part_data.
6437 6. Moved the writing of debug_print from before to after the "senders" test
6440 7. Change 4.31/66 (moving the time when the Received: is generated) caused
6441 problems for message scanning, either using a data ACL, or using
6442 local_scan() because the Received: header was not generated till after they
6443 were called (in order to set the time as the time of reception completion).
6444 I have revised the way this works. The header is now generated after the
6445 body is received, but before the ACL or local_scan() are called. After they
6446 are run, the timestamp in the header is updated.
6452 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
6453 before starting a queue runner without re-exec. This happened only when
6454 deliver_drop_privilege was set or when the Exim user was set to root. The
6455 effect of the bug was that timeouts during subsequent deliveries caused
6456 crashes instead of being properly handled. The handler is now left at its
6457 default (and expected) setting.
6459 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
6460 message, again when deliver_drop_privilege is set or Exim is run as root.
6461 The bug described in (1) was not present in this case, but the tidying up
6462 of the other signals was missing. I have made the two cases consistent.
6464 3. The ignore_target_hosts setting on a manualroute router was being ignored
6465 for hosts that were looked up using the /MX notation.
6467 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
6470 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
6471 operated on the sender address. After changing the $sender_address to <>
6472 for the sender address verify, Exim was re-instated it as the original
6473 (before rewriting) address, but remembering that it had rewritten it, so it
6474 wasn't rewriting it again. This bug also had the effect of breaking the
6475 sender address verification caching when the sender address was rewritten.
6477 6. The ignore_target_hosts option was being ignored by the ipliteral router.
6478 This has been changed so that if the ip literal address matches
6479 ignore_target_hosts, the router declines.
6481 7. Added expansion conditions match_domain, match_address, and match_local_
6482 part (NOT match_host).
6484 8. The placeholder for the Received: header didn't have a length field set.
6486 9. Added code to Exim itself and to exim_lock to test for a specific race
6487 condition that could lead to file corruption when using MBX delivery. The
6488 issue is with the lockfile that is created in /tmp. If this file is removed
6489 after a process has opened it but before that process has acquired a lock,
6490 there is the potential for a second process to recreate the file and also
6491 acquire a lock. This could lead to two Exim processes writing to the file
6492 at the same time. The added code performs the same test as UW imapd; it
6493 checks after acquiring the lock that its file descriptor still refers to
6494 the same named file.
6496 10. The buffer for building added header lines was of fixed size, 8192 bytes.
6497 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
6500 11. Added the smtp_active_hostname option. If used, this will typically be made
6501 to depend on the incoming interface address. Because $interface_address is
6502 not set up until the daemon has forked a reception process, error responses
6503 that can happen earlier (such as "too many connections") no longer contain
6506 12. If an expansion in a condition on a "warn" statement fails because a lookup
6507 defers, the "warn" statement is abandoned, and the next ACL statement is
6508 processed. Previously this caused the whole ACL to be aborted.
6510 13. Added the iplsearch lookup type.
6512 14. Added ident_timeout as a log selector.
6514 15. Added tls_certificate_verified as a log selector.
6516 16. Added a global option tls_require_ciphers (compare the smtp transport
6517 option of the same name). This controls incoming TLS connections.
6519 17. I finally figured out how to make tls_require_ciphers do a similar thing
6520 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
6521 before starting the TLS session.
6523 18. Tabs are now shown as \t in -bP output.
6525 19. If the log selector return_path_on_delivery was set, Exim crashed when
6526 bouncing a message because it had too many Received: header lines.
6528 20. If two routers both had headers_remove settings, and the first one included
6529 a superfluous trailing colon, the final name in the first list and the
6530 first name in the second list were incorrectly joined into one item (with a
6531 colon in the middle).
6537 1. Added -C and -D options to the exinext utility, mainly to make it easier
6538 to include in the automated testing, but these could be helpful when
6539 multiple configurations are in use.
6541 2. The exinext utility was not formatting the output nicely when there was
6542 an alternate port involved in the retry record key, nor when there was a
6543 message id as well (for retries that were specific to a specific message
6544 and a specific host). It was also confused by IPv6 addresses, because of
6545 the additional colons they contain. I have fixed the IPv4 problem, and
6546 patched it up to do a reasonable job for IPv6.
6548 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
6549 delivery, the log line now contains "pipelined" if PIPELINING was used.
6551 4. An SMTP transport process used to panic and die if the bind() call to set
6552 an explicit outgoing interface failed. This has been changed; it is now
6553 treated in the same way as a connect() failure.
6555 5. A reference to $sender_host_name in the part of a conditional expansion
6556 that was being skipped was still causing a DNS lookup. This no longer
6559 6. The def: expansion condition was not recognizing references to header lines
6560 that used bh_ and bheader_.
6562 7. Added the _cache feature to named lists.
6564 8. The code for checking quota_filecount in the appendfile transport was
6565 allowing one more file than it should have been.
6567 9. For compatibility with Sendmail, the command line option
6575 and sets the incoming protocol and host name (for trusted callers). The
6576 host name and its colon can be omitted when only the protocol is to be set.
6577 Note the Exim already has two private options, -pd and -ps, that refer to
6578 embedded Perl. It is therefore impossible to set a protocol value of "d" or
6579 "s", but I don't think that's a major issue.
6581 10. A number of refactoring changes to the code, none of which should affect
6584 (a) The number of logging options was getting close to filling up the
6585 32-bit word that was used as a bit map. I have split them into two classes:
6586 those that are passed in the argument to log_write(), and those that are
6587 only ever tested independently outside of that function. These are now in
6588 separate 32-bit words, so there is plenty of room for expansion again.
6589 There is no change in the user interface or the logging behaviour.
6591 (b) When building, for example, log lines, the code previously used a
6592 macro that called string_cat() twice, in order to add two strings. This is
6593 not really sufficiently general. Furthermore, there was one instance where
6594 it was actually wrong because one of the argument was used twice, and in
6595 one call a function was used. (As it happened, calling the function twice
6596 did not affect the overall behaviour.) The macro has been replaced by a
6597 function that can join an arbitrary number of extra strings onto a growing
6600 (c) The code for expansion conditions now uses a table and a binary chop
6601 instead of a serial search (which was left over from when there were very
6602 few conditions). Also, it now recognizes conditions like "pam" even when
6603 the relevant support is not compiled in: a suitably worded error message is
6604 given if an attempt is made to use such a condition.
6606 11. Added ${time_interval:xxxxx}.
6608 12. A bug was causing one of the ddress fields not to be passed back correctly
6609 from remote delivery subprocesses. The field in question was not being
6610 subsequently used, so this caused to problems in practice.
6612 13. Added new log selectors queue_time and deliver_time.
6614 14. Might have fixed a bug in maildirsizefile handling that threw up
6615 "unexpected character" debug warnings, and recalculated the data
6616 unnecessarily. In any case, I expanded the warning message to give more
6619 15. Added the message "Restricted characters in address" to the statements in
6620 the default ACL that block characters like @ and % in local parts.
6622 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
6623 Three changes have been made:
6625 (a) There was a serious bug; a negative response to MAIL caused the whole
6626 recipient domain to be cached as invalid, thereby blocking all messages
6627 to all local parts at the same domain, from all senders. This bug has
6628 been fixed. The domain is no longer cached after a negative response to
6629 MAIL if the sender used is not empty.
6631 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
6634 (c) A new callout option, "use_sender" has been added for people who want
6635 the modified behaviour.
6641 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
6644 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
6645 indeed breaks things for older releases.
6647 3. Added additional logging to the case where there is a problem reading data
6648 from a filter that is running in a subprocess using a pipe, in order to
6649 try to track down a specific problem.
6651 4. Testing facility fudge: when running in the test harness and attempting
6652 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
6653 getting "No route to host". Convert this to a timeout.
6655 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
6658 6. Some OS don't have socklen_t but use size_t instead. This affects the
6659 fifth argument of getsockopt() amongst other things. This is now
6660 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
6661 can be set for individual OS. I have set it for SunOS5, OSF1, and
6662 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
6663 some earlier ones do not.
6665 7. Change 4.30/15 was not doing the test caselessly.
6667 8. The standard form for an IPv6 address literal was being rejected by address
6668 parsing in, for example, MAIL and RCPT commands. An example of this kind of
6669 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6670 this, as well as the form without the "IPv6" on the front (but only when
6671 address literals are enabled, of course).
6673 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
6675 10. Exim crashed if a message with an empty sender address specified by -f
6676 encountered a router with an errors_to setting. This could be provoked only
6677 by a command such as
6681 where an empty string was supplied; "<>" did not hit this bug.
6683 11. Installed PCRE release 4.5.
6685 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
6686 remained set. It is now erased.
6688 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
6689 times from message ids (which are base 36 rather than the normal 62).
6691 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
6692 were being counted as actual protocol errors, and logged if the log
6693 selector +smtp_protocol_error was set. One cannot be perfect in this test,
6694 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
6695 and DATA following a set of rejected RCPTs do not count as protocol errors.
6696 In other words, Exim assumes they were pipelined, though this may not
6697 actually be the case. Of course, in all cases the client gets an
6698 appropriate error code.
6700 15. If a lookup fails in an ACL condition, a message about the failure may
6701 be available; it is used if testing the ACL cannot continue, because most
6702 such messages specify what the cause of the deferral is. However, some
6703 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
6704 that caused an old message to be retained and used if a later statement
6705 caused a defer, replacing the real cause of the deferral.
6707 16. If an IP address had so many PTR records that the DNS lookup buffer
6708 was not large enough to hold them, Exim could crash while trying to process
6709 the truncated data. It now detects and logs this case.
6711 17. Further to 4.21/58, another change has been made: if (and only if) the
6712 first line of a message (the first header line) ends with CRLF, a bare LF
6713 in a subsequent header line has a space inserted after it, so as not to
6714 terminate the header.
6716 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
6717 unnecessarily, used atoi() instead of strtol(), and didn't check the
6718 termination when getting file sizes from file names by regex.
6720 19. Completely re-implemented the support for maildirsize files, in the light
6721 of a number of problems with the previous contributed implementation
6722 (4.30/29). In particular:
6724 . If the quota is zero, the maildirsize file is maintained, but no quota is
6727 . If the maildir directory does not exist, it is created before any attempt
6728 to write a maildirsize file.
6730 . The quota value in the file is just a cache; if the quota is changed in
6731 the transport, the new value overrides.
6733 . A regular expression is available for excluding directories from the
6736 20. The autoreply transport checks the characters in options that define the
6737 message's headers; it allows continued headers, but it was checking with
6738 isspace() after an embedded newline instead of explicitly looking for a
6741 21. If all the "regular" hosts to which an address was routed had passed their
6742 expiry times, and had not reached their retry times, the address was
6743 bounced, even if fallback hosts were defined. Now Exim should go on to try
6746 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
6747 equivalent code in the SMTP transport. Some hosts send humungous responses
6748 to HELO/EHLO, more than 1024 it seems.
6750 23. Refactoring: code in filter.c used (void *) for "any old type" but this
6751 gives compiler warnings in some environments. I've now done it "properly",
6754 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
6755 (because of problems with the built-in one) was declared to return uschar *
6756 instead of char *, causing compiler failure.
6758 25. Fixed a file descriptor leak when processing alias/forward files.
6760 26. Fixed a minor format string issue in dbfn.c.
6762 27. Typo in exim.c: ("dmbnz" for "dbmnz").
6764 28. If a filter file refered to $h_xxx or $message_headers, and the headers
6765 contained RFC 2047 "words", Exim's memory could, under certain conditions,
6768 29. When a sender address is verified, it is cached, to save repeating the test
6769 when there is more than one recipient in a message. However, when the
6770 verification involves a callout, it is possible for different callout
6771 options to be set for different recipients. It is too complicated to keep
6772 track of this in the cache, so now Exim always runs a verification when a
6773 callout is required, relying on the callout cache for the optimization.
6774 The overhead is duplication of the address routing, but this should not be
6777 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
6778 to be verified with callout=postmaster, and the main callout worked but the
6779 postmaster check failed, the verification correctly failed. However, if a
6780 subsequent RCPT command asked for sender verification *without* the
6781 postmaster check, incorrect caching caused this verification also to fail,
6784 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
6785 it was not caching the DNS options (qualify_single, search_parents) that
6786 were used when the lookup failed. A subsequent lookup with different
6787 options therefore always gave the same answer, though there were cases
6788 where it should not have. (Example: a "domains = !$mx_any" option on a
6789 dnslookup router: the "domains" option is always processed without any
6790 widening, but the router might have qualify_single set.) Now Exim uses the
6791 cached value only when the same options are set.
6793 32. Added John Jetmore's "exipick" utility to the distribution.
6795 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
6796 than a timeout (e.g. a certificate is required, and is not provided), an
6797 Exim server now closes the connection immediately. Previously it waited for
6798 the client to close - but if the client is SSL, it seems that they each
6799 wait for each other, leading to a delay before one of them times out.
6801 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
6802 maintained 0.8.x compatibility because I don't think many are using it, and
6803 it is clearly obsolete.
6805 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
6808 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
6809 client certificate was expired. A simple patch fixes this, though I don't
6810 understand the full logic of why the verify callback is called multiple
6813 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
6814 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
6815 which causes problems with some clients (such as the Certicom SSL Plus
6816 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
6817 disables the coutermeasure allowing Eudora to connect."
6819 38. Exim was not checking that a write() to a log file succeeded. This could
6820 lead to Bad Things if a log got too big, in particular if it hit a file
6821 size limit. Exim now panics and dies if it cannot write to a log file, just
6822 as it does if it cannot open a log file.
6824 39. Modified OS/Makefile-Linux so that it now contains
6826 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
6828 The two -D definitions ensure that Exim is compiled with large file
6829 support, which makes it possible to handle log files that are bigger than
6832 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
6833 instance) a domain was checked against a named list that involved a lookup,
6834 causing $domain_data to be set, then another domain was checked against the
6835 same list, then the first domain was re-checked, the value of $domain_data
6836 after the final check could be wrong. In particular, if the second check
6837 failed, it could be set empty. This bug probably also applied to
6840 41. The strip_trailing_dot option was not being applied to the address given
6841 with the -f command-line option.
6843 42. The code for reading a message's header from the spool was incrementing
6844 $received_count, but never initializing it. This meant that the value was
6845 incorrect (doubled) while delivering a message in the same process in which
6846 it was received. In the most common configuration of Exim, this never
6847 happens - a fresh exec is done - but it can happen when
6848 deliver_drop_privilege is set.
6850 43. When Exim logs an SMTP synchronization error - client data sent too soon -
6851 it now includes up to 150 characters of the unexpected data in the log
6854 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6855 and building data strings. The size of both of these buffers was 10 000
6856 bytes - far larger than anybody would *ever* want, thought I. Needless to
6857 say, somebody hit the limit. I have increased the maximum line length to
6858 20 000 and the maximum data length of concatenated lines to 100 000. I have
6859 also fixed two bugs, because there was no checking on these buffers. Tsk,
6860 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6861 buffer is too small.
6863 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6864 lsearch lookups. Now it does.
6866 46. When parsing a route_list item in a manualroute router, a fixed-length
6867 buffer was used for the list of hosts. I made this 1024 bytes long,
6868 thinking that nobody would ever have a list of hosts that long. Wrong.
6869 Somebody had a whole pile of complicated expansion conditions, and the
6870 string was silently truncated, leading to an expansion error. It turns out
6871 that it is easier to change to an unlimited length (owing to other changes
6872 that have happened since this code was originally written) than to build
6873 structure for giving a limitation error. The length of the item that
6874 expands into the list of hosts is now unlimited.
6876 47. The lsearch lookup could not handle data where the length of text line was
6877 more than 4095 characters. Such lines were truncated, leading to shortened
6878 data being returned. It should now handle lines of any length.
6880 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6881 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6884 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6885 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6886 no longer happens in comment lines. A list of replacements is now placed
6887 at the head of all of the source files, except those whose only change is
6888 to replace PERL_COMMAND in the very first #! line.
6890 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6891 messages on the queue, with a bottom-up merge sort, using code contributed
6892 by Michael Haardt. This should make operations like -bp somewhat faster on
6893 large queues. It won't affect queue runners, except when queue_run_in_order
6896 51. Installed eximstats 1.31 in the distribution.
6898 52. Added support for SRV lookups to the dnslookup router.
6900 53. If an ACL referred to $message_body or $message_body_end, the value was not
6901 reset for any messages that followed in the same SMTP session.
6903 54. The store-handling optimization for building very long strings was not
6904 differentiating between the different store pools. I don't think this
6905 actually made any difference in practice, but I've tidied it.
6907 55. While running the routers to verify a sender address, $sender_address
6908 was still set to the sender address. This is wrong, because when routing to
6909 send a bounce to the sender, it would be empty. Therefore, I have changed
6910 it so that, while verifying a sender address, $sender_address is set to <>.
6911 (There is no change to what happens when verifying a recipient address.)
6913 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6914 target A or AAAA records (if not already returned) without resetting the
6915 qualify_single or search_parents options of the DNS resolver. These are
6916 inappropriate in this case because the targets of MX and SRV records must
6917 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6918 target that, when qualified, matched something in the local domain. These
6919 two options are now turned off when doing these lookups.
6921 57. It seems that at least some releases of Reiserfs (which does not have the
6922 concept of a fixed number of inodes) returns zero and not -1 for the
6923 number of available inodes. This interacted badly with check_spool_inodes,
6924 which assumed that -1 was the "no such thing" setting. What I have done is
6925 to check that the total number of inodes is greater than zero before doing
6926 the test of how many are available.
6928 58. When a "warn" ACL statement has a log_message modifier, the message is
6929 remembered, and not repeated. This is to avoid a lot of repetition when a
6930 message has many recipients that cause the same warning to be written.
6931 However, Exim was preserving the list of already written lines for an
6932 entire SMTP session, which doesn't seem right. The memory is now reset if a
6933 new message is started.
6935 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6936 debugging output unless log_rewrite was also set.
6938 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6939 of (int)(handle) when we know that handle contains (void *)(-1).
6941 61. The Exim daemon panic-logs an error return when it closes the incoming
6942 connection. However "connection reset by peer" seems to be common, and
6943 isn't really an error worthy of noting specially, so that particular error
6946 62. When Exim is trying to find all the local interfaces, it used to panic and
6947 die if the ioctl to get the interface flags failed. However, it seems that
6948 on at least one OS (Solaris 9) it is possible to have an interface that is
6949 included in the list of interfaces, but for which you get a failure error
6950 for this call. This happens when the interface is not "plumbed" into a
6951 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6952 failure of the "get flags" call assumes that the interface is down.
6954 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6955 makes life easier for people who are doing arithmetic on fields extracted
6956 from dates, where you often get leading zeros that should not be
6957 interpreted as octal.
6959 64. Added qualify_domain to the redirect router, to override the global
6962 65. If a pathologically long header line contained very many addresses (the
6963 report of this problem mentioned 10 000) and each of them was rewritten,
6964 Exim could use up a very large amount of memory. (It kept on making new
6965 copies of the header line as it rewrote, and never released the old ones.)
6966 At the expense of a bit more processing, the header rewriting function has
6967 been changed so that it no longer eats memory in this way.
6969 66. The generation of the Received: header has been moved from the time that a
6970 message starts to be received, to the time that it finishes. The timestamp
6971 in the Received: header should now be very close to that of the <= log
6972 line. There are two side-effects of this change:
6974 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6975 logged header lines no longer include the local Received: line, because
6976 it has not yet been created. The same applies to a copy of the message
6977 that is returned to a non-SMTP sender when a message is rejected.
6979 (b) When a filter file is tested using -bf, no additional Received: header
6980 is added to the test message. After some thought, I decided that this
6983 This change does not affect the value of $received_for. It is still set
6984 after address rewriting, but before local_scan() is called.
6986 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6988 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6989 gave an unhelpful panic error message, and a defer error. I have managed to
6990 change this behaviour so that it now rejects any supplied certificate,
6991 which seems right, as the list of acceptable certificates is empty.
6993 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6994 gave an unhelpful defer error. I have not managed to make this reject any
6995 supplied certificates, but the error message it gives is "no certificate
6996 supplied", which is not helpful.
6998 70. exigrep's output now also includes lines that are not associated with any
6999 message, but which match the given pattern. Implemented by a patch from
7000 Martin Sluka, which also tidied up the Perl a bit.
7002 71. Recipient callout verification, like sender verification, was using <> in
7003 the MAIL FROM command. This isn't really the right thing, since the actual
7004 sender may affect whether the remote host accepts the recipient or not. I
7005 have changed it to use the actual sender in the callout; this means that
7006 the cache record is now keyed on a recipient/sender pair, not just the
7007 recipient address. There doesn't seem to be a real danger of callout loops,
7008 since a callout by the remote host to check the sender would use <>.
7009 [SEE ABOVE: changed after hitting problems.]
7011 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
7012 temporary errors. However, in the case of such a code being given after
7013 the end of a data transmission (i.e. after ".") Exim was failing to write
7014 a retry record for the message. (Yes, there was some broken host that was
7015 actually sending 8xx at this point.)
7017 73. An unknown lookup type in a host list could cause Exim to panic-die when
7018 the list was checked. (An example that provoked this was putting <; in the
7019 middle of a list instead of at the start.) If this happened during a DATA
7020 ACL check, a -D file could be left lying around. This kind of configuration
7021 error no longer causes Exim to die; instead it causes a defer error. The
7022 incident is still logged to the main and panic logs.
7024 74. Buglet left over from Exim 3 conversion. The message "too many messages
7025 in one connection" was written to the rejectlog but not the mainlog, except
7026 when address rewriting (yes!) was being logged.
7028 75. Added write_rejectlog option.
7030 76. When a system filter was run not as root (that is, when system_filter_user
7031 was set), the values of the $n variables were not being returned to the
7032 main process; thus, they were not subsequently available in the $sn
7035 77. Added +return_path_on_delivery log selector.
7037 78. A connection timeout was being treated differently from recipients deferred
7038 when testing hosts_max_try with a message that was older than the host's
7039 retry timeout. (The host should not be counted, thus allowing all hosts to
7040 be tried at least once before bouncing.) This may have been the cause of an
7041 occasionally reported bug whereby a message would remain on the queue
7042 longer than the retry timeout, but would be bounced if a delivery was
7043 forced. I say "may" because I never totally pinned down the problem;
7044 setting up timeout/retry tests is difficult. See also the next item.
7046 79. The ultimate address timeout was not being applied to errors that involved
7047 a combination of host plus message (for example, a timeout on a MAIL
7048 command). When an address resolved to a number of possible hosts, and they
7049 were not all tried for each delivery (e.g. because of hosts_max_try), a
7050 message could remain on the queue longer than the retry timeout.
7052 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
7055 81. Fixed an obscure SMTP outgoing bug which required at least the following
7056 conditions: (a) there was another message waiting for the same server;
7057 (b) the server returned 5xx to all RCPT commands in the first message so
7058 that the message was not completed; (c) the server dropped the connection
7059 or gave a negative response to the RSET that Exim sends to abort the
7060 transaction. The observed case was a dropped connection after DATA that had
7061 been sent in pipelining mode. That is, the server had advertised PIPELINING
7062 but was not implementing it correctly. The effect of the bug was incorrect
7063 behaviour, such as trying another host, and this could lead to a crash.
7069 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
7070 and daemon.c were passed as pointers to ints; they should have been
7071 pointers to socklen_t variables (which are typically unsigned ints).
7073 2. Some signed/unsigned type warnings in the os.c file for Linux have been
7076 3. Fixed a really odd bug that affected only the testing scheme; patching a
7077 certain fixed string in the binary changed the value of another string that
7078 happened to be identical to the end of the original first string.
7080 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
7081 name", it returns that address as the IP address. On some operating
7082 systems (e.g. Solaris), it also passes back the IP address string as the
7083 "host name". However, on others (e.g. Linux), it passes back an empty
7084 string. Exim wasn't checking for this, and was changing the host name to an
7085 empty string, assuming it had been canonicalized.
7087 5. Although rare, it is permitted to have more than one PTR record for a given
7088 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
7089 all the names associated with an address, because they do in Solaris.
7090 However, it seems that they do not in Linux for data that comes from the
7091 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
7092 I found this out when I moved to a new Linux workstation and tried to run
7093 the Exim test suite.
7095 To get round this problem I have changed the code so that it now does its
7096 own call to the DNS to look up PTR records when searching for a host name.
7097 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
7098 addresses that are only in /etc/hosts are still found.
7100 This behaviour is, however, controlled by an option called host_lookup_
7101 order, which defaults to "bydns:byaddr". If people want to use the other
7102 order, or indeed, just use one or the other means of lookup, they can
7103 specify it in this variable.
7105 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
7106 some operating systems, this comes back from gethostbyaddr() as an empty
7107 string, and this is what Exim used to test for. However, it seems that in
7108 other systems, "." is yielded. Exim now tests for this case too.
7110 7. The values of check_spool_space and check_log_space are now held internally
7111 as a number of kilobytes instead of an absolute number of bytes. If a
7112 numbers is specified without 'K' or 'M', it is rounded up to the nearest
7113 kilobyte. This means that much larger values can be stored.
7115 8. Exim monitor: an attempt to get the action menu when not actually pointing
7116 at a message produces an empty menu entitled "No message selected". This
7117 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
7118 no entries in it ("Shell widget menu has zero width and/or height"). So I
7119 have added a single, blank menu entry in this case.
7121 9. Added ${quote_local_part.
7123 10. MIME decoding is now applied to the contents of Subject: header lines when
7126 11. Now that a reference to $sender_host_address automatically causes a reverse
7127 lookup to occur if necessary (4.13/18), there is no need to arrange for a
7128 host lookup before query-style lookups in lists that might use this
7129 variable. This has therefore been abolished, and the "net-" prefix is no
7130 longer necessary for query-style lookups.
7132 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
7133 have been a typo for LFLAGS, so it has been changed.
7135 13. The install script calls Exim with "-C /dev/null" in order to find the
7136 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
7137 to be output. However, since Exim outputs its version number before the
7138 error, it didn't break the script. It just looked ugly. I fixed this by
7139 always allowing "-C /dev/null" if the caller is root.
7141 14. Ignore overlarge ACL variable number when reading spool file - insurance
7142 against a later release with more variables having written the file.
7144 15. The standard form for an IPv6 address literal was being rejected by EHLO.
7145 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
7146 this, as well as the form without the "IPv6" on the front.
7148 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
7149 OS/Makefile-Darwin file.
7151 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
7152 only to LDAP libraries that do not have LDAP_OPT_DEREF.
7154 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
7155 variable. However, some OS use ints for inodes. I've added cast to long int
7156 to get rid of the compiler warning.
7158 19. I had forgotten to lock out "/../" in configuration file names when
7159 ALT_CONFIG_PREFIX was set.
7161 20. Routers used for verification do not need to specify transports. However,
7162 if such a router generated a host list, and callout was configured, Exim
7163 crashed, because it could not find a port number from the (non-existent)
7164 transport. It now assumes port 25 in this circumstance.
7166 21. Added the -t option to exigrep.
7168 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
7169 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
7170 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
7171 them is set, LOOKUP_LSEARCH is forced.
7173 23. "exim -bV" now outputs a list of lookups that are included in the binary.
7175 24. Added sender and host information to the "rejected by local_scan()" log
7176 line; previously there was no indication of these.
7178 25. Added .include_if_exists.
7180 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
7181 Linux. It turns out that not all file systems support this. Apparently some
7182 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
7183 do it.) To cope with this, the error EINVAL, which means that sync-ing is
7184 not supported on the file descriptor, is now ignored when Exim is trying to
7185 sync a directory. This applies only to Linux.
7187 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
7189 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
7190 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
7193 29. Added support for maildirsize files from supplied patch (modified a bit).
7195 30. The use of :fail: followed by an empty string could lead Exim to respond to
7196 sender verification failures with (e.g.):
7198 550 Verification failed for <xxx>
7199 550 Sender verify failed
7201 where the first response line was missing the '-' that indicates it is not
7202 the final line of the response.
7204 31. The loop for finding the name of the user that called Exim had a hardwired
7205 limit of 10; it now uses the value of finduser_retries, which is used for
7206 all other user lookups.
7208 32. Added $received_count variable, available in data and not_smtp ACLs, and at
7211 33. Exim was neglecting to zero errno before one call of strtol() when
7212 expanding a string and expecting an integer value. On some systems this
7213 resulted in spurious "integer overflow" errors. Also, it was casting the
7214 result into an int without checking.
7216 34. Testing for a connection timeout using "timeout_connect" in the retry rules
7217 did not work. The code looks as if it has *never* worked, though it appears
7218 to have been documented since at least release 1.62. I have made it work.
7220 35. The "timeout_DNS" error in retry rules, also documented since at least
7221 1.62, also never worked. As it isn't clear exactly what this means, and
7222 clearly it isn't a major issue, I have abolished the feature by treating it
7223 as "timeout", and writing a warning to the main and panic logs.
7225 36. The display of retry rules for -brt wasn't always showing the error code
7228 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
7229 timeout_connect_A, timeout_connect_MX.
7231 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
7232 to the empty sender.
7234 39. The daemon was not analysing the content of -oX till after it had closed
7235 stderr and disconnected from the controlling terminal. This meant that any
7236 syntax errors were only noted on the panic log, and the return code from
7237 the command was 0. By re-arranging the code a little, I've made the
7238 decoding happen first, so such errors now appear on stderr, and the return
7239 code is 1. However, the actual setting up of the sockets still happens in
7240 the disconnected process, so errors there are still only recorded on the
7243 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
7244 connections (as happens on some IP stacks) was logged at start up time as
7245 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
7246 it from "IPv6 and IPv4", which means that two separate sockets are being
7249 41. The debug output for gethostbyname2() or getipnodebyname() failures now
7250 says whether AF_INET or AF_INET6 was passed as an argument.
7252 42. Exiwhat output was messed up when time zones were included in log
7255 43. Exiwhat now gives more information about the daemon's listening ports,
7256 and whether -tls-on-connect was used.
7258 44. The "port" option of the smtp transport is now expanded.
7260 45. A "message" modifier in a "warn" statement in a non-message ACL was being
7261 silently ignored. Now an error message is written to the main and panic
7264 46. There's a new ACL modifier called "logwrite" which writes to a log file
7265 as soon as it is encountered.
7267 47. Added $local_user_uid and $local_user_gid at routing time.
7269 48. Exim crashed when trying to verify a sender address that was being
7272 49. Exim was recognizing only a space character after ".include". It now also
7273 recognizes a tab character.
7275 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
7276 extracting the relevant information from the specification. The man page no
7277 longer contains scrambled data for the -d option, and I've added a section
7278 at the front about calling Exim under different names.
7280 51. Added "extra_headers" argument to the "mail" command in filter files.
7282 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
7285 53. Installed eximstats 1.29.
7287 54. Added transport_filter_timeout as a generic transport option.
7289 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
7290 Cc: header lines. This was required by RFC 822, but it not required by RFC
7293 56. Exim used to add From:, Date:, and Message-Id: header lines to any
7294 incoming messages that did not have them. Now it does so only if the
7295 message originates locally, that is, if there is no associated remote host
7296 address. When Resent- header lines are present, this applies to the Resent-
7297 lines rather than the non-Resent- lines.
7299 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
7300 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
7302 58. Messages for configuration errors now include the name of the main
7303 configuration file - useful now that there may be more than one file in a
7304 list (.included file names were always shown).
7306 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
7307 for those rare installations that do not start the daemon as root or run it
7308 setuid root. I've cut out the call to initgroups() if the daemon is not
7311 60. The Exim user and group can now be bound into the binary as text strings
7312 that are looked up at the start of Exim's processing.
7314 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
7316 62. Added $mailstore_basename variable.
7318 63. Installed patch to sieve.c from Michael Haardt.
7320 64. When Exim failed to open the panic log after failing to open the main log,
7321 the original message it was trying to log was written to stderr and debug
7322 output, but if they were not available (the usual case in production), it
7323 was lost. Now it is written to syslog before the two lines that record the
7324 failures to open the logs.
7326 65. Users' Exim filters run in subprocesses under the user's uid. It is
7327 possible for a "deliver" command or an alias in a "personal" command to
7328 provoke an address rewrite. If logging of address rewriting is configured,
7329 this fails because the process is not running as root or exim. There may be
7330 a better way of dealing with this, but for the moment (because 4.30 needs
7331 to be released), I have disabled address rewrite logging when running a
7332 filter in a non-root, non-exim process.
7338 1. The buildconfig auxiliary program wasn't quoting the value set for
7339 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
7340 not defined. This bug was masked in 4.22 by the effect that was fixed in
7343 2. Some messages that were rejected after a message id was allocated were
7344 shown as "incomplete" by exigrep. It no longer does this for messages that
7345 are rejected by local_scan() or the DATA or non-SMTP ACLs.
7347 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
7348 have allow_domain_literals set, the ID did not get logged in the <= line.
7349 Domain literals are now always recognized in Message-ID: header lines.
7351 4. The first argument for a ${extract expansion item is the key name or field
7352 number. Leading and trailing spaces in this item were not being ignored,
7353 causing some misleading effects.
7355 5. When deliver_drop_privilege was set, single queue runner processes started
7356 manually (i.e. by the command "exim -q") or by the daemon (which uses the
7357 same command in the process it spins off) were not dropping privilege.
7359 6. When the daemon running as "exim" started a queue runner, it always
7360 re-executed Exim in the spun-off process. This is a waste of effort when
7361 deliver_drop_privilege is set. The new process now just calls the
7362 queue-runner function directly.
7368 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
7371 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
7372 ignored. Though the use of -oP was forcing the writing of a pid file, it
7373 was always written to the default place.
7375 3. If the message "no IP address found for host xxxx" is generated during
7376 incoming verification, it is now followed by identification of the incoming
7377 connection (so you can more easily find what provoked it).
7379 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
7381 5. Added some features to "harden" Exim a bit more against certain attacks:
7383 (a) There is now a build-time option called FIXED_NEVER_USERS that can
7384 be put in Local/Makefile. This is like the never_users runtime option,
7385 but it cannot be overridden. The default setting is "root".
7387 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
7388 prefix string with which any file named in a -C command line option
7391 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
7392 is retained for -C and -D only if the caller of Exim is root. Without
7393 it, the exim user may also use -C and -D and retain privilege.
7395 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
7396 command line option is disabled.
7398 6. Macro names set by the -D option must start with an upper case letter, just
7399 like macro names defined in the configuration file.
7401 7. Added "dereference=" facility to LDAP.
7403 8. Two instances of the typo "uknown" in the source files are fixed.
7405 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
7406 the Configure-Makefile script screwed up while processing it.
7408 10. Incorporated PCRE 4.4.
7410 11. The SMTP synchronization check was not operating right at the start of an
7411 SMTP session. For example, it could not catch a HELO sent before the client
7412 waited for the greeting. There is now a check for outstanding input at the
7413 point when the greeting is written. Because of the duplex, asynchronous
7414 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
7415 way, but not yet received, when the check is performed.
7417 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
7418 on TCP/IP sockets, because this apparently causes some broken clients to
7421 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
7422 unchanged) from the Cygwin maintainer.
7424 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
7425 was supported instead of testing for mbx. Effectively a typo.
7427 15. The spa authenticator server code was not checking that the input it
7428 received was valid base64.
7430 16. The debug output line for the "set" modifier in ACLs was not showing the
7431 name of the variable that was being set.
7433 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
7435 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
7436 lookup on its own. Something else was needed to trigger the lookup. For
7437 example, a match in host_lookup or the need for a host name in a host list.
7438 Now, if $sender_host_name is referenced and the host name has not yet been
7439 looked up, a lookup is performed. If the lookup fails, the variable remains
7440 empty, and $host_lookup_failed is set to "1".
7442 19. Added "eqi" as a case-independent comparison operator.
7444 20. The saslauthd authentication condition could segfault if neither service
7445 nor realm was specified.
7447 21. If an overflowing value such as "2048M" was set for message_size_limit, the
7448 error message that was logged was misleading, and incoming SMTP
7449 connections were dropped. The message is now more accurate, and temporary
7450 errors are given to SMTP connections.
7452 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
7453 (except RSET) with a 421 error, until QUIT is received. However, it was
7454 failing to send a response to QUIT.
7456 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
7457 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
7458 ACL is now run after the helo_try_verify_hosts code.
7460 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
7461 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
7462 case-independent, so other case variants are also recognized). Apparently
7463 some systems use these upper case variants.
7465 25. If more than two messages were waiting for the same host, and a transport
7466 filter was specified for the transport, Exim sent two messages over the
7467 same TCP/IP connection, and then failed with "socket operation on non-
7468 socket" when it tried to send the third.
7470 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
7472 27. The extern definition of crypt16() in expand.c was not being excluded when
7473 the OS had its own crypt16() function.
7475 28. Added bounce_return_body as a new option, and bounce_return_size_limit
7476 as a preferred synonym for return_size_limit, both as an option and as an
7479 29. Added LIBS=-liconv to OS/Makefile-OSF1.
7481 30. Changed the default configuration ACL to relax the local part checking rule
7482 for addresses that are not in any local domains. For these addresses,
7483 slashes and pipe symbols are allowed within local parts, but the sequence
7484 /../ is explicitly forbidden.
7486 31. SPA server authentication was not clearing the challenge buffer before
7489 32. log_message in a "warn" ACL statement was writing to the reject log as
7490 well as to the main log, which contradicts the documentation and doesn't
7491 seem right (because no rejection is happening). So I have stopped it.
7493 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
7494 However, I am unable to do any testing of this.
7496 34. Fixed an infelicity in the appendfile transport. When checking directories
7497 for a mailbox, to see if any needed to be created, it was accidentally
7498 using path names with one or more superfluous leading slashes; tracing
7499 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
7501 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
7502 contents are added to the log line that is written for every discarded
7503 recipient. (Previously a log_message setting was ignored.)
7505 36. The ${quote: operator now quotes the string if it is empty.
7507 37. The install script runs exim in order to find its version number. If for
7508 some reason other than non-existence or emptiness, which it checks, it
7509 could not run './exim', it was installing it with an empty version number,
7510 i.e. as "exim-". This error state is now caught, and the installation is
7513 38. An argument was missing from the function that creates an error message
7514 when Exim fails to connect to the socket for saslauthd authentication.
7515 This could cause Exim to crash, or give a corrupted message.
7517 39. Added isip, isip4, and isip6 to ${if conditions.
7519 40. The ACL variables $acl_xx are now saved with the message, and can be
7520 accessed later in routers, transports, and filters.
7522 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
7523 strings in the file are not string-expanded.
7525 42. If a MAIL command specified a SIZE value that was too large to fit into an
7526 int variable, the check against message_size_limit failed. Such values are
7527 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
7528 day this will have to be increased, but I don't think I want to be around
7529 when emails are that large.
7536 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
7537 iconv() is not standard in FreeBSD.
7539 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
7540 IPv6 enabled. The observed symptom was a segmentation fault on return from
7541 the function os_common_find_running_interfaces() in src/os.c.
7543 3. In the check_special_case() function in daemon.c I had used "errno" as an
7544 argument name, which causes warnings on some systems. This was basically a
7545 typo, since it was named "eno" in the comments!
7547 4. The code that waits for the clock to tick (at a resolution of some fraction
7548 of a second) so as to ensure message-id uniqueness was always waiting for
7549 at least one whole tick, when it could have waited for less. [This is
7550 almost certainly not relevant at current processor speeds, where it is
7551 unlikely to ever wait at all. But we try to future-proof.]
7553 5. The function that sleeps for a time interval that includes fractions of a
7554 second contained a race. It did not block SIGALRM between setting the
7555 timer, and suspending (a couple of lines later). If the interval was short
7556 and the sigsuspend() was delayed until after it had expired, the suspension
7557 never ended. On busy systems this could lead to processes getting stuck for
7560 6. Some uncommon configurations may cause a lookup to happen in a queue runner
7561 process, before it forks any delivery processes. The open lookup caching
7562 mechanism meant that the open file or database connection was passed into
7563 the delivery process. The problem was that delivery processes always tidy
7564 up cached lookup data. This could cause a problem for the next delivery
7565 process started by the queue runner, because the external queue runner
7566 process does not know about the closure. So the next delivery process
7567 still has data in the lookup cache. In the case of a file lookup, there was
7568 no problem because closing a file descriptor in a subprocess doesn't affect
7569 the parent. However, if the lookup was caching a connection to a database,
7570 the connection was closed, and the second delivery process was likely to
7571 see errors such as "PGSQL: query failed: server closed the connection
7572 unexpectedly". The problem has been fixed by closing all cached lookups
7573 in a queue runner before running a delivery process.
7575 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
7576 seem to have the "const" qualifier which it has on other OS. I've
7579 8. Change 4.21/2 was too strict. It is only if there are two authenticators
7580 *of the same type* (client or server) with the same public name that an
7581 error should be diagnosed.
7583 9. When Exim looked up a host name for an IP address, but failed to find the
7584 original IP address when looking up the host name (a safety check), it
7585 output the message "<ip address> does not match any IP for NULL", which was
7586 confusing, to say the least. The bug was that the host name should have
7587 appeared instead of "NULL".
7589 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
7590 user that is built into the binary, and the -C or -D options is used, root
7591 privilege is dropped before the configuration file is read. In addition,
7592 logging is switched to stderr instead of the normal log files. If the
7593 configuration then re-defines the Exim user, the unprivileged environment
7594 is probably not what is expected, so Exim logs a panic warning message (but
7597 However, if deliver_drop_privilege is set, the unprivileged state may well
7598 be exactly what is intended, so the warning has been cut out in that case,
7599 and Exim is allowed to try to write to its normal log files.
7605 1. smtp_return_error_details was not giving details for temporary sender
7606 or receiver verification errors.
7608 2. Diagnose a configuration error if two authenticators have the same public
7611 3. Exim used not to create the message log file for a message until the first
7612 delivery attempt. This could be confusing when incoming messages were held
7613 for policy or load reasons. The message log file is now created at the time
7614 the message is received, and an initial "Received" line is written to it.
7616 4. The automatically generated man page for command line options had a minor
7617 bug that caused no ill effects; however, a more serious problem was that
7618 the procedure for building the man page automatically didn't always
7619 operate. Consequently, release 4.20 contains an out-of-date version. This
7620 shouldn't happen again.
7622 5. When building Exim with embedded Perl support, the script that builds the
7623 Makefile was calling 'perl' to find its compile-time parameters, ignoring
7624 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
7626 6. The freeze_tell option was not being used for messages that were frozen on
7627 arrival, either by an ACL or by local_scan().
7629 7. Added the smtp_incomplete_transaction log selector.
7631 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
7632 was accepting AUTH without a new EHLO.
7634 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
7635 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
7638 10. Logging of TCP/IP connections (when configured) now happens in the main
7639 daemon process instead of the child process, so that the TCP/IP connection
7640 count is more accurate (but it can never be perfect).
7642 11. The use of "drop" in a nested ACL was not being handled correctly in the
7643 outer ACL. Now, if condition failure induced by the nested "drop" causes
7644 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
7645 or "require"), the connection is dropped.
7647 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
7648 that yield "discard" can now be used with an "accept" or a "discard" verb,
7649 but an error is generated for any others (because I can't see a useful way
7650 to define what should happen).
7652 13. When an ACL is read dynamically from a file (or anywhere else), the lines
7653 are now processed in the same way as lines in the Exim configuration file.
7654 In particular, continuation lines are supported.
7656 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
7658 15. Added -ti meaning -t -i.
7660 16. Check for letters, digits, hyphens, and dots in the names of dnslist
7661 domains, and warn by logging if others are found.
7663 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
7664 returned from GIFCONF when Exim is trying to find the list of interfaces on
7665 a host. The code in os.c has been modified to copy each ifreq to an aligned
7666 structure in all cases.
7668 Also, in some cases, the returned ifreq's were being copied to a 'struct
7669 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
7670 means the last couple of bytes of an IPv6 address could be chopped if the
7671 ifreq contained only a normal sockaddr (14 bytes storage).
7673 18. Named domain lists were not supported in the hosts_treat_as_local option.
7674 An entry such as +xxxx was not recognized, and was treated as a literal
7677 19. Ensure that header lines added by a DATA ACL are included in the reject log
7678 if the ACL subsequently rejects the message.
7680 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
7681 MD5 (which is deprecated).
7683 21. When testing a filter file using -bf, Exim was writing a message when it
7684 took the sender from a "From " line in the message, but it was not doing so
7685 when it took $return_path from a Return-Path: header line. It now does.
7687 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
7688 with a valid header line field name (a series of printing characters
7689 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
7691 23. Changed "disc" in the source to "disk" to conform to the documentation and
7692 the book and for uniformity.
7694 24. Ignore Sendmail's -Ooption=value command line item.
7696 25. When execve() failed while trying to run a command in a pipe transport,
7697 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
7698 could be confused with a return value of 69 from the command itself. This
7699 has been changed to 127, the value the shell returns if it is asked to run
7700 a non-existent command. The wording for the related log line suggests a
7701 non-existent command as the problem.
7703 26. If received_header_text expands to an empty string, do not add a Received:
7704 header line to the message. (Well, it adds a token one on the spool, but
7705 marks it "old" so that it doesn't get used or transmitted.)
7707 27. Installed eximstats 1.28 (addition of -nt option).
7709 28. There was no check for failure on the call to getsockname() in the daemon
7710 code. This can fail if there is a shortage of resources on the system, with
7711 ENOMEM, for example. A temporary error is now given on failure.
7713 29. Contrary to the C standard, it seems that in some environments, the
7714 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
7715 program. Exim now does this explicitly; it affects the formatting of
7716 timestamps using strftime().
7718 30. If exiqsumm was given junk data, it threw up some uninitialized variable
7719 complaints. I've now initialized all the variables, to avoid this.
7721 32. Header lines added by a system filter were not being "seen" during
7722 transport-time rewrites.
7724 33. The info_callback() function passed to OpenSSL is set up with type void
7725 (*)(SSL *, int, int), as described somewhere. However, when calling the
7726 function (actually a macro) that sets it up, the type void(*)() is
7727 expected. I've put in a cast to prevent warnings from picky compilers.
7729 34. If a DNS black list lookup found a CNAME record, but there were no A
7730 records associated with the domain it pointed at, Exim crashed.
7732 35. If a DNS black list lookup returned more than one A record, Exim ignored
7733 all but the first. It now scans all returned addresses if a particular IP
7734 value is being sought. In this situation, the contents of the
7735 $dnslist_value variable are a list of all the addresses, separated by a
7738 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
7739 to accept a host name and all its aliases if the forward lookup for any of
7740 them yielded the IP address of the incoming connection. Now it accepts only
7741 those names whose forward lookup yields the correct IP address. Any other
7742 names are discarded. This closes a loophole whereby a rogue DNS
7743 administrator could create reverse DNS records to break through a
7744 wildcarded host restriction in an ACL.
7746 37. If a user filter or a system filter that ran in a subprocess used any of
7747 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
7748 the wrong values were passed to the pipe command ($thisaddress had the
7749 value of $0, $0 had the value of $1, etc). This bug was introduced by
7750 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
7752 38. Improved the line breaking for long SMTP error messages from ACLs.
7753 Previously, if there was no break point between 40 and 75 characters, Exim
7754 left the rest of the message alone. Two changes have been made: (a) I've
7755 reduced the minimum length to 35 characters; (b) if it can't find a break
7756 point between 35 and 75 characters, it looks ahead and uses the first one
7757 that it finds. This may give the occasional overlong line, but at least the
7758 remaining text gets split now.
7760 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
7761 file descriptors might be low, and that setting 1000 would always raise it.
7762 It turns out that in some environments, the limit is already over 1000 and
7763 that lowering it causes trouble. So now Exim takes care not to decrease it.
7765 40. When delivering a message, the value of $return_path is set to $sender_
7766 address at the start of routing (routers may change the value). By an
7767 oversight, this default was not being set up when an address was tested by
7768 -bt or -bv, which affected the outcome if any router or filter referred to
7771 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
7772 log only when "message" or "log_message" are set. However, if one of the
7773 conditions was an address verification, or a call to a nested ACL, the
7774 messages generated by the underlying test were being passed through. This
7775 no longer happens. The underlying message is available in $acl_verify_
7776 message for both "message" and "log_message" expansions, so it can be
7777 passed through if needed.
7779 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
7780 new expansion $bh_ to give the encoded byte string without charset
7781 translation. Translation happens only if iconv() is available; HAVE_ICONV
7782 indicates this at build time. HEADERS_CHARSET gives the charset to
7783 translate to; headers_charset can change it in the configuration, and
7784 "headers charset" can change it in an individual filter file.
7786 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
7787 that creates RFC 2047 encoded "words" labels them as that charset instead
7788 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
7789 expansion operator; (ii) when Exim creates a From: line for a local
7790 message; (iii) when a header line is rewritten to include a "phrase" part.
7792 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
7793 buggy, causing it to skip the first lines of messages whose message ID
7794 ended in 'D'. This would not have bitten before Exim release 4.14, because
7795 message IDs were unlikely to end in 'D' before then. The effect was to have
7796 incorrect size information for certain domains.
7798 45. #include "config.h" was missing at the start of the crypt16.c module. This
7799 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
7802 46. If there was a timeout during a "random" callout check, Exim treated it as
7803 a failure of the random address, and carried on sending RSET and the real
7804 address. If the delay was just some slowness somewhere, the response to the
7805 original RCPT would be taken as a response to RSET and so on, causing
7806 mayhem of various kinds.
7808 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
7809 when I implemented it. It didn't allow for the fact that some option values
7810 may legitimately be negative (e.g. size_addition), and it didn't even do
7811 the right test for positive values.
7813 48. Domain names in DNS records are case-independent. Exim always looks them up
7814 in lower case. Some resolvers return domain names in exactly the case they
7815 appear in the zone file, that is, they may contain uppercase letters. Not
7816 all resolvers do this - some return always lower case. Exim was treating a
7817 change of case by a resolver as a change of domain, similar to a widening
7818 of a domain abbreviation. This triggered its re-routing code and so it was
7819 trying to route what was effectively the same domain again. This normally
7820 caused routing to fail (because the router wouldn't handle the domain
7821 twice). Now Exim checks for this case specially, and just changes the
7822 casing of the domain that it ultimately uses when it transmits the message
7825 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
7828 50. If a filter generated a file delivery with a non-absolute name (possible if
7829 no home directory exists for the router), the forbid_file option was not
7832 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
7833 the existing equality matching.
7835 52. Exim was using ints instead of ino_t variables in some places where it was
7836 dealing with inode numbers.
7838 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
7839 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
7840 called TMPDIR, and if it finds it is different, it changes its value.
7842 54. The smtp_printf() function is now made available to local_scan() so
7843 additional output lines can be written before returning. There is also an
7844 smtp_fflush() function to enable the detection of a dropped connection.
7845 The variables smtp_input and smtp_batched_input are exported to
7848 55. Changed the default runtime configuration: the message "Unknown user"
7849 has been removed from the ACL, and instead placed on the localuser router,
7850 using the cannot_route_message feature. This means that any verification
7851 failures that generate their own messages won't get overridden. Similarly,
7852 the "Unrouteable address" message that was in the ACL for unverifiable
7853 relay addresses has also been removed.
7855 56. Added hosts_avoid_esmtp to the smtp transport.
7857 57. The exicyclog script was not checking for the esoteric option
7858 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7859 will work only if exicyclog is run under the appropriate euid.
7861 58. Following a discussion on the list, the rules by which Exim recognises line
7862 endings on incoming messages have been changed. The -dropcr and drop_cr
7863 options are now no-ops, retained only for backwards compatibility. The
7864 following line terminators are recognized: LF CRLF CR. However, special
7865 processing applies to CR:
7867 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7868 nor a local message in the state where . is a terminator.
7870 (ii) If a bare CR is encountered in a header line, an extra space is added
7871 after the line terminator so as not to end the header. The reasoning
7872 behind this is that bare CRs in header lines are most likely either
7873 to be mistakes, or people trying to play silly games.
7875 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7876 was being incorrectly given as 18 bytes larger than it should have been.
7877 This is a VOB (very old bug).
7879 60. This may never have affected anything current, but just in case it has:
7880 When the local host is found other than at the start of a list of hosts,
7881 the local host, those with the same MX, and any that follow, are discarded.
7882 When the list in question was part of a longer list of hosts, the following
7883 hosts (not currently being processed) were also being discarded. This no
7884 longer happens. I'm not sure if this situation could ever has previously
7887 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7890 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7891 header. This is something that is recommended in a new Internet Draft, and
7892 is something that is documented as being done by Sendmail. There are two
7893 possible values. For messages generated by the autoreply transport, Exim
7896 Auto-Submitted: auto-replied
7898 whereas for all other generated messages (e.g. bounces) it adds
7900 Auto-Submitted: auto-generated
7902 63. The "personal" condition in filters now includes a test for the
7903 Auto-Submitted: header. If it contains the string "auto-" the message it
7904 not considered personal.
7906 64. Added rcpt_include_affixes as a generic transport option.
7908 65. Added queue_only_override (default true).
7910 66. Added the syslog_duplication option.
7912 67. If what should have been the first header line of a message consisted of
7913 a space followed by a colon, Exim was mis-interpreting it as a header line.
7914 It isn't of course - it is syntactically invalid and should therefore be
7915 treated as the start of the message body. The misbehaviour could have
7916 caused a number of strange effects, including loss of data in subsequent
7917 header lines, and spool format errors.
7919 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7920 client host had authenticated. This control can now be exercised by an ACL
7921 for more flexibility.
7923 69. By default, callouts do not happen when testing with -bh. There is now a
7924 variant, -bhc, which does actually run the callout code, including
7925 consulting and updating the callout cache.
7927 70. Added support for saslauthd authentication, courtesy of Alexander
7930 71. If statvfs() failed on the spool or log directories while checking their
7931 size for availability, Exim confusingly gave the error "space shortage".
7932 Furthermore, in debugging mode it crashed with a floating point exception.
7933 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7934 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7935 serious problem, Exim now writes to the main and panic logs when this
7936 happens, with details of the failure. It then refuses to accept the
7937 incoming message, giving the message "spool directory problem" or "log
7938 directory problem" with a 421 code for SMTP messages.
7940 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7941 0, 1, and 2 exist, because some OS complain for execs without them (see
7942 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7943 descriptors. However, the code omitted to check that the open succeeded,
7944 causing mysterious errors if for some reason the permissions on /dev/null
7945 got screwed. Now Exim writes a message to the main and panic logs, and
7946 bombs out if it can't open /dev/null.
7948 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7949 interact so that it is all more flexible. It is supposed to remain
7950 backwards compatible. Also added extra_local_interfaces.
7952 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7953 to bomb out with an assertion failure - to the client this appears as a
7954 connection drop. This problem occurs in the part of the code that was taken
7955 from the Samba project. Fortunately, the assertion is in a very simple
7956 function, so I have fixed this by reproducing the function inline in the
7957 one place where it is called, and arranging for authentication to fail
7958 instead of killing the process with assert().
7960 75. The SPA client code was not working when the server requested OEM rather
7961 than Unicode encoding.
7963 76. Added code to make require_files with a specific uid setting more usable in
7964 the case where statting the file as root fails - usually a non-root-mounted
7965 NFS file system. When this happens and the failure is EACCES, Exim now
7966 forks a subprocess and does the per-uid checking as the relevant uid.
7968 77. Added process_log_path.
7970 78. If log_file_path was not explicitly set, a setting of check_log_space or
7971 check_log_inodes was ignored.
7973 79. If a space check for the spool or log partitions fails, the incident is now
7974 logged. Of course, in the latter case the data may get lost...
7976 80. Added the %p formatting code to string_format() so that it can be used to
7977 print addresses in debug_print(). Adjusted all the address printing in the
7978 debugging in store.c to use %p rather than %d.
7980 81. There was a concern that a line of code in smtp_in.c could overflow a
7981 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7982 initially expressed, the concern was not well-founded, because trailing
7983 spaces are removed early. However, if the trailing spaces were followed by
7984 a NULL, they did not get removed, so the overflow was possible. Two fixes
7987 (a) I re-wrote the offending code in a cleaner fashion.
7988 (b) If an incoming SMTP command contains a NULL character, it is rejected
7991 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7992 runs initgroups(), so that if the Exim user is in any additional groups,
7993 they will be used during message reception.
7999 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob